Final Report Summary - PRIMA-EDS (Polypharmacy in chronic diseases: Reduction of Inappropriate Medication and Adverse drug events in elderly populations by electronic Decision Support)
Executive Summary:
Polypharmacy constitutes an increasing problem in older multimorbid patients. The prevalence of polypharmacy in the population >75 years has been found to be between 25 and 50%. While there exists only limited evidence regarding the benefit, the evidence regarding potential harm of polypharmacy is increasing. Thus polypharmacy and inappropriate prescribing have been shown to increase the risk of adverse drug events, and contribute substantially to morbidity, hospitalization, and mortality. An Austria study found that 10% of hospitalizations of patients >70 years are directly related to an adverse drug event. This clearly indicates that there is a strong need to reduce inappropriate drug prescription. Diverse heterogeneous approaches to reduce polypharmacy and inappropriate prescribing have been proposed, but none of these have been evaluated regarding practicability in primary care and clinically relevant endpoints within a sufficiently powered randomized controlled trial. The most popular approach to reduce inappropriate prescribing is at present the implementation of Potentially Inappropriate Medication (PIM)-lists as like the Beers-List. These lists are based on Delphi-rounds of experts (mostly pharmacologists and geriatricians) and not on evidence derived from clinical studies, and they never have been shown to have any effect on clinical outcome. Other approaches include the STOPP-criteria of Gallagher 2008, updated in 2015, or the Medication Appropriateness Index (MAI). None of these tools have been sufficiently evaluated regarding clinical outcome, and they are hardly applicable in every day practice. Using a combination of these methods, we found in a study of 169 primary care patients from 22 practices an average of 2.7 inappropriate medications per patient. But physicians in every day clinical practice, and especially general practitioners (GPs) are unable to do thorough medication reviews with their patients due to knowledge gaps and limited time. Therefore in the PRIMA-eDS project, we developed an electronic decision support (eDS) tool to assist physicians and patients to avoid inappropriate medication and polypharmacy, and to identify prescriptions where risks outweigh possible benefits of the drug treatment. This tool is based on current best evidence to optimize treatment of those diseases that are most common in elderly patients like cardiovascular disease, heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders, COPD, and mental diseases. We collated this evidence from Duodecim’s Evidence based Medicine Guidelines and several systematic reviews and transformed it into recommendations to optimize drug treatment of multimorbid older patients. These recommendations were integrated in the existing EbMeDS electronic decision support together with various commercially available databases to check medication for renal dosing, interactions, indications, contraindications and potential adverse drug events. The tool was then evaluated in a cluster randomized controlled trial with 3904 patients >75 years taking at least 8 different medications. The trial revealed a trend towards a reduction of the composite endpoint of mortality and first hospitalisation in intention to treat analysis, and a significant reduction of this endpoint in per protocol analysis. Also, a significant reduction in the number of drugs prescribed to the patients in the intervention group could be shown. Qualitative and quantitative usability analysis of the tool revealed that it must be integrated in existing electronic health records for wider use. Overall, the PRIM-eDS-tool could be shown to be a promising approach to deal with the increasing problem of polypharmacy in our society. It will help to reduce inappropriate prescribing and improve health and outcome in older patients with polypharmacy.
Project Context and Objectives:
1.2.1 Context of the PRIMA-eDS-project
The prevalence of chronic comorbidity and multimorbidity is substantial, and affects more than half of the population > 75 years of age. The patterns of chronic multimorbidity include mainly the following diseases: cardiovascular disease (including coronary heart disease, cerebrovascular disease and peripheral vascular disease), heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders including pain, COPD, and mental diseases like depression and dementia, with heart failure showing the highest rate of comorbidity, being accompanied by an average of 2.9 additional chronic diseases. As a result of multimorbidity, polypharmacy seems to be ever increasing as guideline adherence regarding drug treatment of chronic disease is recommended by policymakers and medical colleges. Depending on definitions and setting between 25 and 50% of all patients > 75 years are exposed to five or more drugs. The benefits of drugs are usually shown in studies including mostly younger subjects without comorbidity and multiple medications. The patients involved in clinical trials usually are characterized by a life expectancy of many years. Therefore the results of these studies may not be applied to older comorbid patients on polypharmacy and with reduced life expectancy. It has been shown that most clinical practice guidelines do not modify or discuss the applicability of their recommendations for older patients with multiple comorbid diseases, and following all the guidelines might lead to distinct polypharmacy with increased risk of adverse reactions and interactions between drugs, outweighing the potential benefits of drug treatment.
According to a systematic review including all settings, the overall adverse drug event (ADE) rate lies at a median of 14.9% per 1000 person-months and increases with age and polypharmacy. In a recent systematic review focusing particularly on the primary care setting, we found the prevalence rate for ADEs to be about 10% in prospective studies. Up to 50% of these ADEs are judged to be preventable. Various international studies found a hospitalisation rate due to adverse drug events between 2.4 and 16.6%, depending largely on the age-group evaluated. Polypharmacy and inappropriate medication have been shown to contribute substantially to the burden of morbidity, hospitalisation and death.
According to the 2017 drug prescribing report of the AOK (German statutory health insurance) 78.7% of all prescriptions are handed to patients by primary care physicians (primary care physicians in Germany are General Practitioners and General Internists). Any intervention aimed at the reduction of polypharmacy will therefore be most effective, if the professional group of primary care physicians is targeted. Although there exists only limited evidence regarding the benefit of guideline-recommended drugs in polymorbid older patients, GPs are reluctant to discontinue medication initiated by specialists and hospital physicians. Recently, distinct guidelines to deal with polypharmacy have been developed, but physicians are insecure in the reduction of polypharmacy, and there is significant risk of legal consequences when deviating from guidelines dealing with single diseases. Also, it may take too much time to convince patients that they should discontinue drugs they have been prescribed by a specialist. Adequate tools and an easily accessible knowledge base to aid GPs and patients in their shared decision to reduce polypharmacy do not exist.
1.2.2 Objectives of PRIMA-eDS
With the background of the increasing problem of polypharmacy regarding health outcomes and the inability of primary care physicians as the main prescribers of polypharmacy to deal with this problem, PRIMA-eDS set out to develop a usable solution for every day practice in the form of electronic decision support. After the development of the tool, this new approach to dealing with inappropriate prescribing and polypharmacy has been tested in a randomized controlled trial.
In detail, the objectives of PRIMA-eDS were as follows:
• to carry out a series of systematic reviews of the literature, evaluating and defining current best evidence regarding the pharmacological treatment of elderly patients with multiple chronic diseases, specifically addressing the problem of polypharmacy in the treatment of the most common chronic diseases in elderly populations,
• to develop an electronic decision support tool (PRIMA-eDS-tool) based on existing EbMeDS-technology developed by Duodecim ltd. to aid physicians and patients to make use of current best evidence when coming to a shared decision regarding multiple drug treatment of the most common chronic diseases in elderly populations,
• to test the efficacy of the PRIMA-eDS-tool in an appropriately powered, multinational, multicentre cluster-randomised controlled trial with a composite of hospital admission and mortality as the primary endpoint, the number of prescribed drugs as secondary endpoint, and mortality and adverse drug events as safety endpoints,
• to optimise the PRIMA-eDS-tool with the results of a qualitative and quantitative usability test and the randomised controlled trial, and make it available for widespread use within and outside of Europe.
1.2.2.1 Objective 1: Gathering current best evidence
First, we aimed to gather current best evidence to optimize treatment of those diseases that are most common in elderly patients like cardiovascular disease, heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders, COPD, and mental diseases. We carried out a series of systematic reviews of the literature focused on the treatment of multiple chronic diseases in the primary care setting. We made use of existing reviews (e.g. Cochrane-reviews) and guidelines and checked these reviews and guidelines for applicability and validity regarding their use in an elderly multimorbid population. A further systematic review focused on existing strategies and tools to reduce polypharmacy like the Garfinkel-algorithm, the Beers list, the STOPP-criteria, the medication appropriateness index (MAI) and others. Our systematic reviews were carried out with highest standards in methodology as described by the Cochrane Handbook 5.1.0. Objective 1 was the main focus of Work Package 2 of the PRIMA-eDS-project.
1.2.2.2 Objective 2: Development and implementation of the eDS-tool
Secondly, we used the evidence obtained in Objective 1 to develop a sustainable electronic decision support (eDS) tool to reduce polypharmacy and inappropriate medication in older patients that can be easily applied in daily primary care practice. The ttol was based on the existing EbMeDS-technology developed by Duodecim ltd. (www.ebmeds.org). EbMeDS uses predefined decision rules and risk estimation functions based on patient data from the electronic health record, and current best evidence to aid physicians and patients in their shared decision making about optimal therapy. In the PRIMA-eDS-project, diagnoses, medication, and patient baseline data like laboratory test results, renal function, weight, height etc. were transferred to a pseudonomized electronic case report form (eCRF) online which interacted online with the web-based eDS-tool thus providing alerts to the physician, if a specific medication should be avoided or discontinued according to current best evidence. The tool not only supports the GP in the reduction of polypharmacy and inappropriate medication, but also helps to convince the patient when medication should be discontinued for the better, and at the same time provides forensic assurance in the form of background evidence, if discontinuation is contradictory to existing mono-disease guidelines. Objective 2 was the main focus of Work Packages 3-4 of the PRIMA-eDS-project.
1.2.2.3 Objective 3: Evaluation of the eDS-tool
Thirdly, we set out to compare the clinical outcome of existing treatment regimens (applying polypharmacy in the treatment of multimorbidity, i.e. usual care) with the PRIMA-eDS-aided drug regimen reducing polypharmacy and inappropriate prescribing in a multinational, multicentre randomised controlled trial (RCT). From the studies of Garfinkel we have preliminary evidence that the prudent reduction of polypharmacy is beneficial to the patient, but the algorithm Garfinkel used in his study has not been proven to be effective in a randomized controlled trial. There is also limited evidence that Beers’ criteria characterizing inappropriate medication for the elderly are associated with adverse outcome, but this evidence is mostly based on retrospective cohort studies. Similarly, the STOPP-criteria have been shown to be associated with adverse drug events, but it has never been shown convincingly that their application reduces ADEs. In the PRIMA-eDS-trial we set out to prove that the prudential reduction of polypharmacy in the targeted diseases by avoiding inappropriate and non-evidence-based medication in older populations leads to a reduction of the hospital admission rate and adverse events and improves quality of life and mental as well as physical functioning. We intend to achieve the reduction of polypharmacy by eliminating non-evidence based drug therapy and prioritising medication in case of possibly dangerous drug-drug-interactions in a standardized fashion by using the PRIMA-eDS-tool developed for objective 2. Objective 3 was the main focus of Work Packages 5-9 of the PRIMA-eDS-project.
1.2.2.4 Objective 4: Optimization of the eDS-tool and dissemination
Fourthly, we evaluated the PRIMA-eDS-trial and continuously updated and optimized the PRIMA-eDS-tool with the findings of the RCT and continuous evidence-updates. The optimized tool is available as “Comprehensive Medication Review” by Duodecim ltd. and will be disseminated for widespread general utilization in Europe and outside Europe. Objective 4 was the main focus of Work Packages 10-12 of the PRIMA-eDS-project.
Project Results:
1.3.1 Gathering the evidence and development of the PRIMA-eDS-tool
1.3.1.1 Systematic reviews
We prepared a set of systematic reviews (SRs) on the drugs which are the most commonly prescribed to older people and which are associated with an increased risk of hospitalization due to adverse events in this population. The selection of the drugs was done using national prescription databases as well as published studies identified in the literature.
To shorten the review process while maintaining highest quality according to the Cochrane-handbook we developed a distinct methodology. A template for a systematic review study protocol to be used for all reviews carried out within PRIMA-eDS was developed. We then developed Standard Operating Procedures (SOP) for the reviewers to follow, with the aim to standardize our review work.
We aimed at including SRs, meta-analyses, clinical trials and observational studies which included a sufficient number of participants aged 65 years old and older (mean age ≥ 65 or > 80% studies with mean age ≥ 65 or subgroup analysis by age for SRs; >80% participants aged ≥ 65 or subgroup analysis by age for original studies) and which evaluated clinically relevant endpoints. A stepwise approach was carried out for the selection of studies including the following searches: search 1 (systematic reviews and meta-analyses from 2 databases: Cochrane Database of Systematic Reviews [CDSR] and Database of Abstracts of Reviews of Effects [DARE]), search 2 (systematic reviews and meta-analyses from 4 databases: MEDLINE, EMBASE, Health Technology Assessment Database, International Pharmaceutical Abstracts database), search 3A (individual studies meeting inclusion criteria from excluded studies in search 1 and/or 2), search 3B (intervention and observational studies from 4 databases: MEDLINE, EMBASE, Health Technology Assessment Database, International Pharmaceutical Abstracts database). Additional references which were excluded but considered of interest for the development of recommendations were also collected. Each consecutive step of the search strategy was only employed if the preceding step did not yield sufficient results to collate the evidence and derive recommendations.
Each SR has been undertaken by two independent reviewers who have followed the reviews’ protocol and the SOP using piloted excel documents to report the selection of studies, data extraction and quality appraisal. Validated instruments have been used to evaluate the quality of the included studies. Recommendations on when to discontinue or adjust the dose of one of the studied drugs have been developed by the researchers using GRADE methodology. Approximately 2 or 3 team meetings per SR were necessary for the development of recommendations, in which the researchers involved in each SR as well as clinicians and senior researchers took part. Recommendations were developed using a standardized wording schema providing information about the strength of the recommendation and the quality of the evidence. Once recommendations had been developed, they were peer reviewed by a team of experts in evidence based medicine provided by DMP.
Twenty-one SRs on 17 drug classes have been performed. In total, 210 articles have been included (59 SR or MA, 97 CIS, 54 OS), the number of included references per drug class ranging from 0 to 28. The inclusion of studies due to subgroup analysis by age was frequent; information on comorbidity and concomitant use of other drugs was often not available. Often reported outcomes were mortality, hospitalization, cardiovascular events including stroke, adverse drug event, and safety. The quality of the studies included ranged from very low to high.
Forty-two recommendations to discontinue or reconsider the use of the studied drugs or to adjust their doses have been. The quality of the evidence was often downgraded because it was based on subgroup analyses; the strength of the recommendations was most frequently weak. Fifty references were additionally taken into consideration.
Table 1 displays the overview of SRs as well as the information on the number of studies identified in each search, the number of studies included, excluded, considered of interest, and the number of recommendations developed for each SR.
Table 1: Systematic reviews on drug treatment for the development of PRIMA-eDS-stop-recommendations
Legend: The values included in the table are numbers of studies. CG=Clinical Guideline; CR=Cochrane Review; CT=Clinical trials; EC=expert consensus; EO=expert opinion; HTA=Health technology assessment; MA= Meta-analysis; NN=not necessary; NR=narrative review; NSAIDs=Nonsteroidal anti-inflammatory drugs; OS=observational study; PIM= potentially inappropriate medication; RCT=randomized controlled trial; SA=safety assessment; SR=Systematic review; VKA/NOACs=vitamin K antagonists/new oral anticoagulants
Table 2 presents a sample of 12 recommendations developed for 5 drug-indication pairs, in the way they have been incorporated into the tool.
Table 2: Recommendations for a sample of 5 drug-indication pairs derives from the PRIMA-eDS systematic reviews
Recommendation Strength of the recom-mendation/Quality of the evidence
Opioids/Pain
It is suggested not to start transdermal fentanyl delivery system (TTS fentanyl) at a dose of 25 mcg/h or higher for non-malignant chronic pain, especially in older people, because benefits (lower pain, improvement in quality of life) are similar compared to starting with lower doses (12.5 mcg/h) and adverse events are higher. Weak/Low
It is suggested to use the lowest possible dose of oral opioids for the management of chronic non-cancer pain because higher doses (e.g. >=40 mg/d of Morphine - the patient is using [here the eDS-tool automatically inserts the dose the patient is taking]/d) may increase the risk of fractures while there is less confidence of this risk for lower doses. However, adequate pain relief is a priority, and the minimum dose that effectively meets this goal should be used. Weak/Low
It is suggested to discontinue codeine or codeine combinations because codeine may increase the risk of fractures (probably due to falls), cardiovascular outcomes, hospitalizations and all-cause mortality compared to other opioids or non-use of codeine. Weak/Low
Non-steroidal anti-inflammatory drugs (NSAIDs)/Musculosceletal disorders
It is recommended to discontinue the use of NSAIDs because it might increase the risk of cardiovascular events. If necessary, naproxen and the additional use of proton-pump inhibitors (PPI) might be considered if the gastrointestinal risk is high (e.g. history of peptic ulcer, upper gastrointestinal bleeding). If PPIs are used in addition to NSAID, it is recommended to discontinue also PPI after 2 weeks when discontinuing NSAID. Strong/low
Thiazides/Hypertension
It is suggested to reduce high doses of thiazides for the management of hypertension because high doses (50 to 90 mg/day hydrochlorothiazide or equivalent) may be less effective in reducing mortality and coronary artery disease and may be associated with higher risk of gout compared to other antihypertensive medication including low thiazide doses. If the patient has also heart failure, please take the symptoms of heart failure additionally into account. Weak/Low
It is suggested to discontinue both drugs belonging to the combination of hydrochlorothiazide and triamterene in older adults over the age of 80 because the benefits of this treatment may not be established in this age group when compared with placebo and because it may be associated with the development of gouty arthritis, especially in men with high serum uric acid and creatinine levels. Weak/Low
Six of the systematic reviews and the methodology have been published in a peer reviewed journal.
An additional systematic review explored the effectiveness and impact of strategies to reduce polypharmacy (≥4 drugs) on health outcomes, processes of care, health resources, cost-effectiveness, as well as user and patient satisfaction/acceptance in older (≥65 years) patients, with the aim, to optimize the PRIMA-eDS-tool regarding efficacy. A PICOS-framework (population, intervention, control, outcome and study design) was used in order to specify the research questions and their elements used as study selection criteria. We used the GRADE Pro assessment tool to assess the quality of studies reporting on mortality and hospitalization (critical endpoints). In total 25 studies were included in this systematic review (Figure 1: PRISMA Flow Diagram). We identified three main categories of different types of interventions (pharmacist led interventions, physician led interventions and multidisciplinary team led interventions) and three main categories of settings where the interventions were implemented (primary care setting, nursing home setting, hospital setting). Participants’ (appropriate) medication use was reviewed using different methods (tools and instruments), whereof we identified four main methods: checklists (e.g. MAI, the Beers list), drug-drug interactions tools (e.g. software, lists), reconciliation methods and expert opinion. Performance of medication review and the decision criteria were not always described in detail in the included studies. The included studies provide limited evidence due to small sample sizes, poor quality, risk of bias, and short follow-up periods.
Figure 1: PRISMA-Flow-Diagram of the systematic review on strategies to reduce polypharmacy
Remarkable heterogeneity was found between studies. On the one hand, there was clinical heterogeneity in terms of populations, intensity and duration of interventions, outcomes, and follow-up times. On the other hand, there was also methodological heterogeneity such as differences in trial design (17 RCTs, 4 cluster-RCTs and 4 non-randomized controlled trials) and quality. Due to these inequalities a consistent interpretation of the results was seriously limited.
Overall the interventions had very limited effect regarding our outcomes of interest. Neither any single study nor our meta-analysis of effects on all-cause mortality during the study period showed any significant effect in favour of the intervention group (OR: 1.02 95 % CI 0.84 to 1.23) (see fig. 2).
Figure 2: Metaanalysis of the primary endpoint (mortality) of the systematic review on strategies to reduce polypharmacy
Only single studies found improvements in hospitalization in favour of the intervention group. The weighted mean including all prescribed drugs at baseline was in both groups 7.4 drugs per patient. At follow-up the weighted mean of drugs was reduced (-0.2) in the intervention group while it increased (+0.2) in controls but the difference did not reach significance. Further 12 outcome measures of interest were analysed in this SR but only single studies revealed significant differences between intervention and control (probably due to multiple testing and chance).
The evidence regarding effectiveness of interventions in the included studies is rather weak. It is unclear how to ideally organize and implement interventions (pharmacist led, physician led or led by a multidisciplinary team) in order to achieve clinically significant improvements in multimorbid older patients with polypharmacy. There is a great need of large long-term RCTs exploring the effect of strategies to reduce polypharmacy on clinically relevant patient outcome measures such as mortality and hospitalization. Thus the PRIMA-eDS trial is well justified. The harms of polypharmacy are well documented in the scientific literature, but it is not proven yet that interventions to reduce polypharmacy have a positive effect on patients’ outcome. Furthermore best practice models to reduce polypharmacy are still lacking. Due to the demographic development, the burden of chronic diseases and polypharmacy will increase. Polypharmacy is an emergent problem and a societal and economic challenge for most health care systems.
1.3.1.2 Building up the PRIMA-eDS-tool
In addition to the evidence derived from the systematic reviews carried out within the PRIMA-eDS-project, existing evidence was used to build the PRIMA-eDS-tool. DMP produced a list of recommendations which were relevant to older people under polypharmacy developed from the existing EbMeDS data base. Researchers with medical background from UWH gave feedback to this list of recommendations and an agreement was reached about important recommendations for older people. Thus, DMP identified a total of 1638 evidence summaries from EBM Guidelines (www.ebm-guidelines.com or www.ebm-guidelines.at) that were relevant for pharmacotherapy in older individuals. Selected evidence summaries were used as a basis of recommendations along with the results of the systematic reviews.
A total of 147 individual rules were included in the eDS tool, containing more than 200 different messages. 42 recommendations (messages from the rules) were developed on the basis of the systematic reviews. A general rule reminding that a drug is on the European PIM list was developed. Specific rules for several of the drugs on the PIM list were developed to guide the use of the drugs more accurately. The rest of the included rules were derived and (if necessary) modified from the existing rule set of EBMeDS. Descriptions of the rules and their evidence-base are available at http://www.ebmeds.org/web/guest/scripts?query=prima-eds+&lang=en.
We then incorporated the databases of indications, contraindications, dosing, drug interactions (SFINX®), renal dosing (Renbase®), and adverse effects (PHARAO®) in the PRIMA-eDS-tool. Each of these has a rule that enables the use of the database as part of the eDS tool. SFINX®, Renbase®, PHARAO® are developed and maintained by Medbase Ltd. (www.medbase.fi). Before use in the eDS, modifications were made to the databases by inactivating alerts on drug interactions of low significance, and renal dosing alerts in mild renal dysfunction for drugs. Specific rules were developed for several classes of drugs in Renbase® and in SFINX® to guide decision making more accurately than standard drug database alerts.
The total numbers of different feedback messages from the drug databases included in the eDS tool is 30171 (Table 3) which is about 30% more than anticipated in the project plan. All the messages are available in English, German and Italian.
Table 3: Messages in drug databases included in the eDS tool
Drug – drug Interactions 14565
Drug and Renal Malfunction 5024
Drug Contraindications 3144
Drug Indications 4141
Dosage Warnings 3297
Total 30171
A database of laboratory test results relevant for medications was developed on the basis of EBM Guidelines. The database contains reference values of tests related to medications, and recommended follow-up intervals of tests in safety monitoring of drug therapy.
Rule descriptions, rule logic, and executable rules were then developed using the EBMeDS content authoring tool. The testing environment of the EBMeDS content authoring tool was used to test each rule with varying test patient data by a minimum of three people independently of each other before publishing.
Feedback from pilot physicians participating in the study as well as from Finnish users of the eDS tool were analyzed and changes were made to some rules. Of the drug databases, the most extensive changes were made to the indications database where a number of coded indications were added, mostly because missing ICD-10 codes and variable use of the codes by the participating physicians.
The frailty scale on the eCRF was used as input variable to guide some decisions suggested by the rules.
1.3.1.3 Electronic case report form (eCRF) and interfaces
With the aim to make the tool accessible to physicians and to gather data for the PRIMA-eDS randomized trial, we developed an electronic case report form (eCRF) in cooperation with Avain Technologies ltd. The forms application of Avain Technologies Ltd. that is used as the platform for the eCRF is conformant with the Medical Device Directive (class 1 software product).
Search engines for drugs (ATC codes) and diagnoses (ICD-10 codes) were developed, and updated on the basis of user feedback during the first phase of (baseline) data entry of the study. The SF-12 form and the frailty scale were implemented as parts of the eCRF. The functions for outcome data entry require the data to be checked and reported at baseline and at every scheduled visit by asking the users to enter the data and tick a box that the data has been updated.
The functionality of the eCRF form was finalized for the study. Several changes to the original version were made on the basis of the pilot testing. Data export from the eCRF was performed and tested by the statistical team of the consortium.
A screenshot of the first page of the eCRF is shown in figure 3 (German version, the eCRF was alos available in English and Italian for the respective study centres. A print out of an eCRF containing patient data of a test-patient is depicted in figure 4.
Figure 3: Screenshot of the eCRF
Figure 4a: Printout of the eCRF, part 1
Figure 4b: Printout of the eCRF, part 2
The eCRF was then connected to the eDS-tool via a specific interface. The interface between the eCRF and eDS has been developed in cooperation with Avain Technologies Ltd. and is based on XML messaging. When data entry on the form has been completed, and the user clicks the "perform medication review" button, the form creates an XML message that is received by the eDS service and analysed using the rules and data tables created for the purpose. The response time of the eDS service is well under 2 seconds. eDS returns an XML message containing suggestions to modify the patient’s medications or to improve care by monitoring of potential adverse effects of the medication. The suggestions are then shown in the user interface of the eCMR. A screenshot of the eCMR is shown in fig. 5. The CMR could also be saved in pdf-format and printed if necessary or desired.
Figure 5a: User interface of the eDS tool – part 1.
Figure 5b: User interface of the eDS tool – part 2.
All transactions between the eCRF and the EBMeDS (eDS) service were stored in a log file for further analysis. For the control group, the recommendations of the eDS tool were recorded in the log file, but were not shown to the clinician.
Each time the form was updated by the clinician, it was stored on a dedicated server hosted in Finland. When the patient came for a follow-up visit, the form containing previously entered data was collected from the server. The user updated the data and entered the results of the patient assessment. The updated version of the eCRF was then stored on the server.
The data and backup copies of the data (with patient study identification codes created at baseline) were stored on the safe servers of Avain Technologies. Anonymous data for analyses were exported and delivered to be analysed by the statistical team of the consortium.
eCRF and data storage on a central server were protected by special security measures provided by Avain Technologies Ltd. Only authorized users (the clinicians participating in the study and researchers monitoring the study) were able to access the data on the forms. User rights administration by Avain assured that any GP was only able to access data of his own patients. Because confidentiality and safe transfer of the patient data on the form was an absolute requirement, an independent external information security audit has been performed, and an archives certificate has been awarded for Avain Technologies (Certificate number 6692-04, by Inspecta).
1.3.1.4 Pilot study and usability studies
Before starting the PRIMA-eDS-trial, a pilot study was carried out to test and optimize the functionality of the eCRF and the eDS-tool. The main subject identified was the expenditure of time needed for entering patient data into the eCRF. The eCRF form was also perceived as too complex and not simple and intuitive enough to use, especially for those users who do not have much technical experience. Study assistants reported a list of minor bugs and errors back to the DMP team so that they could be corrected as soon as possible. Despite the difficulties encountered when using the study tools, most GPs that have taken part in the pilot study expressed their desire to use the CMR in their practice. Several changes to facilitate data entry were suggested and implemented. For instance, the eCRF was adapted to account for faster data entry and better operability. The CMR output page was adapted and optimized several times to offer compact and fast information to the physician. Usability of the tool have been evaluated in qualitative and quantitative studies, and the results have been incorporated in the tool and the eCRF.
1.3.2 Evaluation of the PRIMA-eDS-tool in a randomized controlled trial
1.3.2.1 Study protocol, ethics approval and trial registration
In parallel to the development of the tool, the PRIMA-eDS randomized controlled trial was planned and initiated. The study protocol has been finalized including final sample size calculation. In 2016 the study protocol was published in a peer reviewed journal. Ethics approval for the coordinating study centre (UWH) was granted on December 3rd, 2013. The study has been approved unconditionally by the ethics committee of Salzburg on the 8th of April 2014 for PMU. The study was approved by the Ethics Committee of the Rostock University Medical Centre on February 14th, 2014 (Approval-No. A 2014-0020). Ethics approval for SAGP was granted on the 19th of June 2013 by the Ethics Committee of Belluno. Ethics approval for UNIMAN was obtained on the 6th of June 2014 at the Ethics Committee of NRES Committee North West - Greater Manchester East. Due to extension of the study to other research networks because of recruiting problems we had to obtain additional ethics approvals of local authorities in the UK and of the Bavarian Chamber of Physicians (Bayerische Ärztekammer) which were granted before the start of the trial. Trial registration was performed July 31, 2014 with Current Controlled Trials ltd (http://www.controlled-trials.com/ISRCTN10137559).
