Skip to main content
European Commission logo
polski polski
CORDIS - Wyniki badań wspieranych przez UE
CORDIS
Zawartość zarchiwizowana w dniu 2024-05-30

The antiretroviral roll out for HIV in India - strengthening capacity to promote adherence and patient follow-up in the context.

Final Report Summary - HIVIND (The antiretroviral roll out for HIV in India - strengthening capacity to promote adherence and patient follow-up in the context.)

Executive Summary:
Mobile phone-based interventions are increasingly harnessed to aid medication adherence in HIV infection. There have been no trials efficacy trials in Asia, though some early evidence has been published from Sub Saharan African settings. Under the EUFP7 HIVIND project, we tested our hypothesis that customized mobile phone reminders would enhance adherence to ART among HIV-infected patients initiating anti-retroviral treatment (ART) in Southern India within the Indian national AIDS control program. The mobile phone intervention was also costed. Opportunistic infections, immune reactivation inflammatory syndrome and adverse drug reactions were also studied in the cohort. The proportion of patients failing first line treatment in the cohort was also reported, mutations were typed. The cost of the mobile phone intervention used in the trial was also reported from a national program perspective. Qualitative studies were done to explore patient perception of receiving such a mobile phone based adherence intervention. We also studied in the trial platform, the use of an ELISA based viral load assay, ExaVir load to compare its performance with PCR assays.

Between July 2010 and Aug 2011, 631 ART-naïve patients were recruited into the HIVIND trial cohort, and randomly assigned into the mobile phone intervention and control arms, and monitored for 96 weeks. The intervention consisted of customized weekly automated voice reminders, and a pictorial message. Primary and secondary outcomes were time to virological failure and pill count ART adherence respectively.
There was no observed difference in time to virological failure between the allocation groups. The rate of suboptimal adherence was similar between both groups. The results of analyses adjusted for potential confounders factors were similar. Other secondary outcomes such as death and attrition rates, and subgroup analysis also showed comparable results across allocation groups.
15.5% of patient in the cohort failed first line treatment, similar proportions in both arms.

The cost of the intervention was low, at scale, the intervention would cost less than 2 USD per patient per year.
The ExaVir Load test performed well against PCR in measuring viral load measures in a cohort subgroup. It showed a strong correlation (r=0.96) and the Bland Altman plot showed good agreement.

In this multicentric randomized controlled trial among ART-naïve patients initiating first-line ART within the Indian national program, we found no significant effect of mobile phone voice reminders on either time to virological failure or medication adherence at the end of two years of therapy. The proportion failing treatment based in the trial definition was 15%. The cost of a mobile phone intervention is low, and could possibly be useful to target adherence in specific subgroups, or to support treatment in ways other than adherence reminders. The ExaVir test load test performed well to measure viral load against PCR.

Project Context and Objectives