1.3.2.2 Physician recruitment
Due to regional differences regarding the health system and organisational differences in primary healthcare, each study centre had to develop its own standard operating procedures regarding recruitment. To account for these differences in the trial and to avoid bias due to the existing differences between countries, randomisation was stratified by country. The procedures of GP recruitment were as follows:
UWH
Recruitment of GP practices took place in three waves. We obtained lists of all practicing GPs under statutory health insurance of the Westfalen-Lippe region and adjacent regions from the “Kassenärztliche Vereinigung Westfalen-Lippe” and the “Kassenärztliche Vereinigung Nordrhein”. We informed the GPs in writing about the PRIMA-eDS trial and asked for a response by fax expressing interest to participate. In total we contacted 1192 physicians. The overall first response rate expressing interest was approximately 10 %, finally 76 GPs recruited patients (response rate 6.4%).
UMR
At the Rostock study centre, recruitment of GPs via mail started in February 2014. In May 2014 a total of 74 GPs decided to enroll for the study. Unfortunately, between June 2014 and April 2015 18 of these GPs changed their mind and withdrew from the study. Reasons included e.g. a loss of interest, or too long waiting time until the start of the trial. Rostock study centre therefore decided to further recruit in the area of Mecklenburg-Western Pomerania as well as in Berlin. Over the entire recruitment period a total of 1.730 GPs from the regions Mecklenburg Western-Pomerania (1.120 GPs), Schleswig-Holstein (50 GPs) and Berlin (560 GPs) have been contacted over seven waves. Positive feedback and interest in the study was declared by 175 GPs. After telephone contact a total of 87 GPs (5.0%) finally took part in the trial.
SAGP
SAGP recruited GPs in the neighboring region of Veneto, Italy. The local collaboration partner of SAGP was SVeMG (Scuola Veneta di Medicina Generale = School of General Practice in Veneto), which has extensive history and experience with the performance of scientific projects in the primary care setting. SVeMG supported SAGP in all aspects of recruitment of GPs and later in the recruitment of patients for PRIMAeDS. Italian GPs for PRIMAeDS were recruited within the Primary Care Health Research Network of Veneto “ MilleInRete”. All GPs of the network were invited by email to participate in the project. The invitation contained information about the project, requirements for participation and an attached participation form, which GPs were asked to sign in case of interest. Seventy GPs replied and declared to be interested in participating in PRIMAeDS by sending us the signed participating form. Finally, 72 Italian GPs signed the informed consent and received eCRF training.
PMU
The recruitment strategy of PMU was planned stepwise and manifold due to the fact that we knew from previous experiences that recruitment of study practices in the area is a demanding challenge. We informed GPs via letter, via mail and via fax in different regions (Austria, Bavaria) in several waves through different providers such as the PMU, general practitioners federation in Bavaria, laboratory community, general practitioners society in Salzburg (SAGAM). Written information was followed by phone calls. Common reasons for not participating were lack of time, not treating enough polypharmacy patients, and software (medical health record) in the surgery was not suitable for patient identification (e.g. lack of filter function). The PRIMA-eDS study including participation information was published in a regional journal of the Physician’s chamber of Salzburg (Medium 11/2013, circulation of 3.600); and in a national journal, ÖGAM News (06/2015). Several face-to-face presentations at different occasions were held: e.g. project information event at Paracelsus Medical University, Salzburg (10/2013), external GP meetings “Bezirksärzte-Fortbildung Salzburg” (03/2014). An additional effort was an advertisement of the study via the Chamber of Physicians (Ärztekammer) Salzburg homepage (07/2014). Several personal, phone and mail contacts were performed to recruit GPs via “snow-ball strategy”. In summary, all efforts during the whole recruitment period resulted in about 80 GPs interested in the study. Eventually sixty-seven GPs signed the contract to participate; however eight withdrew their consent before randomization, and thus only 59 GPs finally participated. PMU recruitment strategy was presented at the annual meeting of the German College of General Practitioners and Family Medicine 2015.
UNIMAN
UNIMAN collaborated with five National Institute for Health Research (NIHR) Clinical Research Networks (CRN). The five CRNs were Eastern, North Eastern, Wales, North West Coast and Greater Manchester. The CRNs are widely spread across four different UK regions, the North West, North East, Wales, and the South East. From these, 76 practices provided expressions of interest from which 67 signed up for the trial and recruited patients.
1.3.2.3 Physician instruction
Just like physician recruitment, GP instruction for the trial had to beorganized by each study centre separately due to differences in local conditions.
UWH
Instruction of GPs was carried out using several approaches to reach a maximum effectiveness. It differed between intervention and control group. All GPs were offered general information about the RCT in writing and by phone, Instruction on using the eCRF via personal visit of the practice by a study nurse, and by providing a manual on eCRF-usage as well as a video on eCRF-usage. Both the manual and the video were made available on the PRIMA-eDS-website in English, German and Italian for all GPs of the PRIMA-eDS-trial in all study centres. We also provided continuing instruction via a telephone hotline which was implemented specifically for the study.
In addition, all GPs of the intervention group received instruction on the use of the PRIMAS-eDS-tool and on the shared decision making process of reducing polypharmacy in collaboration with the patients. We provided a manual and a video on PRIMA-eDS-tool usage and shared decision making. The manual and the video were made available in English, German and Italian for all GPs of the intervention group of the PRIMA-eDS-trial in all study centres. Furthermore, all GPs of the intervention group were offered participation in a webinar explaining and detailing eDS-tool usage and shared decision making (duration approximately 45 min. The webinar was offered to all German speaking GPs of the intervention group, also in the other study centres). Personal visits by a study nurse to explain the eDS-tool were offered but not widely made use off. A telephone hotline during regular office hours was implemented to provide continuous support on eDS-tool usage and shared decision making.
UMR
Instructions on proper use of the eCRF were provided by the study personnel during the initial visits of the GPs. After randomization, GPs of the intervention group received a tutorial for proper use of the eDS-Tool (DVD) and written information. In addition, via telephone a physician was ready to answer all questions of the GPs. The manuals, webinars and videos described under “UWH” were also offered to the GPs.
SAGP
eCRF training: The research personnel of SAGP trained all Italian PRIMAeDS GPs in using the eCRF. eDS training: eDS training meetings were held for GPs of the intervention group. Individual eDS training was provided for GPs not being able to participate in the training meetings. All PRIMA-eDS meetings (regarding information, recruitment, eCRF-and eDS- training and feedback) with the participating GPs were held in collaboration with SVeMG at its headquarters in Caselle di Selvazzano, PD. The manuals and videos described under “UWH” were also offered to the GPs.
PMU
Training in the usage of eCRF and eDS was adapted to the regional setting. Because of regional characteristics (e.g. GP practices are foremost organized as single handed offices), workload and limited time resources, participating GPs refused to attend group training sessions in a centralized location. Project staff delivered the training face-to face within each practice in a two to four-hour session covering identification of eligible participants and entering data in the eCRF. As the eCRF was not ready for use when we started the GP recruitment the first GP visits focused on identification of eligible participants only. Patient identification was started as soon as possible as the procedure had to be performed via manual search by the project staff due to limited software options in regional GP offices. After randomization GPs in the intervention group were trained on how to use eDS and eCRF within a second visit by project staff in the respective offices. GPs in the control groups were instructed via telephone, and the eCRF manual mentioned above was sent to the GP. To support the participating GPs, we installed a telephone hotline for PRIMA-eDS GPs.
UNIMAN
Training in the eCRF and eDS was adapted to the UK context. Because UK GPs are only undertaking Medication Reviews, not the non-Medication Review tasks required for the trial (approaching and consenting patients, completing patient questionnaires and completion of the eCRF), the GP recruitment process included negotiating whether CRN, practice or research team staff would complete the non-Medication-Review tasks, as agreed in the original CRN negotiations. Therefore some parts of the training given to GPs in other countries, in the UK was delivered to other staff groups (mainly nurses). In the Eastern and North Eastern CRN areas peripatetic Research Nurses employed by the CRNs cascaded training to GPs and carried out the non-Medication Review tasks themselves. The study team provided training days in Cambridge and Newcastle for the teams of Research Nurses there, with hands-on training in both non-Medication-Review tasks and Medication Reviews for Intervention practices and only non-Medication-Review tasks for Control practices. In Wales, North West Coast and Greater Manchester all training was delivered by the study team to practices directly. Face-to-face non-Medication-Review training was completed with the CRN/practice personnel identified when the practices were recruited, and timed to facilitate patient recruitment. Medication Review training was undertaken post-Randomisation, initially face-to-face, then by teleconference for speed only for the intervention practices.
1.3.2.4 Patient recruitment
Like GP- recruitment, patient recruitment hat to be organized locally due to different local conditions.
UWH
GPs were instructed to perform searches in their electronic health records to identify eligible patients (patients ≥75 years of age taking ≥8 drugs). Eligible patients were informed about the study by the GP in writing or personally when entering the practice at the discretion of the GP. If more than 15 eligible patients were willing to participate in the study, GPs were asked to randomly select patients or recruit a random sample by continuous recruitment of patients coming to the practice. GPs were asked to recruit at least eight and no more than 15 patients to assure uniformity of cluster size, but GPs with less than 8 or more than 15 patients were not excluded. After signing informed consent, baseline data of the patients were entered into the eCR. SF-12 was performed on paper, and the filled out forms were sent to the UWH study centre for data entry. After completing recruitment, GPs were asked to inform the study centre. To assure concealment of allocation during recruitment, GPs were randomized after the completion of recruitment. After randomization GPs could not include additional patients in the study. Recruitment ended September 30, 2015, and all GPs not randomized yet were randomized on Oct 5, 2015, no matter how many patients they had recruited. In total, 742 patients were recruited by 76 GPs for the UWH study centre.
UMR
Based on standard operating procedures eligible patients were selected from the electronic health record of the GP practices. The list was presented to the GP and he/she was asked to exclude patients not suitable for the study (exclusion criteria). A random sample of a maximum of 30 patients was then selected from the remaining list. These patients received an invitation letter to participate in the study. Patients interested in participating contacted their GP and made an appointment to receive further information, to sign informed consent and to record baseline data. In total, at Rostock study centre 987 patients have been included by 87 GPs.
SAGP
Patients were selected from the electronic health records of the GP surgeries through two special software querys, according to the inclusion criteria of PRIMA-eDS. This was feasible, because all members of the research network, in which we recruited GPs, use the same practice software MilleWin, which is able to extract the patients according to the PRIMAeDS inclusion criteria. For the selected patients, GPs had to apply exclusion criteria one by one. The remaining patients were put in random order by blinded research-personnel of SAGP. The first 30 patients of this random list of eligible patients were invited by letter to take part in the study. All GPs were asked to recruit 11 patients. In total, 70 GPs recruited 905 patients for the PRIMA-eDS-trial with a maximum of 15 patients and a minimum of 8 patients per GP.
PMU
All eligible patients were selected by an electronic search of the electronic health records in the respective GP practice. The list of all eligible patients was numbered serially and brought in random order. Following the random list, the first 30 patients fulfilling inclusion criteria were identified by manual search. Invitation letters for the selected patients were prepared by project staff during the first visit at the GP surgery. In the invitation letter patients were informed about the project and asked to make an appointment at their GP-practice for further information and to sign informed consent. The GP decided whether to use the invitation letters or personally contact patients or both. Baseline data was obtained for the eCRF after informed consent was signed by the patient. As soon as the last patient consented to participate, the respective GP sent a confirmation fax. The project team checked whether all patients met inclusion criteria and submitted a randomization request for the respective practice to the randomization centre. In total, 587 patients were recruited by 59 GPs.
UNIMAN
A total of 702 patients have been recruited from across 67 practices. Four of these patients were subsequently found to be ineligible for the trial, leaving a total eligible sample of 698; very close to the target of 735 patients from 67 practices. Patients were recruited via practices, who posted an invitation pack (letter on practice stationery, study information, reply slip and prepaid envelope) to selected patients identified from the practice list. Interested patients met with a CRN/practice nurse/university researcher who explained the study further, gave the opportunity to ask questions and to consider participation further. Patients who were still interested in the study signed the informed consent form and with this permission could then complete the patient questionnaires and have their data entered into the eCRF.
1.3.2.5 Randomization
A statistician created lists for blockwise randomization stratified by study centre. GP practices were then allocated to either the intervention group or the control group when they had completed patient recruitment. Cluster randomization was performed by the Department of Medical Informatics, Biometry and Epidemiology of the Ruhr-University Bochum (AMIB) starting January 10, 2015 and ending October 5, 2015. Weekly randomization reports were provided by AMIB.
1.3.2.6 Blinding
To minimize bias and confounding, all patients were blinded, i.e. patients were not informed explicitly whether their doctor was part of the intervention group or the control group. Inherent to the study protocol general practitioners could not be blinded. Also, it was not possible to blind all research-personnel. As research-personnel did not interact directly with patients or GPs, the risk of bias due to the lack of blinding appears to be negligible. The statistician performing final analysis of the trial was blinded regarding GPs belonging to the intervention or the control group.
1.3.2.7 Carrying out the trial
Intervention
After completion of recruitment and randomization, the GPs of the intervention group were instructed to perform the medication review with their patients using the PRIMA-eDS-tool. GP-instruction was carried out via face-to-face-instruction by study personnel as well as video-instruction provided via the PRIMA-eDS website (www.prima-eds.eu) and webinars held by study personnel. A hotline was available throughout the study, if GPs encountered technical or medical problems using the tool. In the instruction, special emphasis was put on a shared decision making process to come to an agreement with the patient regarding discontinuation of inappropriate medication. The GPs of the control group received instructions and continuous support on the use of the eCRF, and were advised to prescribe medication to their patients as usual according to current guidelines.
Recording outcome measures and assessment of data
GPs performed the baseline visit and documented patient data in the electronic case report form (eCRF) after recruitment and informed consent, but before randomization to assure concealment of allocation. All GPs were instructed to complete any baseline data still missing after randomization. The medication review was then carried out using the PRIMA-eDS-tool in the intervention group while the control group checked patient medication as usual. GPs were reimbursed for recruitment and baseline data acquisition after completing all their baseline examinations. For the follow up examinations scheduled after 8, 16 and 24 months, GPs were reminded to prepare/perform these follow-ups and document all patient data as required by the eCRF. Letters and faxes were sent to the GPs and followed up by telephone calls if necessary. We allowed a time frame from 4 to 12 months after randomization for follow up 1 (FU1), from 12 to 20 months for follow up 2 (FU2), and from 20 to 28 months for follow up 3 (FU3).
The project team continuously supported GPs to assure the conduction of the intervention (e.g. GP instructions on how to use eCRF/eDS correctly, data collection and data entry of SF-12, baseline and first follow-up examination) in the intervention group and to assure correct completion of the eCRF in the control group.
Data management and statistical analysis
All data were stared securely by Avain Technologies ltd. Anonymized data files were transferred to AMIB on a weekly basis for data cleaning and quality check. Any deviations and errors in data recordings were reported to the monitors who in turn contacted the respective GP for data correction or clarification.
A statistical analysis plan was developed by AMIB in cooperation with UWH and UNIMAN.
Monitoring, quality assurance, and Safety and Data Monitoring Committee (SDMC)
To ensure safety of all patients throughout the study, an independent data and safety monitoring board (SDMC) was set up. The SDMC met periodically throughout the study. The members reviewed and discussed the current data (at baseline and at the various follow-up visits. The SDMC was informed about dropouts, deaths, hospitalisations, and any other relevant safety issue if applicable. The main focus of the interim analyses presented to the SDMC was, to detect discrepancies in relevant safety parameters and their effects on the (non-) achievement of the outcome parameters. All interim analyses were kept blind regarding events in the intervention or the control group. The SDMC received a blinded descriptive synopsis of the most important variables (especially primary outcomes such as deaths, hospitalizations, reported symptoms related to changes in medication, dropouts, monitoring results etc.) in preparation for their regular conferences. The SDMS checked the monitoring of the study and approved of the continuation of the study. The SDMC at no point had any concerns about patient safety throughout the trial, and a discontinuation of the trial was never recommended.
The designated site monitors at each study site performed the local monitoring activities and reported their results to the coordinating monitor at UMR. Throughout the study, site monitors established and maintained contact to general practitioners and practice teams of the respective study site, clarified issues and difficulties during the monitoring contact and prepared status reports on completed on- and offsite visits. All monitoring activities were coordinated and reviewed by UMR including the distribution of relevant information and the review of all monitoring reports provided by the site monitors.
Overall, vivid and frequent communication as well as sharing of experiences between site monitors and coordinating monitors were carried out, mainly via individual phone calls, monthly phone conferences, e-mail, and at steering-committee- and project meetings. In order to facilitate a transparent and quick communication, the online-forum freedcamp.com was used by all monitoring personnel, not only to forward and share documents and information, but also to explicitly exchange experiences between site monitors of all study sites. Additionally, the coordinating monitors regularly reported on the current monitoring trends and results to the consortium via oral presentations held at the project meetings or telephone conferences.
To assure correct and standardized monitoring as well as data quality, a monitoring manual was developed and updated during the study. The monitoring manual and its updates were delivered to every cooperating trial site and all beneficiaries via e-mail. The monitoring manual served as a quality plan that complies with Good Clinical Practice (GCP) guidelines. Additionally, monitoring documentation forms (checklists, report forms, log- & communication-sheets, etc.) and standard operation procedures (SOPs) to support site monitors performing standardized local monitoring activities were developed and adapted throughout the study. Before monitoring started, two online monitoring trainings were conducted to prepare site monitors for the onsite- and offsite-visits at Baseline- (T0) and FU1-(T1) Monitoring. Site monitors and the coordinating monitors discussed all arising questions. Baseline monitoring was performed from April 2015 until April 2016. Altogether, site monitors conducted onsite-visits in 51 of 359 randomized practices and checked 3,919 out of 3,919 Informed Consent Forms during baseline-monitoring. 15 patients had to be excluded due to a lack of a signed informed consent form or withdrawal of consent before randomization, leaving 3904 patients in the intention-to-treat study-population. The most time-consuming aspects mentioned by the site monitors were the search for source data due to huge differences in the quality of documentation within the normal practice routine. Written documentation in practices often lacked information, especially regarding medication, symptoms, start & stop dates, dosage and diagnoses.
FU1-(T1)-monitoring was conducted from March to October 2016. In total, site monitors performed offsite-visits in 80 of 348 included practices at T1. For two random samples of 10% each of the practices, site monitors checked 558 out of 3,904 included patients at T1. Additionally, 100% of all dropouts were monitored. All conspicuous cases detected were reported by using the respective monitoring forms.
The final monitoring of FU3 (T3) was conducted from October 2016 to November 2017. Overall, 3,210 patients in 342 practices were monitored for primary endpoints during the offsite-visits at T3. Moreover, 656 reported dropouts were monitored regarding the primary endpoints (hospitalization, death). Additionally, further items (e. g. medication) were checked for three patients each in 10% of all monitored practices. The majority of GPs and practice personnel were cooperative during monitoring, but some complained about the time needed. For 694 patients of the intention-to-treat-population, monitoring could not be completed or performed at all due to refusal of the GP, drop out, lost-to-follow-up or other reasons. These patients were included in the intention-to-treat analysis with their respective known time-under-risk (date of last available data on hospitalization or death). All detected discrepancies were documented in the respective monitoring documents and discussed with the coordinating monitor.
1.3.3 Main results of the PRIMA-eDS-trial
Baseline characteristics of the study sample are presented in tables 4-7. On average, participants were taking 10.5 substances (±2.4) and had 9.5 diagnoses (±4.9). HMG CoA reductase inhibitors were the most commonly used drug, followed by proton pump inhibitors, selective beta blocking agents, platelet aggregation inhibitors, ACE inhibitors, loop diuretics, and dihydropyridine derivatives. Essential (primary) hypertension was the most common diagnosis, followed by disorders of lipoprotein metabolism, diabetes mellitus type 2, and arthrosis.
Table 4 Demographic and clinical characteristics of the study population
Characteristics All subjects
(n=3904) Control group (n=1951) Intervention group (n=1953)
N N N
Sociodemographic data
Age
<85 (n, %) 3036 77.8 1511 77.5 1525 78.1
≥85 (n, %) 868 22.2 440 22.6 428 21.9
mean ±SD (years) 3904 81.5 ±4.4 1951 81.5 ±4.5 1953 81.5 ±4.4
Female, n (%)
2240 57.4 1137 58.3 1103 56.5
Male, n (%) 1664 42.6 814 41.7 850 43.5
Educational level, n (%)*
Low 1536 39.3 748 38.3 788 40.4
Medium 1465 37.5 726 37.2 739 37.8
High 577 14.8 292 15.0 285 14.6
Health-related factors
Smokers, n (%)* 154 3.9 88 4.5 66 3.4
BMI, n (%)*
BMI <18.5 34 0.9 19 1.0 15 0.8
BMI 18.5-24 957 24.5 474 24.3 483 24.7
BMI 25-29 1606 41.1 790 40.5 816 41.8
BMI ≥30 1307 33.5 668 34.2 639 32.7
Frailty level, n (%)*
Managing well 1643 42.1 771 39.5 872 44.6
Vulnerable 868 22.2 442 22.7 426 21.8
Mildly frail 660 16.9 362 18.6 298 15.3
Moderately frail 505 13.0 250 12.8 255 13.1
Severely frail 97 2.5 53 2.7 44 2.3
Very severely frail 8 0.2 5 0.3 3 0.2
Physical health composite score, median (range)*
3484 36.6
(10-68) 1710 36.6
(10-63) 1774 36.7
(11-68)
Mental health composite score, median (range)*
3483 48.7
(12-76) 1710 48.7
(12-76) 1773 48.7
(12-72)
Study centre, n (%)
PMU 587 15.0 295 15.1 292 15.0
UMR 981 25.1 506 25.9 475 24.3
UWH 742 19.0 369 18.9 373 19.1
SAGP 901 23.1 450 23.1 451 23.1
UNIMAN 693 17.8 331 17.0 362 18.5
Substances, n (mean ±SD) 3904 10.5 ±2.4 1951 10.5 ±2.4 1953 10.6 ±2.5
Diagnoses, n (mean ±SD) 3904 9.5 ±4.9 1951 9.7 ±5.4 1953 9.3 ±4.4
Legend: SD= Standard deviation, BMI= Body Mass Index, N= number; CG= control group, IG= intervention group, * the total n given above does not correspond to the n within this variable due to missing data.
Table 5 Percentage of the population using substances (ATC level 4)
Substances All subjects Control group Intervention group
N (%) N (%) N (%)
C10AA HMG CoA reductase inhibitors 2479 63.5 1254 64.3 1225 62.7
A02BC Proton pump inhibitors 2328 59.6 1152 59.1 1176 60.2
C07AB Beta blocking agents, selective 2240 57.4 1088 55.8 1152 59.0
B01AC Platelet aggregation inhibitors excl. heparin 1952 50.0 977 50.1 975 49.9
C09AA ACE inhibitors, plain 1751 44.9 860 44.1 891 45.6
C03CA Loop diuretics 1715 43.9 838 43.0 877 44.9
C08CA Dihydropyridine derivates 1611 41.3 836 42.9 775 39.7
C09CA Angiotensin II antagonists, plain 1355 34.7 707 36.2 648 33.2
C03AA Thiazides, plain 1354 34.7 716 36.7 638 32.7
A11CC Vitamin D and analogues 1120 28.7 548 28.1 572 29.3
Legend: CG= control group, IG= intervention group
Table 6 Most common diagnoses of the study population
Diagnoses All subjects
N (%) N (%) N (%)
I10 Essential (primary) hypertension 3428 87.8 1708 87.6 1720 88.1
E78 Disorders of lipoprotein metabolism 2078 53.2 1060 54.3 1018 52.1
E11 Diabetes mellitus type 2 1850 47.4 925 47.4 925 47.4
M19 Osteoarthritis 1752 44.9 821 42.1 931 47.7
I25 Chronic ischaemic heart disease 1473 37.7 719 36.9 754 38.6
M54 Dorsalgia 1442 36.9 715 36.7 727 37.2
I48 Atrial fibrillation and flutter 1172 30.0 580 29.7 592 30.3
I50 Heart failure 1142 29.3 536 27.5 606 31.0
K21 Gastro-oesophageal reflux disease 982 25.2 497 25.5 485 24.8
F32 Depressive episode 853 21.9 416 21.3 437 22.4
Legend: CG= control group, IG= intervention group
Table 7 Demographic and clinical characteristics per study centre
Characteristics
PMU UMR UWH SAGP UNIMAN
N N N N N
Sociodemo-graphic data
Age
<85 (n, %)
429 73.1 771 78.6 568 76.5 712 79.0 556 80.2
≥85 (n, %) 158 26.9 210 21.4 174 23.5 189 21.0 137 19.8
mean ±SD (y) 587 82.2 ±4.6 981 81.5 ±4.3 742 81.6 ±4.5 901 81.3 ±4.4 693 81.1 ±4.4
Female, n (%) 362 61.7 606 61.8 415 55.9 521 57.8 336 48.5
Male, n (%)
225 38.3 375 38.2 327 44.1 380 42.2 357 51.5
Educational level, n (%)*
Low
230 39.2 152 15.5 245 33.0 598 66.4 311 44.9
Medium
222 37.8 608 62.0 346 46.6 92 10.2 197 28.4
High
59 10.0 217 22.1 132 17.8 21 2.3 148 21.4
Health-related factors
Smokers, n (%)*
11 1.9 45 4.6 23 3.1 44 4.9 31 4.5
BMI, n (%)*
BMI <18.5
5 0.8 8 0.8 8 1.0 7 0.8 6 0.9
BMI 18.524
174 29.6 195 19.9 157 21.2 255 28.3 176 25.4
BMI 25-29
224 38.2 384 39.1 316 42.6 385 42.7 297 42.8
BMI ≥30
184 31.4 394 40.2 261 35.2 254 28.2 214 30.9
Frailty level, n (%)*
Managing well
245 41.7 415 42.3 255 34.4 376 41.7 352 50.8
Vulnerable
124 21.1 229 23.3 178 24.0 170 18.9 167 24.1
Mildly frail
91 15.5 158 16.1 125 16.8 171 19.0 115 16.6
Moderately
frail
84 14.3 139 14.2 144 19.4 92 10.2 46 6.6
Severely frail
21 3.6 28 2.9 19 2.6 23 2.5 6 0.9
Very severely
frail
2 0.3 0.2 4 0.5 0 0.0 0 0.00
Physical health composite score, median (range)* 509 36.7 (13-62) 948 35.9 (12-63) 713 35.0 (11-62) 675
38.7 (11-68) 639
37.5 (10-60)
Mental health composite
score, median (range)* 506 47.6 (16-74) 946 48.6 (12-72) 715 47.5 (17-76) 676 46.8 (14-72) 640 53.5 (16-74)
Substances
<8 259 44.1 351 35.8 334 45.0 439 48.7 261 37.7
≥10 328 55.9 630 64.2 408 55.0 462 51.3 432 62.3
n (mean ±SD)
587 10.4 ±2.3 981 10.9 ±2.7 742 10.4 ±2.4 901 10.0 ±2.0 693 10.9 ±2.7
Diagnoses, n (mean ±SD)
587 8.5±3.5 981 12.9 ±6.2 742 10.4 ±4.5 901 7.6 ±2.7 693 6.8 ±3.5
Legend: N= Number SD= Standard deviation, BMI= Body Mass Index, * the total n given above does not correspond to the n within this variable due to missing data.
Figure 6 depicts details of the consort flow chart. 3904 patients were randomized and considered for the ITT analysis. 3492 patients were included in the per-protocol analysis.
Figure 6: Consort-Flow Chart Intention to treat analysis and per-protocol analyses
Abbrevations: ITT= intention to treat; PP= per protocol, IG= intervention group, CG= control group
In the intention-to-treat analysis at follow-up 3 (after 24 months), the hazard ratio (HR) for the combined endpoint for all study centers was 0.92 (95% CI 0.82-1.04) thus just missing significance (see table 8). The time-to-even-graph (survival-plot) of the composite primary endpoint shows a clear trend in favor of the intervention (see fig. 4). When the combined endpoint was stratified by research center, PMU had a significantly lower hazard ratio between intervention group and control group in favor of the intervention [0.68 (95% CI 0.50-0.94) than UWH, SAGP, UNIMAN, and UMR (see table 9). The HR for death was 0.92 (95% CI 0.74-1.14) and for the first hospitalization 0.93 (95% CI 0.82-1.06). Table 10 presents a descriptive analysis of the number of events (death/first hospitalization, death, and hospitalization) according to research center.
Table 8 Cox regression analysis of the composite primary endpoint (death/first unplanned hospitalization), and the secondary endpoints all cause death, and first unplanned hospitalization in ITT analysis
Outcome Treatment Hazard ratio (95% CI) p
Composite endpoint death/first unlanned hospitalization IG vs CG 0.92 (0.82-1.04) 0.19
All cause death IG vs CG 0.92 (0.74-1.14) 0.43
First unplanned hospitalization IG vs CG 0.93 (0.82-1.06) 0.28
Table 9 Cox regression analysis of the composite primary endpoint (death/first unplanned hospitalization) stratified by research center in ITT analysis
Research center Treatment Hazard ratio
(95% CI) P
SAGP IG vs CG 0.97 (0.76-1.24) 0.79
UNIMAN IG vs CG 0.94 (0.73-1.20) 0.61
UMR IG vs CG 0.99 (0.75-1.31) 0.93
PMU IG vs CG 0.68 (0.50-0.94) 0.02
UWH IG vs CG 1.05 (0.82-1.33) 0.72
Table 10 Number of events (death/first unplanned hospitalization, death, and hospitalization) according to research center
Death and/or hospitalisation Death Hospitalisation
No Yes No Yes No Yes
Research Center Treatment N % N % N % N % N % N %
SAGP CG 296 65.8 154 34.2 390 86.7 60 13.3 312 69.3 138 30.7
IG 300 66.5 151 33.5 411 91.1 40 8.9 310 68.7 141 31.3
Total 596 66.1 305 33.9 801 88.9 100 11.1 622 69.0 279 31.0
UNIMAN CG 207 62.5 124 37.5 304 91.8 27 8.2 217 65.6 114 34.4
IG 237 65.5 125 34.5 339 93.6 23 6.4 244 67.4 118 32.6
Total 444 64.1 249 35.9 643 92.8 50 7.2 461 66.5 232 33.5
UMR CG 255 50.4 251 49.6 465 91.9 41 8.1 265 52.4 241 47.6
IG 251 52.8 224 47.2 428 90.1 47 9.9 262 55.2 213 44.8
Total 506 51.6 475 48.4 893 91.0 88 9.0 527 53.7 454 46.3
PMU CG 103 34.9 192 65.1 248 84.1 47 15.9 119 40.3 176 59.7
IG 142 48.6 150 51.4 256 87.7 36 12.3 153 52.4 139 47.6
Total 245 41.7 342 58.3 504 85.9 83 14.1 272 46.3 315 53.7
UWH CG 146 39.6 223 60.4 317 85.9 52 14.1 159 43.1 210 56.9
IG 152 40.8 221 59.2 314 84.2 59 15.8 165 44.2 208 55.8
Total 298 40.2 444 59.8 631 85.0 111 15.0 324 43.7 418 56.3
Total CG 1007 51,6 944 48,4 1724 89,7 197 10,3 1072 54,9 879 45,1
IG 1082 55,4 871 44,6 1748 89,5 205 10,5 1134 58,1 819 41,9
Total 2089 53.5 1815 46.5 3472 88.9 432 11.1 2206 56.5 1698 43.5
Legend: CG: control group, IG: intervention group
Figure 7: Survival plot combined endpoint (ITT Analysis)
Table 11 describes the number of drugs at baseline, at the end of the study and the change between the two points according to study centre.
Table 11: Number of drugs per patient and study centre
Research center N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline SAGP (Italy) 901 10.00 1.95 10.0 6 25 9.0 11.0
UNIMAN (UK) 693 10.86 2.72 10.0 5 24 9.0 12.0
UMR (Germany1) 981 10.89 2.68 10.0 5 26 9.0 12.0
PMU (Austria) 587 10.43 2.31 10.0 7 21 9.0 12.0
UWH (Germany2) 742 10.35 2.33 10.0 6 23 9.0 12.0
Overall 3904 10.51 2.44 10.0 5 26 9.0 12.0
Number of drugs (total) End of study SAGP (Italy) 901 9.20 2.54 9.0 1 18 8.0 11.0
UNIMAN (UK) 693 9.95 3.19 10.0 0 26 8.0 12.0
UMR (Germany1) 981 10.56 3.26 10.0 0 26 8.0 12.0
PMU (Austria) 587 9.73 3.04 9.0 0 21 8.0 12.0
UMR (Germany2) 742 10.01 3.03 10.0 0 26 8.0 12.0
Overall 3904 9.91 3.05 10.0 0 26 8.0 12.0
Change number of drugs (total) End-Baseline SAGP (Italy) 901 -0.80 2.05 0.0 -13 6 -2.0 0.0
UNIMAN (UK) 693 -0.92 2.45 -1.0 -16 7 -2.0 0.0
UMR (Germany1) 981 -0.33 2.45 0.0 -14 11 -1.0 1.0
PMU (Austria) 587 -0.70 2.44 0.0 -15 7 -2.0 1.0
UWH (Germany2) 742 -0.34 2.30 0.0 -14 10 -1.0 1.0
Overall 3904 -0.60 2.35 0.0 -16 11 -2.0 1.0
Table 12 depicts the number of drugs at baseline, at the end of the study, and the change between the two points of measurement according to treatment group. Overall, the number of drugs per patient were reduced by a mean of -0.599 drugs per patient. The mean reduction was -0.841 in the intervention group, compared to -0.356 in the control group.
Table 12: Number of drugs per patient and treatment group
Treatment N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline Control group (CG) 1951 10.46 2.41 10.0 5 23 9.0 12.0
Intervention group (IG) 1953 10.55 2.48 10.0 5 26 9.0 12.0
Overall 3904 10.5 2.44 10.0 5 26 9.0 12.0
Number of drugs (total)
End of study Control group (CG) 1951 10.11 3.01 10.0 0 26 8.0 12.0
Intervention group (IG) 1953 9.71 3.09 9.0 0 26 8.0 11.0
Overall 3904 9.91 3.05 10.0 0 26 8.0 12.0
Change number of drugs (total)
End-Baseline Control group (CG) 1951 -0.36 2.29 0.0 -16 11 -1.0 1.0
Intervention group (IG) 1953 -0.84 2.38 0.0 -15 9 -2.0 0.0
Overall 3904 -0.60 2.35 0.0 -16 11 -2.0 1.0
Table 13 depicts the mean number of drugs at baseline, at the end of the study, and the change between baseline and the end of the study according to research centre and treatment group. The largest reduction in the number of drugs in the intervention group was achieved in Italy (SAGP), with a reduction of 1.26 drugs per patient. The smallest reduction was seen in Germany 2 (UWH) with a reduction of only 0.5 drugs per patient. Interestingly, the reduction of drugs in the UK was larger in the control group than in the intervention group.
Table 13: Number of drugs per patient, research center and treatment
Study centre Treatment group N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline SAGP (Italy) Control group (CG)
450 9.96 1.93 9.0 6 17 8.0 11.0
Intervention group (IG) 451 10.03 1.97 10.0 7 25 9.0 11.0
UNIMAN (UK) Control group (CG) 331 10.80 2.69 10.0 5 21 9.0 12.0
Intervention group (IG) 362 10.92 2.75 10.0 5 24 9.0 12.0
UMR (Germany1) Control group (CG) 506 10.83 2.50 10.0 5 23 9.0 12.0
Intervention group (IG) 475 10.95 2.87 10.0 5 26 9.0 12.0
PMU (Austria) Control group (CG) 295 10.33 2.39 10.0 8 21 8.0 12.0
Intervention group (IG) 292 10.54 2.22 10.0 7 17 9.0 12.0
UWH (Germany2) Control group (CG) 369 10.39 2.41 10.0 6 19 9.0 12.0
Intervention group (IG) 373 10.31 2.26 10.0 7 23 9.0 11.0
Number of drugs (total) End of study SAGP (Italy) Control group (CG) 450 9.63 2.45 9.0 3 18 8.0 11.0
Intervention group (IG) 451 8.77 2.56 9.0 1 18 7.0 10.0
UNIMAN (UK) Control group (CG) 331 9.80 3.11 10.0 0 20 8.0 12.0
Intervention group (IG) 362 10.08 3.26 10.0 3 26 8.0 12.0
UMR (Germany1) Control group (CG) 506 10.78 3.05 10.0 0 21 9.0 12.0
Intervention group (IG) 475 10.33 3.46 10.0 0 26 8.0 12.0
PMU (Austria) Control group (CG) 295 9.93 3.24 10.0 0 21 8.0 12.0
Intervention group (IG) 292 9.54 2.82 9.0 0 19 8.0 11.0
UWH (Germany2) Control group (CG) 369 10.21 3.12 10.0 1 26 8.0 12.0
Intervention group (IG) 373 9.81 2.93 10.0 0 22 8.0 11.0
Change number of drugs (total) End-Baseline SAGP (Italy) Control group (CG) 450 -0.34 1.89 0.0 -13 6 -1.0 0.0
Intervention group (IG) 451 -1.26 2.10 -1.0 -10 4 -3.0 0.0
UNIMAN (UK) Control group (CG) 331 -1.01 2.61 -1.0 -16 7 -2.0 0.0
Intervention group (IG) 362 -0.84 2.29 -1.0 -9 5 -2.0 0.0
UMR (Germany 1) Control group (CG) 506 -0.05 2.22 0.0 -12 11 -1.0 1.0
Intervention group (IG) 475 -0.62 2.65 0.0 -14 9 -2.0 1.0
PMU (Austria) Control group (CG) 295 -0.40 2.39 0.0 -13 7 -1.0 1.0
Intervention group (IG) 292 -1.00 2.46 -1.0 -15 5 -2.0 0.0
UWH (Germany 2) Control group (CG) 369 -0.18 2.34 0.0 -14 10 -1.0 1.0
Intervention group (IG) 373 -0.50 2.26 0.0 -11 6 -1.0 0.0
The treatment effect on the number of drugs at the end of the study was tested in a multi level analysis of covariance with “numbers of drugs at baseline” as a co-variate. The cluster factors “research centre”, the interaction between “research centre” and “treatment” and the random GP practice effect were all showing significant effects (see table 14). There also was a significant treatment effect between the control and the intervention group (p >0.0003).
Table 14: Ancova-table of treatment effect on the number of medications
Effect p
Number of drugs at baseline <.0001
Treatment 0.0003
Research center 0.0006
Research center *treatment 0.0398
In the per-protocol analysis for the combined endpoint (first hospitalization/death) the HR was 0.880 (95% CI 0.779-0.994) p=0.040 and thus statistical significance was reached. Figure 8 depicts the survival plot of the primary composite endpoint. Table 15 depicts the events of the endpoints death, hospitalization, and the combined endpoint according to research centre.
Figure 8: Survival Plot of the per-protocol analysis regarding the primary endpoint
Table 15: Per-protocol analysis of primary composite endpoint
Research center Treatment Death and/or hospital stay Death Hospital stay
No Yes No Yes No Yes
N % N % N % N % N % N %
SAGP (Italy) Control group (CG) 262 63.1 153 36.9 355 85.5 60 14.5 278 67.0 137 33.0
Intervention group (IG) 280 65.9 145 34.1 385 90.6 40 9.4 290 68.2 135 31.8
Total 542 64.5 298 35.5 740 88.1 100 11.9 568 67.6 272 32.4
UNIMAN (UK) Control group (CG) 184 60.5 120 39.5 277 91.1 27 8.9 194 63.8 110 36.2
Intervention group (IG) 194 63.2 113 36.8 286 93.2 21 6.8 201 65.5 106 34.5
Total 378 61.9 233 38.1 563 92.1 48 7.9 395 64.6 216 35.4
UMR (Germany) Control group (CG) 203 47.1 228 52.9 394 91.4 37 8.6 211 49.0 220 51.0
Intervention group (IG) 194 49.7 196 50.3 349 89.5 41 10.5 200 51.3 190 48.7
Total 397 48.4 424 51.6 743 90.5 78 9.5 411 50.1 410 49.9
PMU (Austria) Control group (CG) 92 32.6 190 67.4 235 83.3 47 16.7 108 38.3 174 61.7
Intervention group (IG) 135 48.9 141 51.1 240 87.0 36 13.0 146 52.9 130 47.1
Total 227 40.7 331 59.3 475 85.1 83 14.9 254 45.5 304 54.5
UWH (Germany) Control group (CG) 123 37.7 203 62.3 276 84.7 50 15.3 135 41.4 191 58.6
Intervention group (IG) 135 40.2 201 59.8 278 82.7 58 17.3 148 44.0 188 56.0
Total 258 39.0 404 61.0 554 83.7 108 16.3 283 42.7 379 57.3
Total Control group (CG) 864 49.1 894 50.9 1537 87.4 221 12.6 926 52.7 832 47.3
Intervention group (IG) 938 54.1 796 45.9 1538 88.7 196 11.3 985 56.8 749 43.2
Total 1802 51.6 1690 48.4 3075 88.1 417 11.9 1911 54.7 1581 45.3
Overall, the effectiveness of the PRIMA-eDS tool regarding the reduction of polypharmacy could be shown in our trial. The primary endpoint – the composite of death and first unplanned hospitalisation could be shown to be significant in per protocol analysis only. This result points to the fact that the tool has not been used as extensively as anticipated. The survey carried out to examine the usability of the tool also revealed that not all GPs in the intervention group made use of the tool as they were instructed to do, due to several barriers. One of the most important barriers turned out to be the lack of integration of the tool in the electronic health record used in the practices. This required extra time and effort of the GPs having to enter patients’ medication and other data already present in the EHR once again in the eCRF. Further analyses are ongoing to take a more detailed look at which drugs have been discontinued, and to check whether there are any correlations between usage of the tool according to the answers in the usability survey and the actual discontinuation of drugs by the respective GP in the trial.
Potential Impact:
1.4 Potential Impact, Main Dissemination Activities and Exploitation of Results
1.4.1 Impact of PRIMA-eDS on population health and health expenditure
Although the PRIMA-eDS intervention could not be shown to have an impact on the primary endpoint of the PRIMA-eDS-trial in the intention-to-treat population, quite an important impact on population health may be expected, as per-protocol-analysis revealed a significant reduction in hospitalization and death. In absolute numbers, the risk for the primary endpoint event was 45.9% in the intervention group and 50.9% in the control group, accounting for an absolute risk reduction of 5% in an observation period of two years. This means that one hospitalisation or death is avoided per year for one out of 40 patients cared for with the PRIMA-eDS-tool used decision support for avoiding inappropriate medication.
Roughly it can be estimated that there are more than 4.4 Million patients with polypharmacy in Germany (approximately 25% of all persons over 65 years). Thus, a nation wide implementation and usage of the PRIMA-eDS-tool would help to avoid 110,000 hospital stays or deaths per year. Extrapolating these numbers to the EU with about 500 Million people would lead to a tremendous impact on health and healthcare. European statistics estimate that 19.2% (96 Million) of the European population are older than 65 years (48). Even a conservative estimate of a polypharmacy rate of only 15% (14.4 Million) would result in avoiding 360,000 hospitalisations or deaths.
The second important result of the PRIMA-eDS trial revealed discontinuation of approximately 0.8 drugs per patient in the intervention group. When estimating average annual costs of only 100 €/drug, the discontinuation of 0.8 drugs/patient corresponds to health expenditure savings of 80 €/year/patient when using the PRIMA-eDS-tool. Using the population estimates from above this would correspond to annual savings of (14.4 Million x 80 €) 1.152 Million €.
1.4.2 Main dissemination activities
Ther PRIMA-eDS project has been made well known by the consortium already during the time of the project. Eleven peer-reviewed journal articles have been published (10–17,46,49,50). Two peer-reviewed articles are currently under review. Several further articles are being prepared, among others the main publication of the trial results. The project has been presented at numerous national and international conferences (see section 2).
1.4.3 Exploitation of results
The PRIMA-eDS-tool has been continuously updated and optimized during the trial by Duodecim and the PRIMA-eDS-consortium. It has been incorporated by Duodecim in its EbMeDS environment thus making the PRIMA-eDS tool available for further use in the Finnish health care system as “Comprehensive Medication Review”. Information about the availability of this “Comprehensive Medication” Reviews has been translated into English and German, and DMP and UWH are currently seeking possibilities to implement the tool in the electronic health records of the participating countries. Contacts with EHR-vendors have been made, and contacts with policy-makers are planned. The information on the “Comprehensive Medication Review” provided by DMP is attached as appendix 1.
1.5 Project Public Website
We established, and are maintaining a PRIMA-eDS-website (www.prima-eds.eu) giving relevant information about the project and the project achievements for the public. The website (www.prima-eds.eu) has been installed during the first months of the project and has been updated regularly since. The website has been used both for publishing of the PRIMA-eDS-project and for internal communication of the consortium. It has also been used as an entrance gateway for the GPs to access the eCRF forms, and it served as an online resource for the GPs regarding instruction on eCRF- and eDS-tool-usage. This service has now been discontinued as the main PRIMA-eDS-trial has been completed.
The website will be updated with the upcoming publications. The main results are withheld at the moment to avoid public availability of main results before the acceptance of the main manuscript by a peer-reviewed journal.
2 Use and Dissemination of Foreground
The following publication plan has been established:
Publication Status
Title: Polypharmacy in chronic diseases–Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): study protocol for a randomized controlled Trial
Authors: Andreas Sönnichsen, Ulrike S. Trampisch, Anja Rieckert, Giuliano Piccoliori, Anna Vögele, Maria Flamm, Tim Johansson, Aneez Esmail, David Reeves, Christin Löffler, Jennifer Höck, Renate Klaassen-Mielke, Hans Joachim Trampisch, Ilkka Kunnamo
Journal: Trials
Language: English
Published
Title/ concept: Polypharmacy in chronic diseases– Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): a cross-sectional analysis of factors predisposing for excessive polypharmacy
Authors: Anja Rieckert, Ulrike Trampisch, Renate Klaasen-Mielke, Eva Drewelow, Annez Esmail, Tim Johansson, Sophie Keller, Ilkka Kunnamo, Christin Löffler,
Joonas Mäkinen, Giuliano Piccoliori, Anna Vögele, Andreas Sönnichsen
Data: Baseline data
Journal: BMC Family Practice
Language: English
Submitted (Peer-review)
Title/ concept: Polypharmacy in chronic diseases– Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): a survey of GPs experiences
Authors: Anja Rieckert, Anne-Lisa Teichmann, Eva Drewelow, Celine Kriechmayr, Giuliano Piccoliori, Adrine Woodham, Andreas Sönnichsen
Data: Survey
Journal:
Language: English
Drafted, expected to be submitted end of March
Title/concept: Publication of our primary and secondary endpoints
Authors: Anja Rieckert, (to be determined), Ulrike Trampisch, Sabine Weissbach, Andreas Sönnichsen
Data: primary endpoint: composite endpoint of first non-elective hospital admission or death during the observation period measured as a binary outcome.
Secondary endpoint: all-cause mortality, nonselective hospital admission (number of episodes and duration), falls (number and severity), fractures, quality of life (SF-12v2), the number and types of drugs (total number, number discontinued, number not discontinued despite the recommendation to discontinue, number re-administered for symptom control), adverse event rate, and medication costs over the observation period
Journal: to be determined
Language: English
Currently drafted, analyses not finished yet
Title/concept: Predictors of GP ratings of patient frailty. This will be an analysis of the baseline dataset to identify patient variables that predict GP ratings of patient frailty. Variables include diagnoses, polypharmacy, BMI, falls, SF12, etc. The paper may include construction of a “frailty index” from the data in the eCRF and correlating this with the GP assessment.
Authors: David Reeves, (to be determined)
Data: Baseline Data
Journal:
Language: English
Not started
Title/concept: GP ratings of patient frailty as a predictor of patient outcomes. This will make use of the full dataset (i.e. all two years), to explore whether GP frailty ratings predict future patient outcomes, including death, hospitalisations, falls, QoL, etc, both with and without control for other baseline factors. It may also compare frailty ratings with other baseline predictors of outcomes for predictive ability, including polypharmacy and a frailty index constructed from the eCRF data. It will also explore whether frailty ratings moderate the association between polypharmacy and outcomes.
Authors: David Reeves, (to be determined)
Data: full dataset
Journal:
Language: English
Not started
Title/concept: Barriers for implementing electronic decision support tools, Qualitative synthesis
Authors: Malin Wörster, (to be determined), Andreas Sönnichsen
Data: pilot study & literature search
Journal:
Language: Conducting the search
Title/concept: Validation of the Pharao table. Correlation of symptoms with Pharao table.
Authors: (to be determined), Andreas Sönnichsen
Data: Baseline Data+ CMR (Pharao)
Journal: to be determined
Language: English
Developing an analysable database/development of concept
Title/concept: What is the potential of the tool? Analysis of the recommendations. Which drugs recommends the tool to discontinue?
Authors: Anja Rieckert (?) (to be determined), Andreas Sönnichsen
Giuliano is interested in contribution. Experience with PRIMA Italy (based on experts evaluations).
Data: CMR data
Journal: to be determined
Language: English
Developing an analysable database
Title/ concept: Analysis of what did the tool recommend to discontinue and what was actually discontinued.
Data: CMR+ 1st follow-up
Journal: to be determined
Language: English
Developing an analysable database
Title/concept: Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis
Authors: Tim Johansson, Muna E. Abuzahra, Sophie Keller, Eva Mann, Barbara Faller, Christina Sommerauer, Jennifer Höck, Christin Löffler, Anna Köchling, Jochen Schuler, Maria Flamm and Andreas Sönnichsen
Data: SR strategies to reduce polypharmacy
Journal: Br J Clin Pharmacol
Language: English
Published
Title/ concept: Interventions to reduce inappropriate polypharmacy: Implications for research and practice.
Authors: Tim Johansson, Maria Flamm, Andreas Sönnichsen
Data:
Journal: Elsevier Maturitas
Language: English
Published
Title/ concept: Letter to the editor. Reply to ‘Endpoints in strategies to reduce
polypharmacy’.
Authors: Tim Johansson, Maria Flamm, Andreas Sönnichsen, Jochen Schuler
Data:
Journal: British Journal of Clinical Pharmacology
Language: English
Published
Title/ concept: A set of systematic reviews to help reduce inappropriate prescribing to older people: study protocol
Authors: Yolanda V Martinez, Anna Renom-Guiteras, David Reeves, R Erandie Ediriweera de Silva, Aneez Esmail, Ilkka Kunnamo, Anja Rieckert, Christina Sommerauer, Andreas Sönnichsen
Data:
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Effectiveness and safety of Dipeptidyl peptidase 4 inhibitors in the management of type 2 diabetes in older adults: a systematic review and the development of recommendations to reduce inappropriate prescribing
Authors: Gisela Schott, Yolanda V Martinez, Erandie Ediriweera de Silva, Anna Renom-Guiteras, Anna Vögele, David Reeves, Ilkka Kunnamo, Minna Marttila-Vaara, Andreas Sönnichsen
Data: SR gliptins/diabetes
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review to help reduce inappropriate prescribing
Authors: Anna Vögele, Tim Johansson, Yolanda V Martinez, Lisa Schlender, Anna Renom-Guiteras, Anja Rieckert, Anne-Lisa Teichmann, Andreas Sönnichsen
Data: SR ß-Blocker/Hypertension
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Thiazides in the management of hypertension in older adults – a systematic review
Authors: Christina Sommerauer, Neha Kaushik, Adrine Woodham, Anna Renom-Guiteras, Yolanda V Martinez, David Reeves, Ilkka Kunnamo, Steffen Hübner,
Andreas Sönnichsen
Data: SR Thiazides/Hypertension
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Efficacy and Patient safety with Vitamin-K-antagonists and new anticoagulants in the Treatment/Prevention of Thromboembolism in Atrial Fibrillation in older adults – an overview of reviews
Authors: Christina Sommerauer, Lisa Schlender, Marc Krause, Sabine Weißbach, Anja Rieckert, Yolanda V Martinez, Anna Renom-Guiteras, Ilkka Kunnamo, Andreas Sönnichsen
Data: SR VKA/NOACs atrial fibrillation
Journal: BMC Geriatrics
Language: English Published
Title/concept: Efficacy and Patient safety with platelet aggregation inhibitors in the management of cerebral infarction, transient ischaemic attacks, peripheral artery occlusive disease and coronary disease in older adults – a systematic review
Authors: Maren Meinshausen, Anja Rieckert, Anna Renom-Guiteras, Moritz Kröger, Christina Sommerauer, Ilkka Kunnamo, Yolanda V Martinez, Aneez Esmail, Andreas Sönnichsen
Data: SR PAI
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Efficacy and safety of metformin in the management of type 2 Diabetes mellitus in older adults: a systematic review for the development of recommendations to reduce inappropriate prescriptions
Authors: Lisa Schlender, Yolanda V Martinez, Charles, David Reeves, Barbara Faller, Christina Sommerauer, Ilkka Kunnamo, Andreas Sönnichsen, Anna Renom-Guiteras
Data: SR Metformin
Journal: BMC Geriatrics
Language: English
published
Title/concept: Efficacy and patient safety with NSAIDs in the management of musculoskeletal disorders in older adults – a systematic review
Authors: Anja Rieckert, Barbara Faller, Christina Sommerauer, Lisa Schlender, Anne-Lisa Teichmann, Erandie Ediriweera de Silva, Sabine Weißbach, Anna Renom-Guiteras, Thekraiat Al’Quaran, Ilkka Kunnamo, Andreas Sönnichsen
Data: SR NSAIDs
Journal: to be determined
Language: English
1,4000 hits were screened
14 included; update completed
Title/concept: Insulin –SR
Authors: Raniah El-Jezawi, (more to be determined), Lisa Lechterbeck, Andreas Sönnichsen
Data: SR Insulin
Journal: to be determined
Language: English
Study selection 1,2, 3a completed; comprehensive search needs to be conducted
10,000 hits are currently screened
Title/concept: Insulin – Protocol
Authors: Lisa Schlender, Raniah El-Jezawi, Andreas Sönnichsen
Data: SR Insulin
Journal: to be determined
Language: English
Draft ready
Title/concept: Statins SR of SRs
Authors: Thekraiat Al’Quaran, Moritz Kröger (joint first authorship), Anja Rieckert, Anna Renom-Guiteras, David Reeves, Ilkka Kunnamo, Yolanda Martinez, Andreas Sönnichsen
Data: SR Statins
Journal: to be determined
Language: English
Update completed; currently Search 3a+3b
Title/concept: Using the electronic decision support tool PRIMA-eDS to optimize medication in primary care – a qualitative study for practical implementation
Authors: Anja Rieckert, Christina Sommerauer, Anja Krumeich, Andreas Sönnichsen
Data: Interviews with GPs from Witten
Journal: BMC Family Practice
Language: English
Re-submitted (mid-March)
Working title/concept: Validation of the tool. Comparison with recommendations from Dara Koper.
Authors: Ann-Kathrin Bücherl, (to be determined), Andreas Sönnichsen
Data: Kopers patients are entered into the PRIMA-eDS tool. Recommendations by the PRIMA-eDS tool are compared to recommendations provided by the study of Dara Koper
Journal: (to be determined)
Language: English
Database will be ready by mid-April
Title/concept: Recruitment of general practitioners for research - Experiences from a multicentre cluster-randomized controlled trial (PRIMA-eDS)
Authors: Sophie Keller, Tim Johansson, Eva Drewelow, Christin Löffler, Anja Rieckert, Ulrike Trampisch, Maria Flamm
Data: Questionnaires
Journal: BMC Medical Research Methodology
Language: English
Determination of new Journal after being rejected
Title/concept: On which quality are our clinical decisions based on? Quality of studies & systematic review – Interrater Reliability
Authors: Lisa Schlender, , Andreas Sönnichsen
Data: Search and data extraction documents
Journal: to be determined
Language: English
Analysis started
Title/concept: Validation of the PRIMA-eDS tool – comparison with data from the PRIMA study Italy
Authors: Giuliano Piccoliori, Anna Vögele
Data:
Journal:
Language:
CMR database needed
2.1 Section A: Dissemination Measures and Publications
Dissemination measures of the PRIMA-eDS-project are shown in tables 16 (A1) and 17 (A2). Table 16 (A1) depicts peer-reviewed journal articles already published. Table 17 (A2) shows other dissemination activities.
Table 16 (A1): List of scientific (peer reviewed) publications (starting with the most important ones)
No. Title Main author Title of journal Issue, date Publisher Place of Publication Year of publication Pages link or doi open access (y/n)
1 Polypharmacy in chronic diseases–Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): study protocol for a randomized controlled Trial
Sönnichsen BMC Trials 17:57 29 Jan 2016 Biomed Central UK 2016 1-9 https://doi.org/10.1186/s13063-016-1177-8
y
2 Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis. Johansson Br J Clin Pharmacol. Aug;82(2): Wiley UK 2016 532-48 10.1111/bcp.12959 n
3 A set of systematic reviews to help reduce inappropriate prescribing to older people: study protocol Martinez; Renom-Guiteras BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 1-9 10.1186/s12877-017-0570-9 y
4 Interventions to reduce inappropriate polypharmacy: Implications for research and practice. Johansson Maturitas. 97 Elsevier UK 2017 66-68. https://doi.org/10.1016/j.maturitas.2016.12.007 n
5 User perspectives on an electronic decision-support tool performing comprehensive medication reviews - a focus group study with physicians and nurses Koskela BMC Medical Informatics and Decision Making 16 Biomed Central UK 2015 1-6 http://www.biomedcentral.com/1472-6947/16/6
y
6 Thiazides in the management of hypertension in older adults – a systematic review
Sommerauer BMC Geriatrics 17, Oct. 2017 Biomed Central UK 2017 145-157 10.1186/s12877-017-0576-3 y
7 Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review to help reduce inappropriate prescribing Vögele BMC Geriatrics 17, Oct. 2017 Biomed Central UK 2017 119 - 143 10.1186/s12877-017-0575-4 y
8 Efficacy and safety of metformin in the management of type 2 diabetes mellitus in older adults: a systematic review for the development of recommendations to reduce potentially inappropriate prescribing Schlender; Martinez BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 99-117 10.1186/s12877-017-0574-5 y
9 Effectiveness and safety of vitamin K antagonists and new anticoagulants in the prevention of thromboembolism in atrial fibrillation in older adults – a systematic review of reviews and the development of recommendations to reduce inappropriate prescribing Sommerauer; Schlender BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 73-98 10.1186/s12877-017-0573-6 Y
10 Effectiveness and patient safety of platelet aggregation inhibitors in the prevention of cardiovascular disease and ischemic stroke in older adults – a systematic review Meinshausen; Rieckert BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 41-71 10.1186/s12877-017-0572-7 y
11 Effectiveness and safety of Dipeptidyl peptidase 4 inhibitors in the management of type 2 diabetes in older adults: a systematic review and the development of recommendations to reduce inappropriate prescribing
Schott, Martinez BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 11-39 10.1186/s12877-017-0571-8 Y
12 Reply to "Endpoints in strategies to reduce polypharmacy". Johansson Br J Clin Pharmacol. 83(2) Wiley UK 2017 434 DOI: 10.1111/bcp.13125 n
Table 17 (A2): List of Dissemination Activities (e.g. Conferences)
No Type of activity Main leader Title Date Place Type of audience Size of audience Countries addressed
1 Oral presentation
Johansson 17th annual meeting of the
German Network for Evidencebased Medicine 3/3/2016 Cologne (GER) Scientific community (higher education, research) 500 Austria,
Germany,
Italy
2 Oral presentation Johansson & Vögele 51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
3 Oral presentation Johansson 50th congress for general practice and family medicine of German Association of General Medicine 29/9/2016 Frankfurt am Main (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
4 Oral presentation Keller 49th congress for general practice and family medicine of German Association of General Medicine 18/9/2015 Bolzano (ITA) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
5 Oral presentation Johansson 48th congress for general practice and family medicine of German Association of General Medicine 19/9/2014 Hamburg (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
6 Oral presentation Johansson 15th annual meeting of the German Network for Evidence-based Medicine 13.3.2014 Halle, Saale (GER) Scientific community (higher education, research) 500 Austria, Germany
7 Poster Teichmann 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Germany, Austria, Italy, UK
8 Poster Schlender 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
9 Poster Faller 51th congress for general practice and family medicine of German Association of General Medicine 21.09.2017- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
10 Poster Vögele 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Italy, Austria, Germany
11 Oral presentation/ symposium Renom-Guiteras 21st International Association of Gerontology and Geriatrics (IAGG) World Congress 23.07.2017-27.07.2017 San Francisco, California USA Scientific community (higher education, research)
12 Oral presentation Rieckert 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany, Austria, Italy, UK
13 Oral presentation Rieckert 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
14 Oral presenation Sönnichsen 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017- 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
15 Poster Schlender 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
16 Oral presentation Sönnichsen Polypharmazie: ein Dilemma für Arzt und Patient. Ärztlicher Qualitätszirkel. Ärztlicher Qualitätszirkel 22/09/2016 Wetter a. d. Ruhr, Germany Scientific community, Physicians Germany
17 Oral presentation Sönnichsen 15th german congress for healthcare research 05/10/2016-07/10/2016
Berlin, Germany Scientific community (higher education, research) 600 Germany
18 Oral presentation Sönnichsen 15th german congress for healthcare research 05/10/2016 – 07/10/2016 Berlin, Germany Scientific community (higher education, research) 600 Germany
19 Poster Rieckert 50 th congress for general practice and family medicine of German Association of General Medicine 29/09/2016-01/10/2016 Frankfurt am Main Scientific community (higher education, research) 600 Germany, Austria, Italy, UK
20 Poster Sommerauer 50 th congress for general practice and family medicine of German Association of General Medicine 29/09/2016-01/10/2016 Frankfurt am Main Scientific community (higher education, research) 600 Germany
21 Poster Schlager WONCA SAR Conference 2016, Colombo, Sri Lanka 11/02/16-14/02/16 Colombo, Sri Lanka Scientific community (higher education, research) Germany, Austria, Italy
22 Poster Rieckert 17th annual meeting of the German Network for Evidence-based Medicine 03/03/2016-05/03/2016 Köln, Germany Scientific community (higher education, research) 500 Germany
23 Poster Rieckert 1.Wittener research meeting 06.04.2016 Witten, Germany Scientific community (higher education, research) Germany
24 Poster Celine Kriechmayr (PMU) Einsatz der elektronischen Entscheidungshilfe zur Optimierung der Medikation in der hausärztlichen Versorgung – eine quantitative Untersuchung zur praktischen Umsetzung im Rahmen der cluster-randomisert kontrollierten Studie PRIMA-eDS
51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
25 Poster PMU/UWH/UMR/SAGP Effektivität und Patientensicherheit von Betablockern bei älteren Menschen mit Hypertonie – Update einer systematischen Übersichtsarbeit
51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
26 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Mediation in der hausärztlichen Versorgung – eine quantitative Untersuchung zur praktischen Umsetzung 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community(higher education, research) 600 Germany, Austria, Italy, UK
27 Poster UWH Ist Metformin immer noch die beste Option bei der Behandlung von älteren Typ-2 Diabetikern. Eine systematische Übersichtsarbeit. 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
28 Poster UWH Ein systematisches Review zum Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von muskuloskelettalen Erkrankungen bei älteren Menschen 21.09.2017- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
29 Oral presentation/ symposium UWH The PRIMA-eDS electronic decision support tool for polypharmacy – a multinational European project 23.07.2017-27.07.2017 San Francisco, California USA Scientific community(higher education, research)
30 Oral presentation UWH Welche Faktoren prädisponieren ältere Menschen für extreme Polypharmazie? Eine Querschnittsanalyse mit Daten aus der PRIMA-eDS-Studie
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany, Austria, Italy, UK
31 Oral presentation UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung – eine qualitative Untersuchung zur zukünftigen Implementierung
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific communit(higher education, research)y 500 Germany
32 Poster UWH Die Vorteile und Risiken von Metformin bei der Behandlung von älteren Typ-2 Diabetikern. Eine systematische Übersichtsarbeit 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany
33 Oral presentation UWH Diskrepante Ergebnisse bei Randomisiert kontrollierten Studien, Systematic Reviews und Metanalysen zu identischen Fragestellungen – Wie kann das sein?
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany
34 Journal article PMU/UWH/ Interventions to reduce inappropriate polypharmacy: Implications for research and practice. Maturitas. 2017 Mar;97:66-68 3/2017 Scientific community (higher education, research) All
35 Journal article, PMU/UWH/ Reply to 'Endpoints in strategies to reduce polypharmacy'. Br J Clin Pharmacol. 2017 Feb;83(2):434 2/2017 Scientific community (higher education, research) All
36 Oral presentation UWH Reduktion unangemessener Polypharmazie durch eine elektronische Entscheidungshilfe - das PRIMA-eDS-Projekt
15. Deutscher Kongress für Versorgungsforschung 05/10/2016- 07/10/2016 Berlin, Germany Scientific community(higher education, research) 600 Germany
37 Oral presentation UWH Reduktion von Polypharmazie-Risiken für ältere Patienten - eine Implementierungs-Studie mit einer komplexen Intervention
15. Deutscher Kongress für Versorgungsforschung 05/10/2016 -07/10/2016 Berlin, Germany Scientific community(higher education, research) 600 Germany
38 Poster UWH Welche Faktoren prädisponieren ältere Menschen für extreme Polypharmazie? Eine Querschnittsanalyse mit Daten aus der PRIMA-eDS-Studie Poster UWH
39 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung Poster UWH
40 Poster PMU/UMR/UWH Die Effektivität von elektronischen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie hinsichtlich Morbidität, Lebensqualität und anderer Endpunkte – ´Update´ einer systematischen Übersichtsarbeit Poster PMU/UMR/UWH
41 Poster SAGP Effektivität und Patientensicherheit von Betablockern bei älteren Menschen mit Hypertonie - eine systematische Übersichtsarbeit [Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review]
50th congress for general practice and family medicine of German Association of Family Medicine 29/09/2016 -01/10/2016 Frankfurt, Germany Scientific community(higher education, research) 600 Germany, Austria
42 Poster PMU/UWH/UMR Die Effektivität von elektronischen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie hinsichtlich Morbidität, Lebensqualität und anderer Endpunkte - 'Update' einer systematischen Übersichtarbeit
50th congress for general practice and family medicine of German Association of General Medicine 29/9/2016 Frankfurt am Main (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
43 Journal article PMU/UWH/UMR Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis.Br J Clin Pharmacol. 2016 Aug;82(2):532-48 8/8/2016 Scientific community (higher education, research) All
44 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung
1.Wittener Forschertreffen 06.04.2016 Witten, Germany Scientific community(higher education, research) ? Germany
45 Oral presentation Johansson (PMU) Elektronische und nicht-elektronische Interventionen zur Reduktion von Polypharmazie – weiterhin fehlende Evidenz für einen Patientennutzen
17th annual meeting of the German Network for Evidencebased Medicine 03/03/2016 Köln (GER) Scientific community (higher education, research) 500 Austria,
Germany,
Italy
46 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung
Rieckert et al
17th annual meeting of the German Network for Evidence-based Medicine 03/03/2016-05/03/2016 Köln, Germany Scientific community(higher education, research) 500 Germany
47 Oral presentation PMU/UMR/UWH Elektronische und nicht-elektronische Interventionen zur Reduktion von Polypharmazie - weiterhin fehlende Evidenz für einen Patientennutzen
Johansson et al.
17th annual meeting of the German Network for Evidence-based Medicine
03/03/2016-05/03/2016 Köln, Germany Scientific community(higher education, research) 500 Austria, Germany
48 Poster UWH Polypharmacy in chronic diseases: Reduction of Inappropriate Medication and Adverse drug events in elderly populations by electronic Decision Support (PRIMA-eDS)
Schlager et al.
WONCA SAR Conference 2016, Colombo, Sri Lanka 11/02/16-14/02/16 Colombo, Sri Lanka Scientific community(higher education, research) ? All
49 Presentation DMP Clinical Informatics Course 23/2/2016 Toronto, Canada Physicians, medical students, clinical epidemiology students 35 Several, mostly Canadian
50 Presentation DMP WONCA International Classification Committee 14/9/2016 Helsinki, Finland Physicians 25 Several
51 Presentation DMP Guidelines International Conference 29/9/2016 Philadelphia, USA Physicians, guideline developers 20 Several
52 Presentation DMP ICT conference 7/6/2017 Khanty-Mansijsk, Russia Physicians, IT professionals 50 Several (mainly Russian)
53 Presentation DMP Global Evidence Summit, Multimorbidity Working Group Workshop 14/9/2017 Cape Town, South Africa Physicians 16 Several
54 Presentation DMP EBM Conference, Sechenov Medical University 23/10/2017 Moscow, Russia Physicians, medical students 220 Russian
55 Presentation DMP PRIMA-eDS Conference 10/11/2017 Witten, Germany Physicians 35 Several
56 Presentation DMP Nordic-German Health IT Symposium 30/11/2017 Hannover Physicians, IT professionals, investors 28 German
57 seminar UNIMAN UK branch of PRIMA-eDS study
David Reeves, Adrine Woodham, Yolanda Martinez
18/02/2014 Centre for Primary Care, University of Manchester Scientific community 60 UK
58 presentation DMP WONCA World Congress, Prague: Reduction of polypharmacy in the elderly
Ilkka Kunnamo 29.6.2013 Prague Congress Centre Scientific community 65 Several
59 presentation DMP
HIMSS Conference/Nordic Day, New Orleans: Medication and adverse drug events control with decision support systems
Ilkka Kunnamo
3.3.2013 New Orleans Congress Centre Scientific community; Industry 100 Denmark, Finland, Norway, Sweden, Iceland, United States
60 presentation DMP
Guidelines International Network Conference
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 19.8.2013 San Francisco Scientific community 50 Several
61 presentation DMP
Nordic Congress on General Practice
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 23.8.2013 Tampere Scientific community 1000 Denmark, Finland, Norway, Sweden, Iceland, several
62 presentation DMP
Cochrane Colloquium
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.9.2013 Quebec City Scientific community 25 Several
63 presentation DMP
WONCA International Classification Committee (webmeeting presentation)
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.9.2013 Johannesburg Scientific community 50 Several
64 presentation DMP
Kokkola Health Centre
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.12.2012 Kokkola, Finland Civil society 20 Finland
65 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.3.2013 Tampere, Finland Civil society 600 Finland
66 presentation DMP
Aalto University, Health Care Leadership programme
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 16.4.2013 Helsinki, Finland Civil society, policy makers 25 Finlaned
67 presentation DMP
Norbotten and Örebro health care authorities
Reduction of polypharmacy by electronic decision support
Peter Nyberg 18.4.2013 Örebro Policy makers 10 Sweden
68 presentation DMP
National institute for Health and Welfare
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 25.4.2013 Helsinki, Finlalnd Policy makers, civil society 60 Finland, Denmark, Canada
69 presentation DMP
National project on healthcare IT architecture
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 29.4.2013 Helsinki, Finland Policy makers 30 Finland
70 presentation DMP
WoHIT/eHealthWeek
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 14.5.2013 Dublin Scientific community, Policy makers 80 Several
71 presentation DMP
Tieto Healthcare Forum
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 29.8.2013 Helsinki, Finland Policy makers, civil society 200 Finland
72 presentation DMP
Health care management course for physicians in GP specialist training
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 26.9..2013 Helsinki, Finland Civil society 65 Finland
73 presentation DMP
National Guidelines Day
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 28.9.2013 Leuven, Belgium Policy makers, civil society, scientific soviety 160 Belgium
74 presentation DMP
Educational event for young doctors in Finland
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 25.10.2013 Helsinki, Finland Civil society 500 Finland
75 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 7.11.2013 Turku, Finland Civil society 55 Finland
76 presentation DMP
Physicianss’ Association in Tampere
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 7.11.2013 Tampere, Finland Civil society 45 Finland
77 presentation DMP
Healh care leadership educational event
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 8.11.2013 Helsinki, Finland Civil society 80 Finland
78 presentation DMP
EQuiP meeting
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 15.11.2013 Bologna, Italy Civil society, Scientific community 25 Several
79 presentation DMP
Nokia Health Centre
Reduction of polypharmacy by electronic decision support
Peter Nyberg 14.12.2013 Nokia, Finland Civil society 20 Finland
80 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 10.1.2014 Helsinki, Finland Civil society 25 Finland
81 presentation DMP
Central Finland Physicians’ Association
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 15.2.2014 Jyväskylä, Finland Civil society 80 Finland
82 presentation DMP
Ministry of Health, Finland
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo
Timo Haikonen
Peter Nyberg 6.3.2014 Helsinki, Finland Policy makers 5 Finland
83 presentation DMP
Health care leadership course, Finnish Medical Association
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 12.3.2014 Helsinki, Finland Civil society 30 Finland
84 presentation DMP
National institute for Health and Welfare
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 18.3.2014 Helsinki, Finland Civil society 40 Finland
85 presentation DMP
Finnish Innovation Fund (Sitra)
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 6.5.2014 Helsinki, Finland Industry, Civil society, Policy makers 170 Finland
86 journal article UMR Langer, C. (2014). Prima läuft der Arztbesuch. In: Studierendenschaft der Universität Rostock (Hrsg.). HEULER - Das Studentenmagazin der Uni Rostock auf Papier, No. 105; 04-2014, p.26. Apr 2014 Rostock civil society & scientific community 15.000 Germany
87 poster UMR Höck, J. et al. (2013). Polypharmazie bei Patienten mit chronischen Erkrankungen: Reduktion inadäquater Medikation und unerwünschter Arzneimittelwirkungen durch eine elektronische Entscheidungshilfe. Poster presented at F2-Forschung trifft Forschung-Forschungscamp, University of Rostock. 07. Nov 2013 Rostock scientific community 150 Germany
88 presentation PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
Meeting of the regional association of general practitioners 26/03/2014 Salzburg (A) scientific community 20 Austria
89 poster PMU/UWH,UMR A systematic review on strategies to reduce polypharmacy
15th annual meeting of the German Network for Evidence-based Medicine 13/03/2014 Halle (D) scientific community 400 Germany, Austria
90 poster PMU A systematic review on strategies to reduce polypharmacy
47th congress for general practice and familiy medicine of German Association of General Medicine 12/09/2013 Munich (D) scientific community 600 Germany, Austria
91 journal article PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
med.ium 11/2013 (a regional journal for physicians in Salzburg) 11/12/2013 Salzburg (A) scientific community , civil society 3600 Austria
92 presentation PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
Information event at Paracelsus Medical University for physicians of the province of Salzburg 02/10/2013 Salzburg (A) scientific community 30 Austria
93 poster UWH and UNIMAN Efficacy and safety of drugs commonly used in the management of chronic diseases in older adults: a compilation of systematic reviews for the development of an electronic decision support tool. Renom-Guiteras A, Martinez Y et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
94 poster UWH and SAGP Wirksamkeit und Sicherheit von oralen Antidiabetika im Management von Diabetes mellitus Typ 2 bei älteren Erwachsenen – eine systematische Übersichtsarbeit [Efficacy and safety of oral antidiabetics in the management of type 2 Diabetes mellitus in older people – a systematic review]. Vögele A et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
95 poster UWH Efficacy and Patient Safety of Statins in the Prevention of Cardiovascular Events in Older Adults – a Systematic Review. Al Qur’An T et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
96 poster UWH Wirksamkeit und Sicherheit von Antikoagulantien in der Prävention von Thromboembolien bei Vorhofflimmern bei älteren Erwachsenen - eine systematische Übersichtsarbeit. [Efficacy and safety of anticoagulants in the prevention of thromboembolism in atrial fibrillation in older adults – a systematic review]. Sommerauer C et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
97 presentation UWH Polypharmacy in chronic diseases: Reduction of Inappropriate Medication and Adverse drug events in elderly populations by electronic Decision Support. Sommerauer et al. 12th Annual Conference of the German network for health services research October 23rd-25th, 2013 Berlin, Germany Scientific community 700 Germany
98 presentation UWH Welche Medikamente sind evidenzbasiert bei multimorbiden älteren Patienten – eine systematische Literaturrecherche. [Which medications are evidence-based in older adults with comorbidity – compilation of systematic reviews]. Renom-Guiteras A et al. IZVF-Kolloquium at UWH May 7th, 2014 Witten, Germany Scientific community, policy makers 20 Germany
99 poster UWH Entwicklung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie: Das PRIMA-eDS Tool. [Development of an electronic decision tool for the reduction of polypharmacy: the PRIMA-eDS Tool]. Sommerauer et al.
13th Annual Conference of the German network for health services research June 24-27th, 2014 Düsseldorf, Germany Scientific community, policy makers 700 Germany
100 Workshop DMP Clinical decision support for professionals, at DECIDE Consortium meeting 3/6/2014 Edinburgh Scientific community 13 UK, Italy, Norway, Netherl.
101 Demo DMP NHS Education Scotland 5/6/2014 Glasgow Professionals 3 UK
102 Presentation DMP WONCA Europe, session on tools for chronic disease management 3/7/2014 Lisbon Primary care physicians More than 200 Several
103 Presentation DMP Clinical Decision support in guideline implementation, at Guidelines International Network conference 20/8/2014 Melbourne Scientific community, guideline developers 43 Several
104 Presentation DMP Big data and decision support in health care. Symposium by the University of Eastern Finland 9/10/2014 Kuopio, Finland Scientific community, IT develop. 160 Finland
105 Presentation DMP Comprehensive Medication Review, at Welfare ICT Forum by CGI Finland 10/10/2014 Tampere, Finland Clinicians and IT developers 70 Finland
106 Presentation DMP Clinical decision support, at EBM postgraduate course 25/2/2015 Tartu, Estonia Physicians 25 Estonia
107 Presentation DMP Experience of clinical decision support systems among Finnish doctors at Estonian Medical Convention 10/4/2015 Tallinn, Estonia Physicians 60 Estonia
108 Presentation DMP Clinical decision support at Medical Faculty, University of Helsinki 21/5/2015 Helsinki, Finland Medical students 100 Finland
109 Presentation DMP Clinical decision support at NIVA advanced course on occupational health 8/9/2015 Oslo, Norway Physicians 19 Finland, Norway, Denmark, Taiwan
110 Presentations DMP Multimorbidity working group meeting, Guidelines International Network Conference 8/10/2015 Amsterdam, Netherlands Physicians, guideline developers 20 Several
111 Face to face presentation PMU/UWH Die Effektivität von elektroni-schen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie – eine systema-tische Übersichtarbeit. 48th congress of the German Association of Family Medicine 19/09/2014 Hamburg (GER) Scientific community 600 Austria,
Germany,
Italy
112 Poster PMU Ärzte für Studien gewinnen –Erfahrungen aus der Versor-gungsforschung. 49th congress of the German Association of Family Medicine 18/9/2015 Bolzano (ITA) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
113 Journal article PMU Polypharmazie im Praxisalltag bewältigen: das PRIMA-eDS-Tool ÖGAM News (Österreichische Gesellschaft für Allgemein und Familienmedizin – Mitglied der Wonca) Ausgabe 06/15 06/2015 AUT Scientific community (higher education, research) 3600 + online AUT
114 Oral presentation/ abstract publication UNIMAN Safely reducing inappropriate prescribing to the older adults with polypharmacy: the multinational PRIMA-eDS study
(Awarded the prize of best oral paper at the conference) 20/11/2015 Annual Academic Sessions of the Sri Lanka Association of Geriatric Sessions (SLAGM), Waters Edge Hotel, Battaramulla, Sri Lanka Scientific and clinician communities 400 Sri Lanka
115 Poster presentation UNIMAN EvidenceBased Recommendations for the Reduction of Polypharmacy in Older People with Multiple Chronic Conditions: A Set of Systematic Reviews.
15-16/6/2015 Academy Health - Annual Research Meeting. June 14-16, Minneapolis, Minnesota, U.S. Scientific and clinician communities 1500 USA and International
116 Poster presentation UNIMAN EvidenceBased Recommendations for the Reduction of Polypharmacy in Older People with Multiple Chronic Conditions: A Set of Systematic Reviews.
9-10/7/2015 Annual Scientific Meeting of the Society for Academic Primary Care. University of Oxford, Oxford, U.K. Scientific community 200 UK
117 poster UWH Tackling polypharmacy in older adults: A compilation of systematic reviews for the development
of an electronic decision support tool
Schlager et al.
WONCA Europe Conference 2015 22/10/2015-25/10/2015 Istanbul, Turkey Scientific community 3680 79 countries, 42 outside of Europe
118 poster UWH Behandlung der Hypertonie bei älteren Menschen
Sommerauer et al.
49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
119 poster UWH Effektivität und Sicherheit in der Behandlung von Schmerz bei älteren Menschen mit Opioiden und Nichtsteroidalen Antirheumatika - von der Evidenz zur Entwicklung von Empfehlun-gen. Schlager et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
120 poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung. Rieckert et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
121 poster UWH Neue orale Antikoagulantien bei älteren Patienten. Sommerauer et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
122 presentation UWH Polypharmacy in Chronic Diseases: Reduction of Inappropriate Medication and Adverse Drug Events in Older Populations by Electronic Decision Support (PRIMA-eDS). Renom et al. The International Association of Gerontology and Geriatrics 23/04/2015 – 26/04/2015 Dublin, Ireland Scientific communitiy
123 presentation UWH Entwicklung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie: Das PRIMA-eDS Tool. Sönnichsen et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
124 poster UWH Effektivität und Patientensicherheit bei Protonenpumpeninhibitoren bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Sommerauer et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
125 poster UWH Effektivität und Sicherheit von Opioiden in der Behandlung von Schmerz und von Husten bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Schlager et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
126 poster UWH Efficacy and safety of drugs commonly used in the management of chronic diseases in older adults: a compilation of systematic reviews for the development of an electronic decision support tool
Al Qur´an et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
127 poster UWH Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von Erkrankungen des Stütz- und Bewegungsapparates bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Reickert et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
128 poster UWH Über die Einführung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie in der Hausarztpraxis - eine Pilotstudie. Wörster et al. 16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
129 poster UWH Effektivität und Sicherheit von Antihypertensiva vier verschiedener Substanzklassen (Diuretika, ACE-Hemmer, Kalziumantagonisten, ß-Blocker) bei älteren Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Sommerauer et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
130 poster UWH Effektivität und Sicherheit von Herzglykosiden in der Behandlung von Herzinsuffizienz und Vorhofflimmern bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Schlager et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
131 poster UWH Effektivität und Sicherheit von Thrombozytenaggregationshemmern (TAH) bei der Prävention und Behandlung von Herzkreislauferkrankungen älterer Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Meinshausen et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
132 poster UWH Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von Erkrankungen des Stütz- und Bewegungsapparates bei älteren Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Rieckert et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
List of Websites:
www.prima-eds.eu
Polypharmacy constitutes an increasing problem in older multimorbid patients. The prevalence of polypharmacy in the population >75 years has been found to be between 25 and 50%. While there exists only limited evidence regarding the benefit, the evidence regarding potential harm of polypharmacy is increasing. Thus polypharmacy and inappropriate prescribing have been shown to increase the risk of adverse drug events, and contribute substantially to morbidity, hospitalization, and mortality. An Austria study found that 10% of hospitalizations of patients >70 years are directly related to an adverse drug event. This clearly indicates that there is a strong need to reduce inappropriate drug prescription. Diverse heterogeneous approaches to reduce polypharmacy and inappropriate prescribing have been proposed, but none of these have been evaluated regarding practicability in primary care and clinically relevant endpoints within a sufficiently powered randomized controlled trial. The most popular approach to reduce inappropriate prescribing is at present the implementation of Potentially Inappropriate Medication (PIM)-lists as like the Beers-List. These lists are based on Delphi-rounds of experts (mostly pharmacologists and geriatricians) and not on evidence derived from clinical studies, and they never have been shown to have any effect on clinical outcome. Other approaches include the STOPP-criteria of Gallagher 2008, updated in 2015, or the Medication Appropriateness Index (MAI). None of these tools have been sufficiently evaluated regarding clinical outcome, and they are hardly applicable in every day practice. Using a combination of these methods, we found in a study of 169 primary care patients from 22 practices an average of 2.7 inappropriate medications per patient. But physicians in every day clinical practice, and especially general practitioners (GPs) are unable to do thorough medication reviews with their patients due to knowledge gaps and limited time. Therefore in the PRIMA-eDS project, we developed an electronic decision support (eDS) tool to assist physicians and patients to avoid inappropriate medication and polypharmacy, and to identify prescriptions where risks outweigh possible benefits of the drug treatment. This tool is based on current best evidence to optimize treatment of those diseases that are most common in elderly patients like cardiovascular disease, heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders, COPD, and mental diseases. We collated this evidence from Duodecim’s Evidence based Medicine Guidelines and several systematic reviews and transformed it into recommendations to optimize drug treatment of multimorbid older patients. These recommendations were integrated in the existing EbMeDS electronic decision support together with various commercially available databases to check medication for renal dosing, interactions, indications, contraindications and potential adverse drug events. The tool was then evaluated in a cluster randomized controlled trial with 3904 patients >75 years taking at least 8 different medications. The trial revealed a trend towards a reduction of the composite endpoint of mortality and first hospitalisation in intention to treat analysis, and a significant reduction of this endpoint in per protocol analysis. Also, a significant reduction in the number of drugs prescribed to the patients in the intervention group could be shown. Qualitative and quantitative usability analysis of the tool revealed that it must be integrated in existing electronic health records for wider use. Overall, the PRIM-eDS-tool could be shown to be a promising approach to deal with the increasing problem of polypharmacy in our society. It will help to reduce inappropriate prescribing and improve health and outcome in older patients with polypharmacy.
Project Context and Objectives:
1.2.1 Context of the PRIMA-eDS-project
The prevalence of chronic comorbidity and multimorbidity is substantial, and affects more than half of the population > 75 years of age. The patterns of chronic multimorbidity include mainly the following diseases: cardiovascular disease (including coronary heart disease, cerebrovascular disease and peripheral vascular disease), heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders including pain, COPD, and mental diseases like depression and dementia, with heart failure showing the highest rate of comorbidity, being accompanied by an average of 2.9 additional chronic diseases. As a result of multimorbidity, polypharmacy seems to be ever increasing as guideline adherence regarding drug treatment of chronic disease is recommended by policymakers and medical colleges. Depending on definitions and setting between 25 and 50% of all patients > 75 years are exposed to five or more drugs. The benefits of drugs are usually shown in studies including mostly younger subjects without comorbidity and multiple medications. The patients involved in clinical trials usually are characterized by a life expectancy of many years. Therefore the results of these studies may not be applied to older comorbid patients on polypharmacy and with reduced life expectancy. It has been shown that most clinical practice guidelines do not modify or discuss the applicability of their recommendations for older patients with multiple comorbid diseases, and following all the guidelines might lead to distinct polypharmacy with increased risk of adverse reactions and interactions between drugs, outweighing the potential benefits of drug treatment.
According to a systematic review including all settings, the overall adverse drug event (ADE) rate lies at a median of 14.9% per 1000 person-months and increases with age and polypharmacy. In a recent systematic review focusing particularly on the primary care setting, we found the prevalence rate for ADEs to be about 10% in prospective studies. Up to 50% of these ADEs are judged to be preventable. Various international studies found a hospitalisation rate due to adverse drug events between 2.4 and 16.6%, depending largely on the age-group evaluated. Polypharmacy and inappropriate medication have been shown to contribute substantially to the burden of morbidity, hospitalisation and death.
According to the 2017 drug prescribing report of the AOK (German statutory health insurance) 78.7% of all prescriptions are handed to patients by primary care physicians (primary care physicians in Germany are General Practitioners and General Internists). Any intervention aimed at the reduction of polypharmacy will therefore be most effective, if the professional group of primary care physicians is targeted. Although there exists only limited evidence regarding the benefit of guideline-recommended drugs in polymorbid older patients, GPs are reluctant to discontinue medication initiated by specialists and hospital physicians. Recently, distinct guidelines to deal with polypharmacy have been developed, but physicians are insecure in the reduction of polypharmacy, and there is significant risk of legal consequences when deviating from guidelines dealing with single diseases. Also, it may take too much time to convince patients that they should discontinue drugs they have been prescribed by a specialist. Adequate tools and an easily accessible knowledge base to aid GPs and patients in their shared decision to reduce polypharmacy do not exist.
1.2.2 Objectives of PRIMA-eDS
With the background of the increasing problem of polypharmacy regarding health outcomes and the inability of primary care physicians as the main prescribers of polypharmacy to deal with this problem, PRIMA-eDS set out to develop a usable solution for every day practice in the form of electronic decision support. After the development of the tool, this new approach to dealing with inappropriate prescribing and polypharmacy has been tested in a randomized controlled trial.
In detail, the objectives of PRIMA-eDS were as follows:
• to carry out a series of systematic reviews of the literature, evaluating and defining current best evidence regarding the pharmacological treatment of elderly patients with multiple chronic diseases, specifically addressing the problem of polypharmacy in the treatment of the most common chronic diseases in elderly populations,
• to develop an electronic decision support tool (PRIMA-eDS-tool) based on existing EbMeDS-technology developed by Duodecim ltd. to aid physicians and patients to make use of current best evidence when coming to a shared decision regarding multiple drug treatment of the most common chronic diseases in elderly populations,
• to test the efficacy of the PRIMA-eDS-tool in an appropriately powered, multinational, multicentre cluster-randomised controlled trial with a composite of hospital admission and mortality as the primary endpoint, the number of prescribed drugs as secondary endpoint, and mortality and adverse drug events as safety endpoints,
• to optimise the PRIMA-eDS-tool with the results of a qualitative and quantitative usability test and the randomised controlled trial, and make it available for widespread use within and outside of Europe.
1.2.2.1 Objective 1: Gathering current best evidence
First, we aimed to gather current best evidence to optimize treatment of those diseases that are most common in elderly patients like cardiovascular disease, heart failure, hypertension, atrial fibrillation, diabetes mellitus type 2, musculoskeletal disorders, COPD, and mental diseases. We carried out a series of systematic reviews of the literature focused on the treatment of multiple chronic diseases in the primary care setting. We made use of existing reviews (e.g. Cochrane-reviews) and guidelines and checked these reviews and guidelines for applicability and validity regarding their use in an elderly multimorbid population. A further systematic review focused on existing strategies and tools to reduce polypharmacy like the Garfinkel-algorithm, the Beers list, the STOPP-criteria, the medication appropriateness index (MAI) and others. Our systematic reviews were carried out with highest standards in methodology as described by the Cochrane Handbook 5.1.0. Objective 1 was the main focus of Work Package 2 of the PRIMA-eDS-project.
1.2.2.2 Objective 2: Development and implementation of the eDS-tool
Secondly, we used the evidence obtained in Objective 1 to develop a sustainable electronic decision support (eDS) tool to reduce polypharmacy and inappropriate medication in older patients that can be easily applied in daily primary care practice. The ttol was based on the existing EbMeDS-technology developed by Duodecim ltd. (www.ebmeds.org). EbMeDS uses predefined decision rules and risk estimation functions based on patient data from the electronic health record, and current best evidence to aid physicians and patients in their shared decision making about optimal therapy. In the PRIMA-eDS-project, diagnoses, medication, and patient baseline data like laboratory test results, renal function, weight, height etc. were transferred to a pseudonomized electronic case report form (eCRF) online which interacted online with the web-based eDS-tool thus providing alerts to the physician, if a specific medication should be avoided or discontinued according to current best evidence. The tool not only supports the GP in the reduction of polypharmacy and inappropriate medication, but also helps to convince the patient when medication should be discontinued for the better, and at the same time provides forensic assurance in the form of background evidence, if discontinuation is contradictory to existing mono-disease guidelines. Objective 2 was the main focus of Work Packages 3-4 of the PRIMA-eDS-project.
1.2.2.3 Objective 3: Evaluation of the eDS-tool
Thirdly, we set out to compare the clinical outcome of existing treatment regimens (applying polypharmacy in the treatment of multimorbidity, i.e. usual care) with the PRIMA-eDS-aided drug regimen reducing polypharmacy and inappropriate prescribing in a multinational, multicentre randomised controlled trial (RCT). From the studies of Garfinkel we have preliminary evidence that the prudent reduction of polypharmacy is beneficial to the patient, but the algorithm Garfinkel used in his study has not been proven to be effective in a randomized controlled trial. There is also limited evidence that Beers’ criteria characterizing inappropriate medication for the elderly are associated with adverse outcome, but this evidence is mostly based on retrospective cohort studies. Similarly, the STOPP-criteria have been shown to be associated with adverse drug events, but it has never been shown convincingly that their application reduces ADEs. In the PRIMA-eDS-trial we set out to prove that the prudential reduction of polypharmacy in the targeted diseases by avoiding inappropriate and non-evidence-based medication in older populations leads to a reduction of the hospital admission rate and adverse events and improves quality of life and mental as well as physical functioning. We intend to achieve the reduction of polypharmacy by eliminating non-evidence based drug therapy and prioritising medication in case of possibly dangerous drug-drug-interactions in a standardized fashion by using the PRIMA-eDS-tool developed for objective 2. Objective 3 was the main focus of Work Packages 5-9 of the PRIMA-eDS-project.
1.2.2.4 Objective 4: Optimization of the eDS-tool and dissemination
Fourthly, we evaluated the PRIMA-eDS-trial and continuously updated and optimized the PRIMA-eDS-tool with the findings of the RCT and continuous evidence-updates. The optimized tool is available as “Comprehensive Medication Review” by Duodecim ltd. and will be disseminated for widespread general utilization in Europe and outside Europe. Objective 4 was the main focus of Work Packages 10-12 of the PRIMA-eDS-project.
Project Results:
1.3.1 Gathering the evidence and development of the PRIMA-eDS-tool
1.3.1.1 Systematic reviews
We prepared a set of systematic reviews (SRs) on the drugs which are the most commonly prescribed to older people and which are associated with an increased risk of hospitalization due to adverse events in this population. The selection of the drugs was done using national prescription databases as well as published studies identified in the literature.
To shorten the review process while maintaining highest quality according to the Cochrane-handbook we developed a distinct methodology. A template for a systematic review study protocol to be used for all reviews carried out within PRIMA-eDS was developed. We then developed Standard Operating Procedures (SOP) for the reviewers to follow, with the aim to standardize our review work.
We aimed at including SRs, meta-analyses, clinical trials and observational studies which included a sufficient number of participants aged 65 years old and older (mean age ≥ 65 or > 80% studies with mean age ≥ 65 or subgroup analysis by age for SRs; >80% participants aged ≥ 65 or subgroup analysis by age for original studies) and which evaluated clinically relevant endpoints. A stepwise approach was carried out for the selection of studies including the following searches: search 1 (systematic reviews and meta-analyses from 2 databases: Cochrane Database of Systematic Reviews [CDSR] and Database of Abstracts of Reviews of Effects [DARE]), search 2 (systematic reviews and meta-analyses from 4 databases: MEDLINE, EMBASE, Health Technology Assessment Database, International Pharmaceutical Abstracts database), search 3A (individual studies meeting inclusion criteria from excluded studies in search 1 and/or 2), search 3B (intervention and observational studies from 4 databases: MEDLINE, EMBASE, Health Technology Assessment Database, International Pharmaceutical Abstracts database). Additional references which were excluded but considered of interest for the development of recommendations were also collected. Each consecutive step of the search strategy was only employed if the preceding step did not yield sufficient results to collate the evidence and derive recommendations.
Each SR has been undertaken by two independent reviewers who have followed the reviews’ protocol and the SOP using piloted excel documents to report the selection of studies, data extraction and quality appraisal. Validated instruments have been used to evaluate the quality of the included studies. Recommendations on when to discontinue or adjust the dose of one of the studied drugs have been developed by the researchers using GRADE methodology. Approximately 2 or 3 team meetings per SR were necessary for the development of recommendations, in which the researchers involved in each SR as well as clinicians and senior researchers took part. Recommendations were developed using a standardized wording schema providing information about the strength of the recommendation and the quality of the evidence. Once recommendations had been developed, they were peer reviewed by a team of experts in evidence based medicine provided by DMP.
Twenty-one SRs on 17 drug classes have been performed. In total, 210 articles have been included (59 SR or MA, 97 CIS, 54 OS), the number of included references per drug class ranging from 0 to 28. The inclusion of studies due to subgroup analysis by age was frequent; information on comorbidity and concomitant use of other drugs was often not available. Often reported outcomes were mortality, hospitalization, cardiovascular events including stroke, adverse drug event, and safety. The quality of the studies included ranged from very low to high.
Forty-two recommendations to discontinue or reconsider the use of the studied drugs or to adjust their doses have been. The quality of the evidence was often downgraded because it was based on subgroup analyses; the strength of the recommendations was most frequently weak. Fifty references were additionally taken into consideration.
Table 1 displays the overview of SRs as well as the information on the number of studies identified in each search, the number of studies included, excluded, considered of interest, and the number of recommendations developed for each SR.
Table 1: Systematic reviews on drug treatment for the development of PRIMA-eDS-stop-recommendations
Legend: The values included in the table are numbers of studies. CG=Clinical Guideline; CR=Cochrane Review; CT=Clinical trials; EC=expert consensus; EO=expert opinion; HTA=Health technology assessment; MA= Meta-analysis; NN=not necessary; NR=narrative review; NSAIDs=Nonsteroidal anti-inflammatory drugs; OS=observational study; PIM= potentially inappropriate medication; RCT=randomized controlled trial; SA=safety assessment; SR=Systematic review; VKA/NOACs=vitamin K antagonists/new oral anticoagulants
Table 2 presents a sample of 12 recommendations developed for 5 drug-indication pairs, in the way they have been incorporated into the tool.
Table 2: Recommendations for a sample of 5 drug-indication pairs derives from the PRIMA-eDS systematic reviews
Recommendation Strength of the recom-mendation/Quality of the evidence
Opioids/Pain
It is suggested not to start transdermal fentanyl delivery system (TTS fentanyl) at a dose of 25 mcg/h or higher for non-malignant chronic pain, especially in older people, because benefits (lower pain, improvement in quality of life) are similar compared to starting with lower doses (12.5 mcg/h) and adverse events are higher. Weak/Low
It is suggested to use the lowest possible dose of oral opioids for the management of chronic non-cancer pain because higher doses (e.g. >=40 mg/d of Morphine - the patient is using [here the eDS-tool automatically inserts the dose the patient is taking]/d) may increase the risk of fractures while there is less confidence of this risk for lower doses. However, adequate pain relief is a priority, and the minimum dose that effectively meets this goal should be used. Weak/Low
It is suggested to discontinue codeine or codeine combinations because codeine may increase the risk of fractures (probably due to falls), cardiovascular outcomes, hospitalizations and all-cause mortality compared to other opioids or non-use of codeine. Weak/Low
Non-steroidal anti-inflammatory drugs (NSAIDs)/Musculosceletal disorders
It is recommended to discontinue the use of NSAIDs because it might increase the risk of cardiovascular events. If necessary, naproxen and the additional use of proton-pump inhibitors (PPI) might be considered if the gastrointestinal risk is high (e.g. history of peptic ulcer, upper gastrointestinal bleeding). If PPIs are used in addition to NSAID, it is recommended to discontinue also PPI after 2 weeks when discontinuing NSAID. Strong/low
Thiazides/Hypertension
It is suggested to reduce high doses of thiazides for the management of hypertension because high doses (50 to 90 mg/day hydrochlorothiazide or equivalent) may be less effective in reducing mortality and coronary artery disease and may be associated with higher risk of gout compared to other antihypertensive medication including low thiazide doses. If the patient has also heart failure, please take the symptoms of heart failure additionally into account. Weak/Low
It is suggested to discontinue both drugs belonging to the combination of hydrochlorothiazide and triamterene in older adults over the age of 80 because the benefits of this treatment may not be established in this age group when compared with placebo and because it may be associated with the development of gouty arthritis, especially in men with high serum uric acid and creatinine levels. Weak/Low
Six of the systematic reviews and the methodology have been published in a peer reviewed journal.
An additional systematic review explored the effectiveness and impact of strategies to reduce polypharmacy (≥4 drugs) on health outcomes, processes of care, health resources, cost-effectiveness, as well as user and patient satisfaction/acceptance in older (≥65 years) patients, with the aim, to optimize the PRIMA-eDS-tool regarding efficacy. A PICOS-framework (population, intervention, control, outcome and study design) was used in order to specify the research questions and their elements used as study selection criteria. We used the GRADE Pro assessment tool to assess the quality of studies reporting on mortality and hospitalization (critical endpoints). In total 25 studies were included in this systematic review (Figure 1: PRISMA Flow Diagram). We identified three main categories of different types of interventions (pharmacist led interventions, physician led interventions and multidisciplinary team led interventions) and three main categories of settings where the interventions were implemented (primary care setting, nursing home setting, hospital setting). Participants’ (appropriate) medication use was reviewed using different methods (tools and instruments), whereof we identified four main methods: checklists (e.g. MAI, the Beers list), drug-drug interactions tools (e.g. software, lists), reconciliation methods and expert opinion. Performance of medication review and the decision criteria were not always described in detail in the included studies. The included studies provide limited evidence due to small sample sizes, poor quality, risk of bias, and short follow-up periods.
Figure 1: PRISMA-Flow-Diagram of the systematic review on strategies to reduce polypharmacy
Remarkable heterogeneity was found between studies. On the one hand, there was clinical heterogeneity in terms of populations, intensity and duration of interventions, outcomes, and follow-up times. On the other hand, there was also methodological heterogeneity such as differences in trial design (17 RCTs, 4 cluster-RCTs and 4 non-randomized controlled trials) and quality. Due to these inequalities a consistent interpretation of the results was seriously limited.
Overall the interventions had very limited effect regarding our outcomes of interest. Neither any single study nor our meta-analysis of effects on all-cause mortality during the study period showed any significant effect in favour of the intervention group (OR: 1.02 95 % CI 0.84 to 1.23) (see fig. 2).
Figure 2: Metaanalysis of the primary endpoint (mortality) of the systematic review on strategies to reduce polypharmacy
Only single studies found improvements in hospitalization in favour of the intervention group. The weighted mean including all prescribed drugs at baseline was in both groups 7.4 drugs per patient. At follow-up the weighted mean of drugs was reduced (-0.2) in the intervention group while it increased (+0.2) in controls but the difference did not reach significance. Further 12 outcome measures of interest were analysed in this SR but only single studies revealed significant differences between intervention and control (probably due to multiple testing and chance).
The evidence regarding effectiveness of interventions in the included studies is rather weak. It is unclear how to ideally organize and implement interventions (pharmacist led, physician led or led by a multidisciplinary team) in order to achieve clinically significant improvements in multimorbid older patients with polypharmacy. There is a great need of large long-term RCTs exploring the effect of strategies to reduce polypharmacy on clinically relevant patient outcome measures such as mortality and hospitalization. Thus the PRIMA-eDS trial is well justified. The harms of polypharmacy are well documented in the scientific literature, but it is not proven yet that interventions to reduce polypharmacy have a positive effect on patients’ outcome. Furthermore best practice models to reduce polypharmacy are still lacking. Due to the demographic development, the burden of chronic diseases and polypharmacy will increase. Polypharmacy is an emergent problem and a societal and economic challenge for most health care systems.
1.3.1.2 Building up the PRIMA-eDS-tool
In addition to the evidence derived from the systematic reviews carried out within the PRIMA-eDS-project, existing evidence was used to build the PRIMA-eDS-tool. DMP produced a list of recommendations which were relevant to older people under polypharmacy developed from the existing EbMeDS data base. Researchers with medical background from UWH gave feedback to this list of recommendations and an agreement was reached about important recommendations for older people. Thus, DMP identified a total of 1638 evidence summaries from EBM Guidelines (www.ebm-guidelines.com or www.ebm-guidelines.at) that were relevant for pharmacotherapy in older individuals. Selected evidence summaries were used as a basis of recommendations along with the results of the systematic reviews.
A total of 147 individual rules were included in the eDS tool, containing more than 200 different messages. 42 recommendations (messages from the rules) were developed on the basis of the systematic reviews. A general rule reminding that a drug is on the European PIM list was developed. Specific rules for several of the drugs on the PIM list were developed to guide the use of the drugs more accurately. The rest of the included rules were derived and (if necessary) modified from the existing rule set of EBMeDS. Descriptions of the rules and their evidence-base are available at http://www.ebmeds.org/web/guest/scripts?query=prima-eds+&lang=en.
We then incorporated the databases of indications, contraindications, dosing, drug interactions (SFINX®), renal dosing (Renbase®), and adverse effects (PHARAO®) in the PRIMA-eDS-tool. Each of these has a rule that enables the use of the database as part of the eDS tool. SFINX®, Renbase®, PHARAO® are developed and maintained by Medbase Ltd. (www.medbase.fi). Before use in the eDS, modifications were made to the databases by inactivating alerts on drug interactions of low significance, and renal dosing alerts in mild renal dysfunction for drugs. Specific rules were developed for several classes of drugs in Renbase® and in SFINX® to guide decision making more accurately than standard drug database alerts.
The total numbers of different feedback messages from the drug databases included in the eDS tool is 30171 (Table 3) which is about 30% more than anticipated in the project plan. All the messages are available in English, German and Italian.
Table 3: Messages in drug databases included in the eDS tool
Drug – drug Interactions 14565
Drug and Renal Malfunction 5024
Drug Contraindications 3144
Drug Indications 4141
Dosage Warnings 3297
Total 30171
A database of laboratory test results relevant for medications was developed on the basis of EBM Guidelines. The database contains reference values of tests related to medications, and recommended follow-up intervals of tests in safety monitoring of drug therapy.
Rule descriptions, rule logic, and executable rules were then developed using the EBMeDS content authoring tool. The testing environment of the EBMeDS content authoring tool was used to test each rule with varying test patient data by a minimum of three people independently of each other before publishing.
Feedback from pilot physicians participating in the study as well as from Finnish users of the eDS tool were analyzed and changes were made to some rules. Of the drug databases, the most extensive changes were made to the indications database where a number of coded indications were added, mostly because missing ICD-10 codes and variable use of the codes by the participating physicians.
The frailty scale on the eCRF was used as input variable to guide some decisions suggested by the rules.
1.3.1.3 Electronic case report form (eCRF) and interfaces
With the aim to make the tool accessible to physicians and to gather data for the PRIMA-eDS randomized trial, we developed an electronic case report form (eCRF) in cooperation with Avain Technologies ltd. The forms application of Avain Technologies Ltd. that is used as the platform for the eCRF is conformant with the Medical Device Directive (class 1 software product).
Search engines for drugs (ATC codes) and diagnoses (ICD-10 codes) were developed, and updated on the basis of user feedback during the first phase of (baseline) data entry of the study. The SF-12 form and the frailty scale were implemented as parts of the eCRF. The functions for outcome data entry require the data to be checked and reported at baseline and at every scheduled visit by asking the users to enter the data and tick a box that the data has been updated.
The functionality of the eCRF form was finalized for the study. Several changes to the original version were made on the basis of the pilot testing. Data export from the eCRF was performed and tested by the statistical team of the consortium.
A screenshot of the first page of the eCRF is shown in figure 3 (German version, the eCRF was alos available in English and Italian for the respective study centres. A print out of an eCRF containing patient data of a test-patient is depicted in figure 4.
Figure 3: Screenshot of the eCRF
Figure 4a: Printout of the eCRF, part 1
Figure 4b: Printout of the eCRF, part 2
The eCRF was then connected to the eDS-tool via a specific interface. The interface between the eCRF and eDS has been developed in cooperation with Avain Technologies Ltd. and is based on XML messaging. When data entry on the form has been completed, and the user clicks the "perform medication review" button, the form creates an XML message that is received by the eDS service and analysed using the rules and data tables created for the purpose. The response time of the eDS service is well under 2 seconds. eDS returns an XML message containing suggestions to modify the patient’s medications or to improve care by monitoring of potential adverse effects of the medication. The suggestions are then shown in the user interface of the eCMR. A screenshot of the eCMR is shown in fig. 5. The CMR could also be saved in pdf-format and printed if necessary or desired.
Figure 5a: User interface of the eDS tool – part 1.
Figure 5b: User interface of the eDS tool – part 2.
All transactions between the eCRF and the EBMeDS (eDS) service were stored in a log file for further analysis. For the control group, the recommendations of the eDS tool were recorded in the log file, but were not shown to the clinician.
Each time the form was updated by the clinician, it was stored on a dedicated server hosted in Finland. When the patient came for a follow-up visit, the form containing previously entered data was collected from the server. The user updated the data and entered the results of the patient assessment. The updated version of the eCRF was then stored on the server.
The data and backup copies of the data (with patient study identification codes created at baseline) were stored on the safe servers of Avain Technologies. Anonymous data for analyses were exported and delivered to be analysed by the statistical team of the consortium.
eCRF and data storage on a central server were protected by special security measures provided by Avain Technologies Ltd. Only authorized users (the clinicians participating in the study and researchers monitoring the study) were able to access the data on the forms. User rights administration by Avain assured that any GP was only able to access data of his own patients. Because confidentiality and safe transfer of the patient data on the form was an absolute requirement, an independent external information security audit has been performed, and an archives certificate has been awarded for Avain Technologies (Certificate number 6692-04, by Inspecta).
1.3.1.4 Pilot study and usability studies
Before starting the PRIMA-eDS-trial, a pilot study was carried out to test and optimize the functionality of the eCRF and the eDS-tool. The main subject identified was the expenditure of time needed for entering patient data into the eCRF. The eCRF form was also perceived as too complex and not simple and intuitive enough to use, especially for those users who do not have much technical experience. Study assistants reported a list of minor bugs and errors back to the DMP team so that they could be corrected as soon as possible. Despite the difficulties encountered when using the study tools, most GPs that have taken part in the pilot study expressed their desire to use the CMR in their practice. Several changes to facilitate data entry were suggested and implemented. For instance, the eCRF was adapted to account for faster data entry and better operability. The CMR output page was adapted and optimized several times to offer compact and fast information to the physician. Usability of the tool have been evaluated in qualitative and quantitative studies, and the results have been incorporated in the tool and the eCRF.
1.3.2 Evaluation of the PRIMA-eDS-tool in a randomized controlled trial
1.3.2.1 Study protocol, ethics approval and trial registration
In parallel to the development of the tool, the PRIMA-eDS randomized controlled trial was planned and initiated. The study protocol has been finalized including final sample size calculation. In 2016 the study protocol was published in a peer reviewed journal. Ethics approval for the coordinating study centre (UWH) was granted on December 3rd, 2013. The study has been approved unconditionally by the ethics committee of Salzburg on the 8th of April 2014 for PMU. The study was approved by the Ethics Committee of the Rostock University Medical Centre on February 14th, 2014 (Approval-No. A 2014-0020). Ethics approval for SAGP was granted on the 19th of June 2013 by the Ethics Committee of Belluno. Ethics approval for UNIMAN was obtained on the 6th of June 2014 at the Ethics Committee of NRES Committee North West - Greater Manchester East. Due to extension of the study to other research networks because of recruiting problems we had to obtain additional ethics approvals of local authorities in the UK and of the Bavarian Chamber of Physicians (Bayerische Ärztekammer) which were granted before the start of the trial. Trial registration was performed July 31, 2014 with Current Controlled Trials ltd (http://www.controlled-trials.com/ISRCTN10137559).
1.3.2.2 Physician recruitment
Due to regional differences regarding the health system and organisational differences in primary healthcare, each study centre had to develop its own standard operating procedures regarding recruitment. To account for these differences in the trial and to avoid bias due to the existing differences between countries, randomisation was stratified by country. The procedures of GP recruitment were as follows:
UWH
Recruitment of GP practices took place in three waves. We obtained lists of all practicing GPs under statutory health insurance of the Westfalen-Lippe region and adjacent regions from the “Kassenärztliche Vereinigung Westfalen-Lippe” and the “Kassenärztliche Vereinigung Nordrhein”. We informed the GPs in writing about the PRIMA-eDS trial and asked for a response by fax expressing interest to participate. In total we contacted 1192 physicians. The overall first response rate expressing interest was approximately 10 %, finally 76 GPs recruited patients (response rate 6.4%).
UMR
At the Rostock study centre, recruitment of GPs via mail started in February 2014. In May 2014 a total of 74 GPs decided to enroll for the study. Unfortunately, between June 2014 and April 2015 18 of these GPs changed their mind and withdrew from the study. Reasons included e.g. a loss of interest, or too long waiting time until the start of the trial. Rostock study centre therefore decided to further recruit in the area of Mecklenburg-Western Pomerania as well as in Berlin. Over the entire recruitment period a total of 1.730 GPs from the regions Mecklenburg Western-Pomerania (1.120 GPs), Schleswig-Holstein (50 GPs) and Berlin (560 GPs) have been contacted over seven waves. Positive feedback and interest in the study was declared by 175 GPs. After telephone contact a total of 87 GPs (5.0%) finally took part in the trial.
SAGP
SAGP recruited GPs in the neighboring region of Veneto, Italy. The local collaboration partner of SAGP was SVeMG (Scuola Veneta di Medicina Generale = School of General Practice in Veneto), which has extensive history and experience with the performance of scientific projects in the primary care setting. SVeMG supported SAGP in all aspects of recruitment of GPs and later in the recruitment of patients for PRIMAeDS. Italian GPs for PRIMAeDS were recruited within the Primary Care Health Research Network of Veneto “ MilleInRete”. All GPs of the network were invited by email to participate in the project. The invitation contained information about the project, requirements for participation and an attached participation form, which GPs were asked to sign in case of interest. Seventy GPs replied and declared to be interested in participating in PRIMAeDS by sending us the signed participating form. Finally, 72 Italian GPs signed the informed consent and received eCRF training.
PMU
The recruitment strategy of PMU was planned stepwise and manifold due to the fact that we knew from previous experiences that recruitment of study practices in the area is a demanding challenge. We informed GPs via letter, via mail and via fax in different regions (Austria, Bavaria) in several waves through different providers such as the PMU, general practitioners federation in Bavaria, laboratory community, general practitioners society in Salzburg (SAGAM). Written information was followed by phone calls. Common reasons for not participating were lack of time, not treating enough polypharmacy patients, and software (medical health record) in the surgery was not suitable for patient identification (e.g. lack of filter function). The PRIMA-eDS study including participation information was published in a regional journal of the Physician’s chamber of Salzburg (Medium 11/2013, circulation of 3.600); and in a national journal, ÖGAM News (06/2015). Several face-to-face presentations at different occasions were held: e.g. project information event at Paracelsus Medical University, Salzburg (10/2013), external GP meetings “Bezirksärzte-Fortbildung Salzburg” (03/2014). An additional effort was an advertisement of the study via the Chamber of Physicians (Ärztekammer) Salzburg homepage (07/2014). Several personal, phone and mail contacts were performed to recruit GPs via “snow-ball strategy”. In summary, all efforts during the whole recruitment period resulted in about 80 GPs interested in the study. Eventually sixty-seven GPs signed the contract to participate; however eight withdrew their consent before randomization, and thus only 59 GPs finally participated. PMU recruitment strategy was presented at the annual meeting of the German College of General Practitioners and Family Medicine 2015.
UNIMAN
UNIMAN collaborated with five National Institute for Health Research (NIHR) Clinical Research Networks (CRN). The five CRNs were Eastern, North Eastern, Wales, North West Coast and Greater Manchester. The CRNs are widely spread across four different UK regions, the North West, North East, Wales, and the South East. From these, 76 practices provided expressions of interest from which 67 signed up for the trial and recruited patients.
1.3.2.3 Physician instruction
Just like physician recruitment, GP instruction for the trial had to beorganized by each study centre separately due to differences in local conditions.
UWH
Instruction of GPs was carried out using several approaches to reach a maximum effectiveness. It differed between intervention and control group. All GPs were offered general information about the RCT in writing and by phone, Instruction on using the eCRF via personal visit of the practice by a study nurse, and by providing a manual on eCRF-usage as well as a video on eCRF-usage. Both the manual and the video were made available on the PRIMA-eDS-website in English, German and Italian for all GPs of the PRIMA-eDS-trial in all study centres. We also provided continuing instruction via a telephone hotline which was implemented specifically for the study.
In addition, all GPs of the intervention group received instruction on the use of the PRIMAS-eDS-tool and on the shared decision making process of reducing polypharmacy in collaboration with the patients. We provided a manual and a video on PRIMA-eDS-tool usage and shared decision making. The manual and the video were made available in English, German and Italian for all GPs of the intervention group of the PRIMA-eDS-trial in all study centres. Furthermore, all GPs of the intervention group were offered participation in a webinar explaining and detailing eDS-tool usage and shared decision making (duration approximately 45 min. The webinar was offered to all German speaking GPs of the intervention group, also in the other study centres). Personal visits by a study nurse to explain the eDS-tool were offered but not widely made use off. A telephone hotline during regular office hours was implemented to provide continuous support on eDS-tool usage and shared decision making.
UMR
Instructions on proper use of the eCRF were provided by the study personnel during the initial visits of the GPs. After randomization, GPs of the intervention group received a tutorial for proper use of the eDS-Tool (DVD) and written information. In addition, via telephone a physician was ready to answer all questions of the GPs. The manuals, webinars and videos described under “UWH” were also offered to the GPs.
SAGP
eCRF training: The research personnel of SAGP trained all Italian PRIMAeDS GPs in using the eCRF. eDS training: eDS training meetings were held for GPs of the intervention group. Individual eDS training was provided for GPs not being able to participate in the training meetings. All PRIMA-eDS meetings (regarding information, recruitment, eCRF-and eDS- training and feedback) with the participating GPs were held in collaboration with SVeMG at its headquarters in Caselle di Selvazzano, PD. The manuals and videos described under “UWH” were also offered to the GPs.
PMU
Training in the usage of eCRF and eDS was adapted to the regional setting. Because of regional characteristics (e.g. GP practices are foremost organized as single handed offices), workload and limited time resources, participating GPs refused to attend group training sessions in a centralized location. Project staff delivered the training face-to face within each practice in a two to four-hour session covering identification of eligible participants and entering data in the eCRF. As the eCRF was not ready for use when we started the GP recruitment the first GP visits focused on identification of eligible participants only. Patient identification was started as soon as possible as the procedure had to be performed via manual search by the project staff due to limited software options in regional GP offices. After randomization GPs in the intervention group were trained on how to use eDS and eCRF within a second visit by project staff in the respective offices. GPs in the control groups were instructed via telephone, and the eCRF manual mentioned above was sent to the GP. To support the participating GPs, we installed a telephone hotline for PRIMA-eDS GPs.
UNIMAN
Training in the eCRF and eDS was adapted to the UK context. Because UK GPs are only undertaking Medication Reviews, not the non-Medication Review tasks required for the trial (approaching and consenting patients, completing patient questionnaires and completion of the eCRF), the GP recruitment process included negotiating whether CRN, practice or research team staff would complete the non-Medication-Review tasks, as agreed in the original CRN negotiations. Therefore some parts of the training given to GPs in other countries, in the UK was delivered to other staff groups (mainly nurses). In the Eastern and North Eastern CRN areas peripatetic Research Nurses employed by the CRNs cascaded training to GPs and carried out the non-Medication Review tasks themselves. The study team provided training days in Cambridge and Newcastle for the teams of Research Nurses there, with hands-on training in both non-Medication-Review tasks and Medication Reviews for Intervention practices and only non-Medication-Review tasks for Control practices. In Wales, North West Coast and Greater Manchester all training was delivered by the study team to practices directly. Face-to-face non-Medication-Review training was completed with the CRN/practice personnel identified when the practices were recruited, and timed to facilitate patient recruitment. Medication Review training was undertaken post-Randomisation, initially face-to-face, then by teleconference for speed only for the intervention practices.
1.3.2.4 Patient recruitment
Like GP- recruitment, patient recruitment hat to be organized locally due to different local conditions.
UWH
GPs were instructed to perform searches in their electronic health records to identify eligible patients (patients ≥75 years of age taking ≥8 drugs). Eligible patients were informed about the study by the GP in writing or personally when entering the practice at the discretion of the GP. If more than 15 eligible patients were willing to participate in the study, GPs were asked to randomly select patients or recruit a random sample by continuous recruitment of patients coming to the practice. GPs were asked to recruit at least eight and no more than 15 patients to assure uniformity of cluster size, but GPs with less than 8 or more than 15 patients were not excluded. After signing informed consent, baseline data of the patients were entered into the eCR. SF-12 was performed on paper, and the filled out forms were sent to the UWH study centre for data entry. After completing recruitment, GPs were asked to inform the study centre. To assure concealment of allocation during recruitment, GPs were randomized after the completion of recruitment. After randomization GPs could not include additional patients in the study. Recruitment ended September 30, 2015, and all GPs not randomized yet were randomized on Oct 5, 2015, no matter how many patients they had recruited. In total, 742 patients were recruited by 76 GPs for the UWH study centre.
UMR
Based on standard operating procedures eligible patients were selected from the electronic health record of the GP practices. The list was presented to the GP and he/she was asked to exclude patients not suitable for the study (exclusion criteria). A random sample of a maximum of 30 patients was then selected from the remaining list. These patients received an invitation letter to participate in the study. Patients interested in participating contacted their GP and made an appointment to receive further information, to sign informed consent and to record baseline data. In total, at Rostock study centre 987 patients have been included by 87 GPs.
SAGP
Patients were selected from the electronic health records of the GP surgeries through two special software querys, according to the inclusion criteria of PRIMA-eDS. This was feasible, because all members of the research network, in which we recruited GPs, use the same practice software MilleWin, which is able to extract the patients according to the PRIMAeDS inclusion criteria. For the selected patients, GPs had to apply exclusion criteria one by one. The remaining patients were put in random order by blinded research-personnel of SAGP. The first 30 patients of this random list of eligible patients were invited by letter to take part in the study. All GPs were asked to recruit 11 patients. In total, 70 GPs recruited 905 patients for the PRIMA-eDS-trial with a maximum of 15 patients and a minimum of 8 patients per GP.
PMU
All eligible patients were selected by an electronic search of the electronic health records in the respective GP practice. The list of all eligible patients was numbered serially and brought in random order. Following the random list, the first 30 patients fulfilling inclusion criteria were identified by manual search. Invitation letters for the selected patients were prepared by project staff during the first visit at the GP surgery. In the invitation letter patients were informed about the project and asked to make an appointment at their GP-practice for further information and to sign informed consent. The GP decided whether to use the invitation letters or personally contact patients or both. Baseline data was obtained for the eCRF after informed consent was signed by the patient. As soon as the last patient consented to participate, the respective GP sent a confirmation fax. The project team checked whether all patients met inclusion criteria and submitted a randomization request for the respective practice to the randomization centre. In total, 587 patients were recruited by 59 GPs.
UNIMAN
A total of 702 patients have been recruited from across 67 practices. Four of these patients were subsequently found to be ineligible for the trial, leaving a total eligible sample of 698; very close to the target of 735 patients from 67 practices. Patients were recruited via practices, who posted an invitation pack (letter on practice stationery, study information, reply slip and prepaid envelope) to selected patients identified from the practice list. Interested patients met with a CRN/practice nurse/university researcher who explained the study further, gave the opportunity to ask questions and to consider participation further. Patients who were still interested in the study signed the informed consent form and with this permission could then complete the patient questionnaires and have their data entered into the eCRF.
1.3.2.5 Randomization
A statistician created lists for blockwise randomization stratified by study centre. GP practices were then allocated to either the intervention group or the control group when they had completed patient recruitment. Cluster randomization was performed by the Department of Medical Informatics, Biometry and Epidemiology of the Ruhr-University Bochum (AMIB) starting January 10, 2015 and ending October 5, 2015. Weekly randomization reports were provided by AMIB.
1.3.2.6 Blinding
To minimize bias and confounding, all patients were blinded, i.e. patients were not informed explicitly whether their doctor was part of the intervention group or the control group. Inherent to the study protocol general practitioners could not be blinded. Also, it was not possible to blind all research-personnel. As research-personnel did not interact directly with patients or GPs, the risk of bias due to the lack of blinding appears to be negligible. The statistician performing final analysis of the trial was blinded regarding GPs belonging to the intervention or the control group.
1.3.2.7 Carrying out the trial
Intervention
After completion of recruitment and randomization, the GPs of the intervention group were instructed to perform the medication review with their patients using the PRIMA-eDS-tool. GP-instruction was carried out via face-to-face-instruction by study personnel as well as video-instruction provided via the PRIMA-eDS website (www.prima-eds.eu) and webinars held by study personnel. A hotline was available throughout the study, if GPs encountered technical or medical problems using the tool. In the instruction, special emphasis was put on a shared decision making process to come to an agreement with the patient regarding discontinuation of inappropriate medication. The GPs of the control group received instructions and continuous support on the use of the eCRF, and were advised to prescribe medication to their patients as usual according to current guidelines.
Recording outcome measures and assessment of data
GPs performed the baseline visit and documented patient data in the electronic case report form (eCRF) after recruitment and informed consent, but before randomization to assure concealment of allocation. All GPs were instructed to complete any baseline data still missing after randomization. The medication review was then carried out using the PRIMA-eDS-tool in the intervention group while the control group checked patient medication as usual. GPs were reimbursed for recruitment and baseline data acquisition after completing all their baseline examinations. For the follow up examinations scheduled after 8, 16 and 24 months, GPs were reminded to prepare/perform these follow-ups and document all patient data as required by the eCRF. Letters and faxes were sent to the GPs and followed up by telephone calls if necessary. We allowed a time frame from 4 to 12 months after randomization for follow up 1 (FU1), from 12 to 20 months for follow up 2 (FU2), and from 20 to 28 months for follow up 3 (FU3).
The project team continuously supported GPs to assure the conduction of the intervention (e.g. GP instructions on how to use eCRF/eDS correctly, data collection and data entry of SF-12, baseline and first follow-up examination) in the intervention group and to assure correct completion of the eCRF in the control group.
Data management and statistical analysis
All data were stared securely by Avain Technologies ltd. Anonymized data files were transferred to AMIB on a weekly basis for data cleaning and quality check. Any deviations and errors in data recordings were reported to the monitors who in turn contacted the respective GP for data correction or clarification.
A statistical analysis plan was developed by AMIB in cooperation with UWH and UNIMAN.
Monitoring, quality assurance, and Safety and Data Monitoring Committee (SDMC)
To ensure safety of all patients throughout the study, an independent data and safety monitoring board (SDMC) was set up. The SDMC met periodically throughout the study. The members reviewed and discussed the current data (at baseline and at the various follow-up visits. The SDMC was informed about dropouts, deaths, hospitalisations, and any other relevant safety issue if applicable. The main focus of the interim analyses presented to the SDMC was, to detect discrepancies in relevant safety parameters and their effects on the (non-) achievement of the outcome parameters. All interim analyses were kept blind regarding events in the intervention or the control group. The SDMC received a blinded descriptive synopsis of the most important variables (especially primary outcomes such as deaths, hospitalizations, reported symptoms related to changes in medication, dropouts, monitoring results etc.) in preparation for their regular conferences. The SDMS checked the monitoring of the study and approved of the continuation of the study. The SDMC at no point had any concerns about patient safety throughout the trial, and a discontinuation of the trial was never recommended.
The designated site monitors at each study site performed the local monitoring activities and reported their results to the coordinating monitor at UMR. Throughout the study, site monitors established and maintained contact to general practitioners and practice teams of the respective study site, clarified issues and difficulties during the monitoring contact and prepared status reports on completed on- and offsite visits. All monitoring activities were coordinated and reviewed by UMR including the distribution of relevant information and the review of all monitoring reports provided by the site monitors.
Overall, vivid and frequent communication as well as sharing of experiences between site monitors and coordinating monitors were carried out, mainly via individual phone calls, monthly phone conferences, e-mail, and at steering-committee- and project meetings. In order to facilitate a transparent and quick communication, the online-forum freedcamp.com was used by all monitoring personnel, not only to forward and share documents and information, but also to explicitly exchange experiences between site monitors of all study sites. Additionally, the coordinating monitors regularly reported on the current monitoring trends and results to the consortium via oral presentations held at the project meetings or telephone conferences.
To assure correct and standardized monitoring as well as data quality, a monitoring manual was developed and updated during the study. The monitoring manual and its updates were delivered to every cooperating trial site and all beneficiaries via e-mail. The monitoring manual served as a quality plan that complies with Good Clinical Practice (GCP) guidelines. Additionally, monitoring documentation forms (checklists, report forms, log- & communication-sheets, etc.) and standard operation procedures (SOPs) to support site monitors performing standardized local monitoring activities were developed and adapted throughout the study. Before monitoring started, two online monitoring trainings were conducted to prepare site monitors for the onsite- and offsite-visits at Baseline- (T0) and FU1-(T1) Monitoring. Site monitors and the coordinating monitors discussed all arising questions. Baseline monitoring was performed from April 2015 until April 2016. Altogether, site monitors conducted onsite-visits in 51 of 359 randomized practices and checked 3,919 out of 3,919 Informed Consent Forms during baseline-monitoring. 15 patients had to be excluded due to a lack of a signed informed consent form or withdrawal of consent before randomization, leaving 3904 patients in the intention-to-treat study-population. The most time-consuming aspects mentioned by the site monitors were the search for source data due to huge differences in the quality of documentation within the normal practice routine. Written documentation in practices often lacked information, especially regarding medication, symptoms, start & stop dates, dosage and diagnoses.
FU1-(T1)-monitoring was conducted from March to October 2016. In total, site monitors performed offsite-visits in 80 of 348 included practices at T1. For two random samples of 10% each of the practices, site monitors checked 558 out of 3,904 included patients at T1. Additionally, 100% of all dropouts were monitored. All conspicuous cases detected were reported by using the respective monitoring forms.
The final monitoring of FU3 (T3) was conducted from October 2016 to November 2017. Overall, 3,210 patients in 342 practices were monitored for primary endpoints during the offsite-visits at T3. Moreover, 656 reported dropouts were monitored regarding the primary endpoints (hospitalization, death). Additionally, further items (e. g. medication) were checked for three patients each in 10% of all monitored practices. The majority of GPs and practice personnel were cooperative during monitoring, but some complained about the time needed. For 694 patients of the intention-to-treat-population, monitoring could not be completed or performed at all due to refusal of the GP, drop out, lost-to-follow-up or other reasons. These patients were included in the intention-to-treat analysis with their respective known time-under-risk (date of last available data on hospitalization or death). All detected discrepancies were documented in the respective monitoring documents and discussed with the coordinating monitor.
1.3.3 Main results of the PRIMA-eDS-trial
Baseline characteristics of the study sample are presented in tables 4-7. On average, participants were taking 10.5 substances (±2.4) and had 9.5 diagnoses (±4.9). HMG CoA reductase inhibitors were the most commonly used drug, followed by proton pump inhibitors, selective beta blocking agents, platelet aggregation inhibitors, ACE inhibitors, loop diuretics, and dihydropyridine derivatives. Essential (primary) hypertension was the most common diagnosis, followed by disorders of lipoprotein metabolism, diabetes mellitus type 2, and arthrosis.
Table 4 Demographic and clinical characteristics of the study population
Characteristics All subjects
(n=3904) Control group (n=1951) Intervention group (n=1953)
N N N
Sociodemographic data
Age
<85 (n, %) 3036 77.8 1511 77.5 1525 78.1
≥85 (n, %) 868 22.2 440 22.6 428 21.9
mean ±SD (years) 3904 81.5 ±4.4 1951 81.5 ±4.5 1953 81.5 ±4.4
Female, n (%)
2240 57.4 1137 58.3 1103 56.5
Male, n (%) 1664 42.6 814 41.7 850 43.5
Educational level, n (%)*
Low 1536 39.3 748 38.3 788 40.4
Medium 1465 37.5 726 37.2 739 37.8
High 577 14.8 292 15.0 285 14.6
Health-related factors
Smokers, n (%)* 154 3.9 88 4.5 66 3.4
BMI, n (%)*
BMI <18.5 34 0.9 19 1.0 15 0.8
BMI 18.5-24 957 24.5 474 24.3 483 24.7
BMI 25-29 1606 41.1 790 40.5 816 41.8
BMI ≥30 1307 33.5 668 34.2 639 32.7
Frailty level, n (%)*
Managing well 1643 42.1 771 39.5 872 44.6
Vulnerable 868 22.2 442 22.7 426 21.8
Mildly frail 660 16.9 362 18.6 298 15.3
Moderately frail 505 13.0 250 12.8 255 13.1
Severely frail 97 2.5 53 2.7 44 2.3
Very severely frail 8 0.2 5 0.3 3 0.2
Physical health composite score, median (range)*
3484 36.6
(10-68) 1710 36.6
(10-63) 1774 36.7
(11-68)
Mental health composite score, median (range)*
3483 48.7
(12-76) 1710 48.7
(12-76) 1773 48.7
(12-72)
Study centre, n (%)
PMU 587 15.0 295 15.1 292 15.0
UMR 981 25.1 506 25.9 475 24.3
UWH 742 19.0 369 18.9 373 19.1
SAGP 901 23.1 450 23.1 451 23.1
UNIMAN 693 17.8 331 17.0 362 18.5
Substances, n (mean ±SD) 3904 10.5 ±2.4 1951 10.5 ±2.4 1953 10.6 ±2.5
Diagnoses, n (mean ±SD) 3904 9.5 ±4.9 1951 9.7 ±5.4 1953 9.3 ±4.4
Legend: SD= Standard deviation, BMI= Body Mass Index, N= number; CG= control group, IG= intervention group, * the total n given above does not correspond to the n within this variable due to missing data.
Table 5 Percentage of the population using substances (ATC level 4)
Substances All subjects Control group Intervention group
N (%) N (%) N (%)
C10AA HMG CoA reductase inhibitors 2479 63.5 1254 64.3 1225 62.7
A02BC Proton pump inhibitors 2328 59.6 1152 59.1 1176 60.2
C07AB Beta blocking agents, selective 2240 57.4 1088 55.8 1152 59.0
B01AC Platelet aggregation inhibitors excl. heparin 1952 50.0 977 50.1 975 49.9
C09AA ACE inhibitors, plain 1751 44.9 860 44.1 891 45.6
C03CA Loop diuretics 1715 43.9 838 43.0 877 44.9
C08CA Dihydropyridine derivates 1611 41.3 836 42.9 775 39.7
C09CA Angiotensin II antagonists, plain 1355 34.7 707 36.2 648 33.2
C03AA Thiazides, plain 1354 34.7 716 36.7 638 32.7
A11CC Vitamin D and analogues 1120 28.7 548 28.1 572 29.3
Legend: CG= control group, IG= intervention group
Table 6 Most common diagnoses of the study population
Diagnoses All subjects
N (%) N (%) N (%)
I10 Essential (primary) hypertension 3428 87.8 1708 87.6 1720 88.1
E78 Disorders of lipoprotein metabolism 2078 53.2 1060 54.3 1018 52.1
E11 Diabetes mellitus type 2 1850 47.4 925 47.4 925 47.4
M19 Osteoarthritis 1752 44.9 821 42.1 931 47.7
I25 Chronic ischaemic heart disease 1473 37.7 719 36.9 754 38.6
M54 Dorsalgia 1442 36.9 715 36.7 727 37.2
I48 Atrial fibrillation and flutter 1172 30.0 580 29.7 592 30.3
I50 Heart failure 1142 29.3 536 27.5 606 31.0
K21 Gastro-oesophageal reflux disease 982 25.2 497 25.5 485 24.8
F32 Depressive episode 853 21.9 416 21.3 437 22.4
Legend: CG= control group, IG= intervention group
Table 7 Demographic and clinical characteristics per study centre
Characteristics
PMU UMR UWH SAGP UNIMAN
N N N N N
Sociodemo-graphic data
Age
<85 (n, %)
429 73.1 771 78.6 568 76.5 712 79.0 556 80.2
≥85 (n, %) 158 26.9 210 21.4 174 23.5 189 21.0 137 19.8
mean ±SD (y) 587 82.2 ±4.6 981 81.5 ±4.3 742 81.6 ±4.5 901 81.3 ±4.4 693 81.1 ±4.4
Female, n (%) 362 61.7 606 61.8 415 55.9 521 57.8 336 48.5
Male, n (%)
225 38.3 375 38.2 327 44.1 380 42.2 357 51.5
Educational level, n (%)*
Low
230 39.2 152 15.5 245 33.0 598 66.4 311 44.9
Medium
222 37.8 608 62.0 346 46.6 92 10.2 197 28.4
High
59 10.0 217 22.1 132 17.8 21 2.3 148 21.4
Health-related factors
Smokers, n (%)*
11 1.9 45 4.6 23 3.1 44 4.9 31 4.5
BMI, n (%)*
BMI <18.5
5 0.8 8 0.8 8 1.0 7 0.8 6 0.9
BMI 18.524
174 29.6 195 19.9 157 21.2 255 28.3 176 25.4
BMI 25-29
224 38.2 384 39.1 316 42.6 385 42.7 297 42.8
BMI ≥30
184 31.4 394 40.2 261 35.2 254 28.2 214 30.9
Frailty level, n (%)*
Managing well
245 41.7 415 42.3 255 34.4 376 41.7 352 50.8
Vulnerable
124 21.1 229 23.3 178 24.0 170 18.9 167 24.1
Mildly frail
91 15.5 158 16.1 125 16.8 171 19.0 115 16.6
Moderately
frail
84 14.3 139 14.2 144 19.4 92 10.2 46 6.6
Severely frail
21 3.6 28 2.9 19 2.6 23 2.5 6 0.9
Very severely
frail
2 0.3 0.2 4 0.5 0 0.0 0 0.00
Physical health composite score, median (range)* 509 36.7 (13-62) 948 35.9 (12-63) 713 35.0 (11-62) 675
38.7 (11-68) 639
37.5 (10-60)
Mental health composite
score, median (range)* 506 47.6 (16-74) 946 48.6 (12-72) 715 47.5 (17-76) 676 46.8 (14-72) 640 53.5 (16-74)
Substances
<8 259 44.1 351 35.8 334 45.0 439 48.7 261 37.7
≥10 328 55.9 630 64.2 408 55.0 462 51.3 432 62.3
n (mean ±SD)
587 10.4 ±2.3 981 10.9 ±2.7 742 10.4 ±2.4 901 10.0 ±2.0 693 10.9 ±2.7
Diagnoses, n (mean ±SD)
587 8.5±3.5 981 12.9 ±6.2 742 10.4 ±4.5 901 7.6 ±2.7 693 6.8 ±3.5
Legend: N= Number SD= Standard deviation, BMI= Body Mass Index, * the total n given above does not correspond to the n within this variable due to missing data.
Figure 6 depicts details of the consort flow chart. 3904 patients were randomized and considered for the ITT analysis. 3492 patients were included in the per-protocol analysis.
Figure 6: Consort-Flow Chart Intention to treat analysis and per-protocol analyses
Abbrevations: ITT= intention to treat; PP= per protocol, IG= intervention group, CG= control group
In the intention-to-treat analysis at follow-up 3 (after 24 months), the hazard ratio (HR) for the combined endpoint for all study centers was 0.92 (95% CI 0.82-1.04) thus just missing significance (see table 8). The time-to-even-graph (survival-plot) of the composite primary endpoint shows a clear trend in favor of the intervention (see fig. 4). When the combined endpoint was stratified by research center, PMU had a significantly lower hazard ratio between intervention group and control group in favor of the intervention [0.68 (95% CI 0.50-0.94) than UWH, SAGP, UNIMAN, and UMR (see table 9). The HR for death was 0.92 (95% CI 0.74-1.14) and for the first hospitalization 0.93 (95% CI 0.82-1.06). Table 10 presents a descriptive analysis of the number of events (death/first hospitalization, death, and hospitalization) according to research center.
Table 8 Cox regression analysis of the composite primary endpoint (death/first unplanned hospitalization), and the secondary endpoints all cause death, and first unplanned hospitalization in ITT analysis
Outcome Treatment Hazard ratio (95% CI) p
Composite endpoint death/first unlanned hospitalization IG vs CG 0.92 (0.82-1.04) 0.19
All cause death IG vs CG 0.92 (0.74-1.14) 0.43
First unplanned hospitalization IG vs CG 0.93 (0.82-1.06) 0.28
Table 9 Cox regression analysis of the composite primary endpoint (death/first unplanned hospitalization) stratified by research center in ITT analysis
Research center Treatment Hazard ratio
(95% CI) P
SAGP IG vs CG 0.97 (0.76-1.24) 0.79
UNIMAN IG vs CG 0.94 (0.73-1.20) 0.61
UMR IG vs CG 0.99 (0.75-1.31) 0.93
PMU IG vs CG 0.68 (0.50-0.94) 0.02
UWH IG vs CG 1.05 (0.82-1.33) 0.72
Table 10 Number of events (death/first unplanned hospitalization, death, and hospitalization) according to research center
Death and/or hospitalisation Death Hospitalisation
No Yes No Yes No Yes
Research Center Treatment N % N % N % N % N % N %
SAGP CG 296 65.8 154 34.2 390 86.7 60 13.3 312 69.3 138 30.7
IG 300 66.5 151 33.5 411 91.1 40 8.9 310 68.7 141 31.3
Total 596 66.1 305 33.9 801 88.9 100 11.1 622 69.0 279 31.0
UNIMAN CG 207 62.5 124 37.5 304 91.8 27 8.2 217 65.6 114 34.4
IG 237 65.5 125 34.5 339 93.6 23 6.4 244 67.4 118 32.6
Total 444 64.1 249 35.9 643 92.8 50 7.2 461 66.5 232 33.5
UMR CG 255 50.4 251 49.6 465 91.9 41 8.1 265 52.4 241 47.6
IG 251 52.8 224 47.2 428 90.1 47 9.9 262 55.2 213 44.8
Total 506 51.6 475 48.4 893 91.0 88 9.0 527 53.7 454 46.3
PMU CG 103 34.9 192 65.1 248 84.1 47 15.9 119 40.3 176 59.7
IG 142 48.6 150 51.4 256 87.7 36 12.3 153 52.4 139 47.6
Total 245 41.7 342 58.3 504 85.9 83 14.1 272 46.3 315 53.7
UWH CG 146 39.6 223 60.4 317 85.9 52 14.1 159 43.1 210 56.9
IG 152 40.8 221 59.2 314 84.2 59 15.8 165 44.2 208 55.8
Total 298 40.2 444 59.8 631 85.0 111 15.0 324 43.7 418 56.3
Total CG 1007 51,6 944 48,4 1724 89,7 197 10,3 1072 54,9 879 45,1
IG 1082 55,4 871 44,6 1748 89,5 205 10,5 1134 58,1 819 41,9
Total 2089 53.5 1815 46.5 3472 88.9 432 11.1 2206 56.5 1698 43.5
Legend: CG: control group, IG: intervention group
Figure 7: Survival plot combined endpoint (ITT Analysis)
Table 11 describes the number of drugs at baseline, at the end of the study and the change between the two points according to study centre.
Table 11: Number of drugs per patient and study centre
Research center N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline SAGP (Italy) 901 10.00 1.95 10.0 6 25 9.0 11.0
UNIMAN (UK) 693 10.86 2.72 10.0 5 24 9.0 12.0
UMR (Germany1) 981 10.89 2.68 10.0 5 26 9.0 12.0
PMU (Austria) 587 10.43 2.31 10.0 7 21 9.0 12.0
UWH (Germany2) 742 10.35 2.33 10.0 6 23 9.0 12.0
Overall 3904 10.51 2.44 10.0 5 26 9.0 12.0
Number of drugs (total) End of study SAGP (Italy) 901 9.20 2.54 9.0 1 18 8.0 11.0
UNIMAN (UK) 693 9.95 3.19 10.0 0 26 8.0 12.0
UMR (Germany1) 981 10.56 3.26 10.0 0 26 8.0 12.0
PMU (Austria) 587 9.73 3.04 9.0 0 21 8.0 12.0
UMR (Germany2) 742 10.01 3.03 10.0 0 26 8.0 12.0
Overall 3904 9.91 3.05 10.0 0 26 8.0 12.0
Change number of drugs (total) End-Baseline SAGP (Italy) 901 -0.80 2.05 0.0 -13 6 -2.0 0.0
UNIMAN (UK) 693 -0.92 2.45 -1.0 -16 7 -2.0 0.0
UMR (Germany1) 981 -0.33 2.45 0.0 -14 11 -1.0 1.0
PMU (Austria) 587 -0.70 2.44 0.0 -15 7 -2.0 1.0
UWH (Germany2) 742 -0.34 2.30 0.0 -14 10 -1.0 1.0
Overall 3904 -0.60 2.35 0.0 -16 11 -2.0 1.0
Table 12 depicts the number of drugs at baseline, at the end of the study, and the change between the two points of measurement according to treatment group. Overall, the number of drugs per patient were reduced by a mean of -0.599 drugs per patient. The mean reduction was -0.841 in the intervention group, compared to -0.356 in the control group.
Table 12: Number of drugs per patient and treatment group
Treatment N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline Control group (CG) 1951 10.46 2.41 10.0 5 23 9.0 12.0
Intervention group (IG) 1953 10.55 2.48 10.0 5 26 9.0 12.0
Overall 3904 10.5 2.44 10.0 5 26 9.0 12.0
Number of drugs (total)
End of study Control group (CG) 1951 10.11 3.01 10.0 0 26 8.0 12.0
Intervention group (IG) 1953 9.71 3.09 9.0 0 26 8.0 11.0
Overall 3904 9.91 3.05 10.0 0 26 8.0 12.0
Change number of drugs (total)
End-Baseline Control group (CG) 1951 -0.36 2.29 0.0 -16 11 -1.0 1.0
Intervention group (IG) 1953 -0.84 2.38 0.0 -15 9 -2.0 0.0
Overall 3904 -0.60 2.35 0.0 -16 11 -2.0 1.0
Table 13 depicts the mean number of drugs at baseline, at the end of the study, and the change between baseline and the end of the study according to research centre and treatment group. The largest reduction in the number of drugs in the intervention group was achieved in Italy (SAGP), with a reduction of 1.26 drugs per patient. The smallest reduction was seen in Germany 2 (UWH) with a reduction of only 0.5 drugs per patient. Interestingly, the reduction of drugs in the UK was larger in the control group than in the intervention group.
Table 13: Number of drugs per patient, research center and treatment
Study centre Treatment group N Mean SD Median Min Max Lower quartile Upper quartile
Number of drugs at Baseline SAGP (Italy) Control group (CG)
450 9.96 1.93 9.0 6 17 8.0 11.0
Intervention group (IG) 451 10.03 1.97 10.0 7 25 9.0 11.0
UNIMAN (UK) Control group (CG) 331 10.80 2.69 10.0 5 21 9.0 12.0
Intervention group (IG) 362 10.92 2.75 10.0 5 24 9.0 12.0
UMR (Germany1) Control group (CG) 506 10.83 2.50 10.0 5 23 9.0 12.0
Intervention group (IG) 475 10.95 2.87 10.0 5 26 9.0 12.0
PMU (Austria) Control group (CG) 295 10.33 2.39 10.0 8 21 8.0 12.0
Intervention group (IG) 292 10.54 2.22 10.0 7 17 9.0 12.0
UWH (Germany2) Control group (CG) 369 10.39 2.41 10.0 6 19 9.0 12.0
Intervention group (IG) 373 10.31 2.26 10.0 7 23 9.0 11.0
Number of drugs (total) End of study SAGP (Italy) Control group (CG) 450 9.63 2.45 9.0 3 18 8.0 11.0
Intervention group (IG) 451 8.77 2.56 9.0 1 18 7.0 10.0
UNIMAN (UK) Control group (CG) 331 9.80 3.11 10.0 0 20 8.0 12.0
Intervention group (IG) 362 10.08 3.26 10.0 3 26 8.0 12.0
UMR (Germany1) Control group (CG) 506 10.78 3.05 10.0 0 21 9.0 12.0
Intervention group (IG) 475 10.33 3.46 10.0 0 26 8.0 12.0
PMU (Austria) Control group (CG) 295 9.93 3.24 10.0 0 21 8.0 12.0
Intervention group (IG) 292 9.54 2.82 9.0 0 19 8.0 11.0
UWH (Germany2) Control group (CG) 369 10.21 3.12 10.0 1 26 8.0 12.0
Intervention group (IG) 373 9.81 2.93 10.0 0 22 8.0 11.0
Change number of drugs (total) End-Baseline SAGP (Italy) Control group (CG) 450 -0.34 1.89 0.0 -13 6 -1.0 0.0
Intervention group (IG) 451 -1.26 2.10 -1.0 -10 4 -3.0 0.0
UNIMAN (UK) Control group (CG) 331 -1.01 2.61 -1.0 -16 7 -2.0 0.0
Intervention group (IG) 362 -0.84 2.29 -1.0 -9 5 -2.0 0.0
UMR (Germany 1) Control group (CG) 506 -0.05 2.22 0.0 -12 11 -1.0 1.0
Intervention group (IG) 475 -0.62 2.65 0.0 -14 9 -2.0 1.0
PMU (Austria) Control group (CG) 295 -0.40 2.39 0.0 -13 7 -1.0 1.0
Intervention group (IG) 292 -1.00 2.46 -1.0 -15 5 -2.0 0.0
UWH (Germany 2) Control group (CG) 369 -0.18 2.34 0.0 -14 10 -1.0 1.0
Intervention group (IG) 373 -0.50 2.26 0.0 -11 6 -1.0 0.0
The treatment effect on the number of drugs at the end of the study was tested in a multi level analysis of covariance with “numbers of drugs at baseline” as a co-variate. The cluster factors “research centre”, the interaction between “research centre” and “treatment” and the random GP practice effect were all showing significant effects (see table 14). There also was a significant treatment effect between the control and the intervention group (p >0.0003).
Table 14: Ancova-table of treatment effect on the number of medications
Effect p
Number of drugs at baseline <.0001
Treatment 0.0003
Research center 0.0006
Research center *treatment 0.0398
In the per-protocol analysis for the combined endpoint (first hospitalization/death) the HR was 0.880 (95% CI 0.779-0.994) p=0.040 and thus statistical significance was reached. Figure 8 depicts the survival plot of the primary composite endpoint. Table 15 depicts the events of the endpoints death, hospitalization, and the combined endpoint according to research centre.
Figure 8: Survival Plot of the per-protocol analysis regarding the primary endpoint
Table 15: Per-protocol analysis of primary composite endpoint
Research center Treatment Death and/or hospital stay Death Hospital stay
No Yes No Yes No Yes
N % N % N % N % N % N %
SAGP (Italy) Control group (CG) 262 63.1 153 36.9 355 85.5 60 14.5 278 67.0 137 33.0
Intervention group (IG) 280 65.9 145 34.1 385 90.6 40 9.4 290 68.2 135 31.8
Total 542 64.5 298 35.5 740 88.1 100 11.9 568 67.6 272 32.4
UNIMAN (UK) Control group (CG) 184 60.5 120 39.5 277 91.1 27 8.9 194 63.8 110 36.2
Intervention group (IG) 194 63.2 113 36.8 286 93.2 21 6.8 201 65.5 106 34.5
Total 378 61.9 233 38.1 563 92.1 48 7.9 395 64.6 216 35.4
UMR (Germany) Control group (CG) 203 47.1 228 52.9 394 91.4 37 8.6 211 49.0 220 51.0
Intervention group (IG) 194 49.7 196 50.3 349 89.5 41 10.5 200 51.3 190 48.7
Total 397 48.4 424 51.6 743 90.5 78 9.5 411 50.1 410 49.9
PMU (Austria) Control group (CG) 92 32.6 190 67.4 235 83.3 47 16.7 108 38.3 174 61.7
Intervention group (IG) 135 48.9 141 51.1 240 87.0 36 13.0 146 52.9 130 47.1
Total 227 40.7 331 59.3 475 85.1 83 14.9 254 45.5 304 54.5
UWH (Germany) Control group (CG) 123 37.7 203 62.3 276 84.7 50 15.3 135 41.4 191 58.6
Intervention group (IG) 135 40.2 201 59.8 278 82.7 58 17.3 148 44.0 188 56.0
Total 258 39.0 404 61.0 554 83.7 108 16.3 283 42.7 379 57.3
Total Control group (CG) 864 49.1 894 50.9 1537 87.4 221 12.6 926 52.7 832 47.3
Intervention group (IG) 938 54.1 796 45.9 1538 88.7 196 11.3 985 56.8 749 43.2
Total 1802 51.6 1690 48.4 3075 88.1 417 11.9 1911 54.7 1581 45.3
Overall, the effectiveness of the PRIMA-eDS tool regarding the reduction of polypharmacy could be shown in our trial. The primary endpoint – the composite of death and first unplanned hospitalisation could be shown to be significant in per protocol analysis only. This result points to the fact that the tool has not been used as extensively as anticipated. The survey carried out to examine the usability of the tool also revealed that not all GPs in the intervention group made use of the tool as they were instructed to do, due to several barriers. One of the most important barriers turned out to be the lack of integration of the tool in the electronic health record used in the practices. This required extra time and effort of the GPs having to enter patients’ medication and other data already present in the EHR once again in the eCRF. Further analyses are ongoing to take a more detailed look at which drugs have been discontinued, and to check whether there are any correlations between usage of the tool according to the answers in the usability survey and the actual discontinuation of drugs by the respective GP in the trial.
Potential Impact:
1.4 Potential Impact, Main Dissemination Activities and Exploitation of Results
1.4.1 Impact of PRIMA-eDS on population health and health expenditure
Although the PRIMA-eDS intervention could not be shown to have an impact on the primary endpoint of the PRIMA-eDS-trial in the intention-to-treat population, quite an important impact on population health may be expected, as per-protocol-analysis revealed a significant reduction in hospitalization and death. In absolute numbers, the risk for the primary endpoint event was 45.9% in the intervention group and 50.9% in the control group, accounting for an absolute risk reduction of 5% in an observation period of two years. This means that one hospitalisation or death is avoided per year for one out of 40 patients cared for with the PRIMA-eDS-tool used decision support for avoiding inappropriate medication.
Roughly it can be estimated that there are more than 4.4 Million patients with polypharmacy in Germany (approximately 25% of all persons over 65 years). Thus, a nation wide implementation and usage of the PRIMA-eDS-tool would help to avoid 110,000 hospital stays or deaths per year. Extrapolating these numbers to the EU with about 500 Million people would lead to a tremendous impact on health and healthcare. European statistics estimate that 19.2% (96 Million) of the European population are older than 65 years (48). Even a conservative estimate of a polypharmacy rate of only 15% (14.4 Million) would result in avoiding 360,000 hospitalisations or deaths.
The second important result of the PRIMA-eDS trial revealed discontinuation of approximately 0.8 drugs per patient in the intervention group. When estimating average annual costs of only 100 €/drug, the discontinuation of 0.8 drugs/patient corresponds to health expenditure savings of 80 €/year/patient when using the PRIMA-eDS-tool. Using the population estimates from above this would correspond to annual savings of (14.4 Million x 80 €) 1.152 Million €.
1.4.2 Main dissemination activities
Ther PRIMA-eDS project has been made well known by the consortium already during the time of the project. Eleven peer-reviewed journal articles have been published (10–17,46,49,50). Two peer-reviewed articles are currently under review. Several further articles are being prepared, among others the main publication of the trial results. The project has been presented at numerous national and international conferences (see section 2).
1.4.3 Exploitation of results
The PRIMA-eDS-tool has been continuously updated and optimized during the trial by Duodecim and the PRIMA-eDS-consortium. It has been incorporated by Duodecim in its EbMeDS environment thus making the PRIMA-eDS tool available for further use in the Finnish health care system as “Comprehensive Medication Review”. Information about the availability of this “Comprehensive Medication” Reviews has been translated into English and German, and DMP and UWH are currently seeking possibilities to implement the tool in the electronic health records of the participating countries. Contacts with EHR-vendors have been made, and contacts with policy-makers are planned. The information on the “Comprehensive Medication Review” provided by DMP is attached as appendix 1.
1.5 Project Public Website
We established, and are maintaining a PRIMA-eDS-website (www.prima-eds.eu) giving relevant information about the project and the project achievements for the public. The website (www.prima-eds.eu) has been installed during the first months of the project and has been updated regularly since. The website has been used both for publishing of the PRIMA-eDS-project and for internal communication of the consortium. It has also been used as an entrance gateway for the GPs to access the eCRF forms, and it served as an online resource for the GPs regarding instruction on eCRF- and eDS-tool-usage. This service has now been discontinued as the main PRIMA-eDS-trial has been completed.
The website will be updated with the upcoming publications. The main results are withheld at the moment to avoid public availability of main results before the acceptance of the main manuscript by a peer-reviewed journal.
2 Use and Dissemination of Foreground
The following publication plan has been established:
Publication Status
Title: Polypharmacy in chronic diseases–Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): study protocol for a randomized controlled Trial
Authors: Andreas Sönnichsen, Ulrike S. Trampisch, Anja Rieckert, Giuliano Piccoliori, Anna Vögele, Maria Flamm, Tim Johansson, Aneez Esmail, David Reeves, Christin Löffler, Jennifer Höck, Renate Klaassen-Mielke, Hans Joachim Trampisch, Ilkka Kunnamo
Journal: Trials
Language: English
Published
Title/ concept: Polypharmacy in chronic diseases– Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): a cross-sectional analysis of factors predisposing for excessive polypharmacy
Authors: Anja Rieckert, Ulrike Trampisch, Renate Klaasen-Mielke, Eva Drewelow, Annez Esmail, Tim Johansson, Sophie Keller, Ilkka Kunnamo, Christin Löffler,
Joonas Mäkinen, Giuliano Piccoliori, Anna Vögele, Andreas Sönnichsen
Data: Baseline data
Journal: BMC Family Practice
Language: English
Submitted (Peer-review)
Title/ concept: Polypharmacy in chronic diseases– Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): a survey of GPs experiences
Authors: Anja Rieckert, Anne-Lisa Teichmann, Eva Drewelow, Celine Kriechmayr, Giuliano Piccoliori, Adrine Woodham, Andreas Sönnichsen
Data: Survey
Journal:
Language: English
Drafted, expected to be submitted end of March
Title/concept: Publication of our primary and secondary endpoints
Authors: Anja Rieckert, (to be determined), Ulrike Trampisch, Sabine Weissbach, Andreas Sönnichsen
Data: primary endpoint: composite endpoint of first non-elective hospital admission or death during the observation period measured as a binary outcome.
Secondary endpoint: all-cause mortality, nonselective hospital admission (number of episodes and duration), falls (number and severity), fractures, quality of life (SF-12v2), the number and types of drugs (total number, number discontinued, number not discontinued despite the recommendation to discontinue, number re-administered for symptom control), adverse event rate, and medication costs over the observation period
Journal: to be determined
Language: English
Currently drafted, analyses not finished yet
Title/concept: Predictors of GP ratings of patient frailty. This will be an analysis of the baseline dataset to identify patient variables that predict GP ratings of patient frailty. Variables include diagnoses, polypharmacy, BMI, falls, SF12, etc. The paper may include construction of a “frailty index” from the data in the eCRF and correlating this with the GP assessment.
Authors: David Reeves, (to be determined)
Data: Baseline Data
Journal:
Language: English
Not started
Title/concept: GP ratings of patient frailty as a predictor of patient outcomes. This will make use of the full dataset (i.e. all two years), to explore whether GP frailty ratings predict future patient outcomes, including death, hospitalisations, falls, QoL, etc, both with and without control for other baseline factors. It may also compare frailty ratings with other baseline predictors of outcomes for predictive ability, including polypharmacy and a frailty index constructed from the eCRF data. It will also explore whether frailty ratings moderate the association between polypharmacy and outcomes.
Authors: David Reeves, (to be determined)
Data: full dataset
Journal:
Language: English
Not started
Title/concept: Barriers for implementing electronic decision support tools, Qualitative synthesis
Authors: Malin Wörster, (to be determined), Andreas Sönnichsen
Data: pilot study & literature search
Journal:
Language: Conducting the search
Title/concept: Validation of the Pharao table. Correlation of symptoms with Pharao table.
Authors: (to be determined), Andreas Sönnichsen
Data: Baseline Data+ CMR (Pharao)
Journal: to be determined
Language: English
Developing an analysable database/development of concept
Title/concept: What is the potential of the tool? Analysis of the recommendations. Which drugs recommends the tool to discontinue?
Authors: Anja Rieckert (?) (to be determined), Andreas Sönnichsen
Giuliano is interested in contribution. Experience with PRIMA Italy (based on experts evaluations).
Data: CMR data
Journal: to be determined
Language: English
Developing an analysable database
Title/ concept: Analysis of what did the tool recommend to discontinue and what was actually discontinued.
Data: CMR+ 1st follow-up
Journal: to be determined
Language: English
Developing an analysable database
Title/concept: Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis
Authors: Tim Johansson, Muna E. Abuzahra, Sophie Keller, Eva Mann, Barbara Faller, Christina Sommerauer, Jennifer Höck, Christin Löffler, Anna Köchling, Jochen Schuler, Maria Flamm and Andreas Sönnichsen
Data: SR strategies to reduce polypharmacy
Journal: Br J Clin Pharmacol
Language: English
Published
Title/ concept: Interventions to reduce inappropriate polypharmacy: Implications for research and practice.
Authors: Tim Johansson, Maria Flamm, Andreas Sönnichsen
Data:
Journal: Elsevier Maturitas
Language: English
Published
Title/ concept: Letter to the editor. Reply to ‘Endpoints in strategies to reduce
polypharmacy’.
Authors: Tim Johansson, Maria Flamm, Andreas Sönnichsen, Jochen Schuler
Data:
Journal: British Journal of Clinical Pharmacology
Language: English
Published
Title/ concept: A set of systematic reviews to help reduce inappropriate prescribing to older people: study protocol
Authors: Yolanda V Martinez, Anna Renom-Guiteras, David Reeves, R Erandie Ediriweera de Silva, Aneez Esmail, Ilkka Kunnamo, Anja Rieckert, Christina Sommerauer, Andreas Sönnichsen
Data:
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Effectiveness and safety of Dipeptidyl peptidase 4 inhibitors in the management of type 2 diabetes in older adults: a systematic review and the development of recommendations to reduce inappropriate prescribing
Authors: Gisela Schott, Yolanda V Martinez, Erandie Ediriweera de Silva, Anna Renom-Guiteras, Anna Vögele, David Reeves, Ilkka Kunnamo, Minna Marttila-Vaara, Andreas Sönnichsen
Data: SR gliptins/diabetes
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review to help reduce inappropriate prescribing
Authors: Anna Vögele, Tim Johansson, Yolanda V Martinez, Lisa Schlender, Anna Renom-Guiteras, Anja Rieckert, Anne-Lisa Teichmann, Andreas Sönnichsen
Data: SR ß-Blocker/Hypertension
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Thiazides in the management of hypertension in older adults – a systematic review
Authors: Christina Sommerauer, Neha Kaushik, Adrine Woodham, Anna Renom-Guiteras, Yolanda V Martinez, David Reeves, Ilkka Kunnamo, Steffen Hübner,
Andreas Sönnichsen
Data: SR Thiazides/Hypertension
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Efficacy and Patient safety with Vitamin-K-antagonists and new anticoagulants in the Treatment/Prevention of Thromboembolism in Atrial Fibrillation in older adults – an overview of reviews
Authors: Christina Sommerauer, Lisa Schlender, Marc Krause, Sabine Weißbach, Anja Rieckert, Yolanda V Martinez, Anna Renom-Guiteras, Ilkka Kunnamo, Andreas Sönnichsen
Data: SR VKA/NOACs atrial fibrillation
Journal: BMC Geriatrics
Language: English Published
Title/concept: Efficacy and Patient safety with platelet aggregation inhibitors in the management of cerebral infarction, transient ischaemic attacks, peripheral artery occlusive disease and coronary disease in older adults – a systematic review
Authors: Maren Meinshausen, Anja Rieckert, Anna Renom-Guiteras, Moritz Kröger, Christina Sommerauer, Ilkka Kunnamo, Yolanda V Martinez, Aneez Esmail, Andreas Sönnichsen
Data: SR PAI
Journal: BMC Geriatrics
Language: English
Published
Title/concept: Efficacy and safety of metformin in the management of type 2 Diabetes mellitus in older adults: a systematic review for the development of recommendations to reduce inappropriate prescriptions
Authors: Lisa Schlender, Yolanda V Martinez, Charles, David Reeves, Barbara Faller, Christina Sommerauer, Ilkka Kunnamo, Andreas Sönnichsen, Anna Renom-Guiteras
Data: SR Metformin
Journal: BMC Geriatrics
Language: English
published
Title/concept: Efficacy and patient safety with NSAIDs in the management of musculoskeletal disorders in older adults – a systematic review
Authors: Anja Rieckert, Barbara Faller, Christina Sommerauer, Lisa Schlender, Anne-Lisa Teichmann, Erandie Ediriweera de Silva, Sabine Weißbach, Anna Renom-Guiteras, Thekraiat Al’Quaran, Ilkka Kunnamo, Andreas Sönnichsen
Data: SR NSAIDs
Journal: to be determined
Language: English
1,4000 hits were screened
14 included; update completed
Title/concept: Insulin –SR
Authors: Raniah El-Jezawi, (more to be determined), Lisa Lechterbeck, Andreas Sönnichsen
Data: SR Insulin
Journal: to be determined
Language: English
Study selection 1,2, 3a completed; comprehensive search needs to be conducted
10,000 hits are currently screened
Title/concept: Insulin – Protocol
Authors: Lisa Schlender, Raniah El-Jezawi, Andreas Sönnichsen
Data: SR Insulin
Journal: to be determined
Language: English
Draft ready
Title/concept: Statins SR of SRs
Authors: Thekraiat Al’Quaran, Moritz Kröger (joint first authorship), Anja Rieckert, Anna Renom-Guiteras, David Reeves, Ilkka Kunnamo, Yolanda Martinez, Andreas Sönnichsen
Data: SR Statins
Journal: to be determined
Language: English
Update completed; currently Search 3a+3b
Title/concept: Using the electronic decision support tool PRIMA-eDS to optimize medication in primary care – a qualitative study for practical implementation
Authors: Anja Rieckert, Christina Sommerauer, Anja Krumeich, Andreas Sönnichsen
Data: Interviews with GPs from Witten
Journal: BMC Family Practice
Language: English
Re-submitted (mid-March)
Working title/concept: Validation of the tool. Comparison with recommendations from Dara Koper.
Authors: Ann-Kathrin Bücherl, (to be determined), Andreas Sönnichsen
Data: Kopers patients are entered into the PRIMA-eDS tool. Recommendations by the PRIMA-eDS tool are compared to recommendations provided by the study of Dara Koper
Journal: (to be determined)
Language: English
Database will be ready by mid-April
Title/concept: Recruitment of general practitioners for research - Experiences from a multicentre cluster-randomized controlled trial (PRIMA-eDS)
Authors: Sophie Keller, Tim Johansson, Eva Drewelow, Christin Löffler, Anja Rieckert, Ulrike Trampisch, Maria Flamm
Data: Questionnaires
Journal: BMC Medical Research Methodology
Language: English
Determination of new Journal after being rejected
Title/concept: On which quality are our clinical decisions based on? Quality of studies & systematic review – Interrater Reliability
Authors: Lisa Schlender, , Andreas Sönnichsen
Data: Search and data extraction documents
Journal: to be determined
Language: English
Analysis started
Title/concept: Validation of the PRIMA-eDS tool – comparison with data from the PRIMA study Italy
Authors: Giuliano Piccoliori, Anna Vögele
Data:
Journal:
Language:
CMR database needed
2.1 Section A: Dissemination Measures and Publications
Dissemination measures of the PRIMA-eDS-project are shown in tables 16 (A1) and 17 (A2). Table 16 (A1) depicts peer-reviewed journal articles already published. Table 17 (A2) shows other dissemination activities.
Table 16 (A1): List of scientific (peer reviewed) publications (starting with the most important ones)
No. Title Main author Title of journal Issue, date Publisher Place of Publication Year of publication Pages link or doi open access (y/n)
1 Polypharmacy in chronic diseases–Reduction of Inappropriate Medication and Adverse drug events in older populations by electronic Decision Support (PRIMA-eDS): study protocol for a randomized controlled Trial
Sönnichsen BMC Trials 17:57 29 Jan 2016 Biomed Central UK 2016 1-9 https://doi.org/10.1186/s13063-016-1177-8
y
2 Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis. Johansson Br J Clin Pharmacol. Aug;82(2): Wiley UK 2016 532-48 10.1111/bcp.12959 n
3 A set of systematic reviews to help reduce inappropriate prescribing to older people: study protocol Martinez; Renom-Guiteras BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 1-9 10.1186/s12877-017-0570-9 y
4 Interventions to reduce inappropriate polypharmacy: Implications for research and practice. Johansson Maturitas. 97 Elsevier UK 2017 66-68. https://doi.org/10.1016/j.maturitas.2016.12.007 n
5 User perspectives on an electronic decision-support tool performing comprehensive medication reviews - a focus group study with physicians and nurses Koskela BMC Medical Informatics and Decision Making 16 Biomed Central UK 2015 1-6 http://www.biomedcentral.com/1472-6947/16/6
y
6 Thiazides in the management of hypertension in older adults – a systematic review
Sommerauer BMC Geriatrics 17, Oct. 2017 Biomed Central UK 2017 145-157 10.1186/s12877-017-0576-3 y
7 Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review to help reduce inappropriate prescribing Vögele BMC Geriatrics 17, Oct. 2017 Biomed Central UK 2017 119 - 143 10.1186/s12877-017-0575-4 y
8 Efficacy and safety of metformin in the management of type 2 diabetes mellitus in older adults: a systematic review for the development of recommendations to reduce potentially inappropriate prescribing Schlender; Martinez BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 99-117 10.1186/s12877-017-0574-5 y
9 Effectiveness and safety of vitamin K antagonists and new anticoagulants in the prevention of thromboembolism in atrial fibrillation in older adults – a systematic review of reviews and the development of recommendations to reduce inappropriate prescribing Sommerauer; Schlender BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 73-98 10.1186/s12877-017-0573-6 Y
10 Effectiveness and patient safety of platelet aggregation inhibitors in the prevention of cardiovascular disease and ischemic stroke in older adults – a systematic review Meinshausen; Rieckert BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 41-71 10.1186/s12877-017-0572-7 y
11 Effectiveness and safety of Dipeptidyl peptidase 4 inhibitors in the management of type 2 diabetes in older adults: a systematic review and the development of recommendations to reduce inappropriate prescribing
Schott, Martinez BMC Geriatrics 17. Oct. 2017 Biomed Central UK 2017 11-39 10.1186/s12877-017-0571-8 Y
12 Reply to "Endpoints in strategies to reduce polypharmacy". Johansson Br J Clin Pharmacol. 83(2) Wiley UK 2017 434 DOI: 10.1111/bcp.13125 n
Table 17 (A2): List of Dissemination Activities (e.g. Conferences)
No Type of activity Main leader Title Date Place Type of audience Size of audience Countries addressed
1 Oral presentation
Johansson 17th annual meeting of the
German Network for Evidencebased Medicine 3/3/2016 Cologne (GER) Scientific community (higher education, research) 500 Austria,
Germany,
Italy
2 Oral presentation Johansson & Vögele 51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
3 Oral presentation Johansson 50th congress for general practice and family medicine of German Association of General Medicine 29/9/2016 Frankfurt am Main (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
4 Oral presentation Keller 49th congress for general practice and family medicine of German Association of General Medicine 18/9/2015 Bolzano (ITA) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
5 Oral presentation Johansson 48th congress for general practice and family medicine of German Association of General Medicine 19/9/2014 Hamburg (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
6 Oral presentation Johansson 15th annual meeting of the German Network for Evidence-based Medicine 13.3.2014 Halle, Saale (GER) Scientific community (higher education, research) 500 Austria, Germany
7 Poster Teichmann 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Germany, Austria, Italy, UK
8 Poster Schlender 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
9 Poster Faller 51th congress for general practice and family medicine of German Association of General Medicine 21.09.2017- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
10 Poster Vögele 51th congress for general practice and family medicine of German Association of General Medicine 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Italy, Austria, Germany
11 Oral presentation/ symposium Renom-Guiteras 21st International Association of Gerontology and Geriatrics (IAGG) World Congress 23.07.2017-27.07.2017 San Francisco, California USA Scientific community (higher education, research)
12 Oral presentation Rieckert 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany, Austria, Italy, UK
13 Oral presentation Rieckert 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
14 Oral presenation Sönnichsen 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017- 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
15 Poster Schlender 18 th annual meeting of the German Network for Evidence-based Medicine 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community (higher education, research) 500 Germany
16 Oral presentation Sönnichsen Polypharmazie: ein Dilemma für Arzt und Patient. Ärztlicher Qualitätszirkel. Ärztlicher Qualitätszirkel 22/09/2016 Wetter a. d. Ruhr, Germany Scientific community, Physicians Germany
17 Oral presentation Sönnichsen 15th german congress for healthcare research 05/10/2016-07/10/2016
Berlin, Germany Scientific community (higher education, research) 600 Germany
18 Oral presentation Sönnichsen 15th german congress for healthcare research 05/10/2016 – 07/10/2016 Berlin, Germany Scientific community (higher education, research) 600 Germany
19 Poster Rieckert 50 th congress for general practice and family medicine of German Association of General Medicine 29/09/2016-01/10/2016 Frankfurt am Main Scientific community (higher education, research) 600 Germany, Austria, Italy, UK
20 Poster Sommerauer 50 th congress for general practice and family medicine of German Association of General Medicine 29/09/2016-01/10/2016 Frankfurt am Main Scientific community (higher education, research) 600 Germany
21 Poster Schlager WONCA SAR Conference 2016, Colombo, Sri Lanka 11/02/16-14/02/16 Colombo, Sri Lanka Scientific community (higher education, research) Germany, Austria, Italy
22 Poster Rieckert 17th annual meeting of the German Network for Evidence-based Medicine 03/03/2016-05/03/2016 Köln, Germany Scientific community (higher education, research) 500 Germany
23 Poster Rieckert 1.Wittener research meeting 06.04.2016 Witten, Germany Scientific community (higher education, research) Germany
24 Poster Celine Kriechmayr (PMU) Einsatz der elektronischen Entscheidungshilfe zur Optimierung der Medikation in der hausärztlichen Versorgung – eine quantitative Untersuchung zur praktischen Umsetzung im Rahmen der cluster-randomisert kontrollierten Studie PRIMA-eDS
51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
25 Poster PMU/UWH/UMR/SAGP Effektivität und Patientensicherheit von Betablockern bei älteren Menschen mit Hypertonie – Update einer systematischen Übersichtsarbeit
51st congress for general practice and family medicine of German Association of General Medicine 21/9/2017 Düsseldorf (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
26 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Mediation in der hausärztlichen Versorgung – eine quantitative Untersuchung zur praktischen Umsetzung 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community(higher education, research) 600 Germany, Austria, Italy, UK
27 Poster UWH Ist Metformin immer noch die beste Option bei der Behandlung von älteren Typ-2 Diabetikern. Eine systematische Übersichtsarbeit. 21.09.17- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
28 Poster UWH Ein systematisches Review zum Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von muskuloskelettalen Erkrankungen bei älteren Menschen 21.09.2017- 23.09.2017 Düsseldorf, Germany Scientific community (higher education, research) 600 Germany
29 Oral presentation/ symposium UWH The PRIMA-eDS electronic decision support tool for polypharmacy – a multinational European project 23.07.2017-27.07.2017 San Francisco, California USA Scientific community(higher education, research)
30 Oral presentation UWH Welche Faktoren prädisponieren ältere Menschen für extreme Polypharmazie? Eine Querschnittsanalyse mit Daten aus der PRIMA-eDS-Studie
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany, Austria, Italy, UK
31 Oral presentation UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung – eine qualitative Untersuchung zur zukünftigen Implementierung
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific communit(higher education, research)y 500 Germany
32 Poster UWH Die Vorteile und Risiken von Metformin bei der Behandlung von älteren Typ-2 Diabetikern. Eine systematische Übersichtsarbeit 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany
33 Oral presentation UWH Diskrepante Ergebnisse bei Randomisiert kontrollierten Studien, Systematic Reviews und Metanalysen zu identischen Fragestellungen – Wie kann das sein?
18. Jahrestagung Deutsches Netzwerk Evidenzbasierte Medizin 09.03.2017 – 11.03.2017 Hamburg, Germany Scientific community(higher education, research) 500 Germany
34 Journal article PMU/UWH/ Interventions to reduce inappropriate polypharmacy: Implications for research and practice. Maturitas. 2017 Mar;97:66-68 3/2017 Scientific community (higher education, research) All
35 Journal article, PMU/UWH/ Reply to 'Endpoints in strategies to reduce polypharmacy'. Br J Clin Pharmacol. 2017 Feb;83(2):434 2/2017 Scientific community (higher education, research) All
36 Oral presentation UWH Reduktion unangemessener Polypharmazie durch eine elektronische Entscheidungshilfe - das PRIMA-eDS-Projekt
15. Deutscher Kongress für Versorgungsforschung 05/10/2016- 07/10/2016 Berlin, Germany Scientific community(higher education, research) 600 Germany
37 Oral presentation UWH Reduktion von Polypharmazie-Risiken für ältere Patienten - eine Implementierungs-Studie mit einer komplexen Intervention
15. Deutscher Kongress für Versorgungsforschung 05/10/2016 -07/10/2016 Berlin, Germany Scientific community(higher education, research) 600 Germany
38 Poster UWH Welche Faktoren prädisponieren ältere Menschen für extreme Polypharmazie? Eine Querschnittsanalyse mit Daten aus der PRIMA-eDS-Studie Poster UWH
39 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung Poster UWH
40 Poster PMU/UMR/UWH Die Effektivität von elektronischen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie hinsichtlich Morbidität, Lebensqualität und anderer Endpunkte – ´Update´ einer systematischen Übersichtsarbeit Poster PMU/UMR/UWH
41 Poster SAGP Effektivität und Patientensicherheit von Betablockern bei älteren Menschen mit Hypertonie - eine systematische Übersichtsarbeit [Effectiveness and safety of beta blockers in the management of hypertension in older adults: a systematic review]
50th congress for general practice and family medicine of German Association of Family Medicine 29/09/2016 -01/10/2016 Frankfurt, Germany Scientific community(higher education, research) 600 Germany, Austria
42 Poster PMU/UWH/UMR Die Effektivität von elektronischen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie hinsichtlich Morbidität, Lebensqualität und anderer Endpunkte - 'Update' einer systematischen Übersichtarbeit
50th congress for general practice and family medicine of German Association of General Medicine 29/9/2016 Frankfurt am Main (GER) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
43 Journal article PMU/UWH/UMR Impact of strategies to reduce polypharmacy on clinically relevant endpoints: a systematic review and meta-analysis.Br J Clin Pharmacol. 2016 Aug;82(2):532-48 8/8/2016 Scientific community (higher education, research) All
44 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung
1.Wittener Forschertreffen 06.04.2016 Witten, Germany Scientific community(higher education, research) ? Germany
45 Oral presentation Johansson (PMU) Elektronische und nicht-elektronische Interventionen zur Reduktion von Polypharmazie – weiterhin fehlende Evidenz für einen Patientennutzen
17th annual meeting of the German Network for Evidencebased Medicine 03/03/2016 Köln (GER) Scientific community (higher education, research) 500 Austria,
Germany,
Italy
46 Poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung
Rieckert et al
17th annual meeting of the German Network for Evidence-based Medicine 03/03/2016-05/03/2016 Köln, Germany Scientific community(higher education, research) 500 Germany
47 Oral presentation PMU/UMR/UWH Elektronische und nicht-elektronische Interventionen zur Reduktion von Polypharmazie - weiterhin fehlende Evidenz für einen Patientennutzen
Johansson et al.
17th annual meeting of the German Network for Evidence-based Medicine
03/03/2016-05/03/2016 Köln, Germany Scientific community(higher education, research) 500 Austria, Germany
48 Poster UWH Polypharmacy in chronic diseases: Reduction of Inappropriate Medication and Adverse drug events in elderly populations by electronic Decision Support (PRIMA-eDS)
Schlager et al.
WONCA SAR Conference 2016, Colombo, Sri Lanka 11/02/16-14/02/16 Colombo, Sri Lanka Scientific community(higher education, research) ? All
49 Presentation DMP Clinical Informatics Course 23/2/2016 Toronto, Canada Physicians, medical students, clinical epidemiology students 35 Several, mostly Canadian
50 Presentation DMP WONCA International Classification Committee 14/9/2016 Helsinki, Finland Physicians 25 Several
51 Presentation DMP Guidelines International Conference 29/9/2016 Philadelphia, USA Physicians, guideline developers 20 Several
52 Presentation DMP ICT conference 7/6/2017 Khanty-Mansijsk, Russia Physicians, IT professionals 50 Several (mainly Russian)
53 Presentation DMP Global Evidence Summit, Multimorbidity Working Group Workshop 14/9/2017 Cape Town, South Africa Physicians 16 Several
54 Presentation DMP EBM Conference, Sechenov Medical University 23/10/2017 Moscow, Russia Physicians, medical students 220 Russian
55 Presentation DMP PRIMA-eDS Conference 10/11/2017 Witten, Germany Physicians 35 Several
56 Presentation DMP Nordic-German Health IT Symposium 30/11/2017 Hannover Physicians, IT professionals, investors 28 German
57 seminar UNIMAN UK branch of PRIMA-eDS study
David Reeves, Adrine Woodham, Yolanda Martinez
18/02/2014 Centre for Primary Care, University of Manchester Scientific community 60 UK
58 presentation DMP WONCA World Congress, Prague: Reduction of polypharmacy in the elderly
Ilkka Kunnamo 29.6.2013 Prague Congress Centre Scientific community 65 Several
59 presentation DMP
HIMSS Conference/Nordic Day, New Orleans: Medication and adverse drug events control with decision support systems
Ilkka Kunnamo
3.3.2013 New Orleans Congress Centre Scientific community; Industry 100 Denmark, Finland, Norway, Sweden, Iceland, United States
60 presentation DMP
Guidelines International Network Conference
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 19.8.2013 San Francisco Scientific community 50 Several
61 presentation DMP
Nordic Congress on General Practice
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 23.8.2013 Tampere Scientific community 1000 Denmark, Finland, Norway, Sweden, Iceland, several
62 presentation DMP
Cochrane Colloquium
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.9.2013 Quebec City Scientific community 25 Several
63 presentation DMP
WONCA International Classification Committee (webmeeting presentation)
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.9.2013 Johannesburg Scientific community 50 Several
64 presentation DMP
Kokkola Health Centre
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.12.2012 Kokkola, Finland Civil society 20 Finland
65 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 20.3.2013 Tampere, Finland Civil society 600 Finland
66 presentation DMP
Aalto University, Health Care Leadership programme
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 16.4.2013 Helsinki, Finland Civil society, policy makers 25 Finlaned
67 presentation DMP
Norbotten and Örebro health care authorities
Reduction of polypharmacy by electronic decision support
Peter Nyberg 18.4.2013 Örebro Policy makers 10 Sweden
68 presentation DMP
National institute for Health and Welfare
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 25.4.2013 Helsinki, Finlalnd Policy makers, civil society 60 Finland, Denmark, Canada
69 presentation DMP
National project on healthcare IT architecture
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 29.4.2013 Helsinki, Finland Policy makers 30 Finland
70 presentation DMP
WoHIT/eHealthWeek
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 14.5.2013 Dublin Scientific community, Policy makers 80 Several
71 presentation DMP
Tieto Healthcare Forum
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 29.8.2013 Helsinki, Finland Policy makers, civil society 200 Finland
72 presentation DMP
Health care management course for physicians in GP specialist training
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 26.9..2013 Helsinki, Finland Civil society 65 Finland
73 presentation DMP
National Guidelines Day
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 28.9.2013 Leuven, Belgium Policy makers, civil society, scientific soviety 160 Belgium
74 presentation DMP
Educational event for young doctors in Finland
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 25.10.2013 Helsinki, Finland Civil society 500 Finland
75 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 7.11.2013 Turku, Finland Civil society 55 Finland
76 presentation DMP
Physicianss’ Association in Tampere
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 7.11.2013 Tampere, Finland Civil society 45 Finland
77 presentation DMP
Healh care leadership educational event
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 8.11.2013 Helsinki, Finland Civil society 80 Finland
78 presentation DMP
EQuiP meeting
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 15.11.2013 Bologna, Italy Civil society, Scientific community 25 Several
79 presentation DMP
Nokia Health Centre
Reduction of polypharmacy by electronic decision support
Peter Nyberg 14.12.2013 Nokia, Finland Civil society 20 Finland
80 presentation DMP
Regional educational event for physicians
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 10.1.2014 Helsinki, Finland Civil society 25 Finland
81 presentation DMP
Central Finland Physicians’ Association
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 15.2.2014 Jyväskylä, Finland Civil society 80 Finland
82 presentation DMP
Ministry of Health, Finland
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo
Timo Haikonen
Peter Nyberg 6.3.2014 Helsinki, Finland Policy makers 5 Finland
83 presentation DMP
Health care leadership course, Finnish Medical Association
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 12.3.2014 Helsinki, Finland Civil society 30 Finland
84 presentation DMP
National institute for Health and Welfare
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 18.3.2014 Helsinki, Finland Civil society 40 Finland
85 presentation DMP
Finnish Innovation Fund (Sitra)
Reduction of polypharmacy by electronic decision support
Ilkka Kunnamo 6.5.2014 Helsinki, Finland Industry, Civil society, Policy makers 170 Finland
86 journal article UMR Langer, C. (2014). Prima läuft der Arztbesuch. In: Studierendenschaft der Universität Rostock (Hrsg.). HEULER - Das Studentenmagazin der Uni Rostock auf Papier, No. 105; 04-2014, p.26. Apr 2014 Rostock civil society & scientific community 15.000 Germany
87 poster UMR Höck, J. et al. (2013). Polypharmazie bei Patienten mit chronischen Erkrankungen: Reduktion inadäquater Medikation und unerwünschter Arzneimittelwirkungen durch eine elektronische Entscheidungshilfe. Poster presented at F2-Forschung trifft Forschung-Forschungscamp, University of Rostock. 07. Nov 2013 Rostock scientific community 150 Germany
88 presentation PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
Meeting of the regional association of general practitioners 26/03/2014 Salzburg (A) scientific community 20 Austria
89 poster PMU/UWH,UMR A systematic review on strategies to reduce polypharmacy
15th annual meeting of the German Network for Evidence-based Medicine 13/03/2014 Halle (D) scientific community 400 Germany, Austria
90 poster PMU A systematic review on strategies to reduce polypharmacy
47th congress for general practice and familiy medicine of German Association of General Medicine 12/09/2013 Munich (D) scientific community 600 Germany, Austria
91 journal article PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
med.ium 11/2013 (a regional journal for physicians in Salzburg) 11/12/2013 Salzburg (A) scientific community , civil society 3600 Austria
92 presentation PMU PRIMA-eDS: Reduction of polypharmacy by electronic decision support
Information event at Paracelsus Medical University for physicians of the province of Salzburg 02/10/2013 Salzburg (A) scientific community 30 Austria
93 poster UWH and UNIMAN Efficacy and safety of drugs commonly used in the management of chronic diseases in older adults: a compilation of systematic reviews for the development of an electronic decision support tool. Renom-Guiteras A, Martinez Y et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
94 poster UWH and SAGP Wirksamkeit und Sicherheit von oralen Antidiabetika im Management von Diabetes mellitus Typ 2 bei älteren Erwachsenen – eine systematische Übersichtsarbeit [Efficacy and safety of oral antidiabetics in the management of type 2 Diabetes mellitus in older people – a systematic review]. Vögele A et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
95 poster UWH Efficacy and Patient Safety of Statins in the Prevention of Cardiovascular Events in Older Adults – a Systematic Review. Al Qur’An T et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
96 poster UWH Wirksamkeit und Sicherheit von Antikoagulantien in der Prävention von Thromboembolien bei Vorhofflimmern bei älteren Erwachsenen - eine systematische Übersichtsarbeit. [Efficacy and safety of anticoagulants in the prevention of thromboembolism in atrial fibrillation in older adults – a systematic review]. Sommerauer C et al.
15th annual meeting of the German Network for Evidence-based Medicine March 13th-15th, 2014 Halle, Germany Scientific community 400 Germany, Austria
97 presentation UWH Polypharmacy in chronic diseases: Reduction of Inappropriate Medication and Adverse drug events in elderly populations by electronic Decision Support. Sommerauer et al. 12th Annual Conference of the German network for health services research October 23rd-25th, 2013 Berlin, Germany Scientific community 700 Germany
98 presentation UWH Welche Medikamente sind evidenzbasiert bei multimorbiden älteren Patienten – eine systematische Literaturrecherche. [Which medications are evidence-based in older adults with comorbidity – compilation of systematic reviews]. Renom-Guiteras A et al. IZVF-Kolloquium at UWH May 7th, 2014 Witten, Germany Scientific community, policy makers 20 Germany
99 poster UWH Entwicklung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie: Das PRIMA-eDS Tool. [Development of an electronic decision tool for the reduction of polypharmacy: the PRIMA-eDS Tool]. Sommerauer et al.
13th Annual Conference of the German network for health services research June 24-27th, 2014 Düsseldorf, Germany Scientific community, policy makers 700 Germany
100 Workshop DMP Clinical decision support for professionals, at DECIDE Consortium meeting 3/6/2014 Edinburgh Scientific community 13 UK, Italy, Norway, Netherl.
101 Demo DMP NHS Education Scotland 5/6/2014 Glasgow Professionals 3 UK
102 Presentation DMP WONCA Europe, session on tools for chronic disease management 3/7/2014 Lisbon Primary care physicians More than 200 Several
103 Presentation DMP Clinical Decision support in guideline implementation, at Guidelines International Network conference 20/8/2014 Melbourne Scientific community, guideline developers 43 Several
104 Presentation DMP Big data and decision support in health care. Symposium by the University of Eastern Finland 9/10/2014 Kuopio, Finland Scientific community, IT develop. 160 Finland
105 Presentation DMP Comprehensive Medication Review, at Welfare ICT Forum by CGI Finland 10/10/2014 Tampere, Finland Clinicians and IT developers 70 Finland
106 Presentation DMP Clinical decision support, at EBM postgraduate course 25/2/2015 Tartu, Estonia Physicians 25 Estonia
107 Presentation DMP Experience of clinical decision support systems among Finnish doctors at Estonian Medical Convention 10/4/2015 Tallinn, Estonia Physicians 60 Estonia
108 Presentation DMP Clinical decision support at Medical Faculty, University of Helsinki 21/5/2015 Helsinki, Finland Medical students 100 Finland
109 Presentation DMP Clinical decision support at NIVA advanced course on occupational health 8/9/2015 Oslo, Norway Physicians 19 Finland, Norway, Denmark, Taiwan
110 Presentations DMP Multimorbidity working group meeting, Guidelines International Network Conference 8/10/2015 Amsterdam, Netherlands Physicians, guideline developers 20 Several
111 Face to face presentation PMU/UWH Die Effektivität von elektroni-schen und nicht-elektronischen Interventionen zur Reduktion von Polypharmazie – eine systema-tische Übersichtarbeit. 48th congress of the German Association of Family Medicine 19/09/2014 Hamburg (GER) Scientific community 600 Austria,
Germany,
Italy
112 Poster PMU Ärzte für Studien gewinnen –Erfahrungen aus der Versor-gungsforschung. 49th congress of the German Association of Family Medicine 18/9/2015 Bolzano (ITA) Scientific community (higher education, research) 600 Austria,
Germany,
Italy
113 Journal article PMU Polypharmazie im Praxisalltag bewältigen: das PRIMA-eDS-Tool ÖGAM News (Österreichische Gesellschaft für Allgemein und Familienmedizin – Mitglied der Wonca) Ausgabe 06/15 06/2015 AUT Scientific community (higher education, research) 3600 + online AUT
114 Oral presentation/ abstract publication UNIMAN Safely reducing inappropriate prescribing to the older adults with polypharmacy: the multinational PRIMA-eDS study
(Awarded the prize of best oral paper at the conference) 20/11/2015 Annual Academic Sessions of the Sri Lanka Association of Geriatric Sessions (SLAGM), Waters Edge Hotel, Battaramulla, Sri Lanka Scientific and clinician communities 400 Sri Lanka
115 Poster presentation UNIMAN EvidenceBased Recommendations for the Reduction of Polypharmacy in Older People with Multiple Chronic Conditions: A Set of Systematic Reviews.
15-16/6/2015 Academy Health - Annual Research Meeting. June 14-16, Minneapolis, Minnesota, U.S. Scientific and clinician communities 1500 USA and International
116 Poster presentation UNIMAN EvidenceBased Recommendations for the Reduction of Polypharmacy in Older People with Multiple Chronic Conditions: A Set of Systematic Reviews.
9-10/7/2015 Annual Scientific Meeting of the Society for Academic Primary Care. University of Oxford, Oxford, U.K. Scientific community 200 UK
117 poster UWH Tackling polypharmacy in older adults: A compilation of systematic reviews for the development
of an electronic decision support tool
Schlager et al.
WONCA Europe Conference 2015 22/10/2015-25/10/2015 Istanbul, Turkey Scientific community 3680 79 countries, 42 outside of Europe
118 poster UWH Behandlung der Hypertonie bei älteren Menschen
Sommerauer et al.
49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
119 poster UWH Effektivität und Sicherheit in der Behandlung von Schmerz bei älteren Menschen mit Opioiden und Nichtsteroidalen Antirheumatika - von der Evidenz zur Entwicklung von Empfehlun-gen. Schlager et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
120 poster UWH Einsatz der elektronischen Entscheidungshilfe PRIMA-eDS zur Optimierung der Medikation in der hausärztlichen Versorgung - eine qualitative Untersuchung zur praktischen Umsetzung. Rieckert et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
121 poster UWH Neue orale Antikoagulantien bei älteren Patienten. Sommerauer et al. 49th congress for general practice and family medicine of German Association of General Medicine 17/09/2015-19/09/2015 Bozen, Italy Scientific communitiy 600 Germany, Austria, Italy
122 presentation UWH Polypharmacy in Chronic Diseases: Reduction of Inappropriate Medication and Adverse Drug Events in Older Populations by Electronic Decision Support (PRIMA-eDS). Renom et al. The International Association of Gerontology and Geriatrics 23/04/2015 – 26/04/2015 Dublin, Ireland Scientific communitiy
123 presentation UWH Entwicklung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie: Das PRIMA-eDS Tool. Sönnichsen et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
124 poster UWH Effektivität und Patientensicherheit bei Protonenpumpeninhibitoren bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Sommerauer et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
125 poster UWH Effektivität und Sicherheit von Opioiden in der Behandlung von Schmerz und von Husten bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Schlager et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
126 poster UWH Efficacy and safety of drugs commonly used in the management of chronic diseases in older adults: a compilation of systematic reviews for the development of an electronic decision support tool
Al Qur´an et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
127 poster UWH Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von Erkrankungen des Stütz- und Bewegungsapparates bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Reickert et al.
16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
128 poster UWH Über die Einführung einer elektronischen Entscheidungshilfe zur Reduktion von Polypharmazie in der Hausarztpraxis - eine Pilotstudie. Wörster et al. 16th annual meeting of the German Network for Evidence-based Medicine 13/03/2015-14/03/2015 Berlin, Germany Scientific communitiy 400 Germany
129 poster UWH Effektivität und Sicherheit von Antihypertensiva vier verschiedener Substanzklassen (Diuretika, ACE-Hemmer, Kalziumantagonisten, ß-Blocker) bei älteren Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Sommerauer et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
130 poster UWH Effektivität und Sicherheit von Herzglykosiden in der Behandlung von Herzinsuffizienz und Vorhofflimmern bei älteren Menschen - von der Evidenz zur Entwicklung von Empfehlungen
Schlager et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
131 poster UWH Effektivität und Sicherheit von Thrombozytenaggregationshemmern (TAH) bei der Prävention und Behandlung von Herzkreislauferkrankungen älterer Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Meinshausen et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
132 poster UWH Einsatz nichtsteroidaler Antirheumatika (NSAR) in der Behandlung von Erkrankungen des Stütz- und Bewegungsapparates bei älteren Menschen – von der Evidenz zur Entwicklung von Empfehlungen
Rieckert et al.
48th congress for general practice and family medicine of German Association of Family Medicine 18/09/2015 – 20/09/2014 Hamburg,
Germany Scientific communitiy 600 Germany
List of Websites:
www.prima-eds.eu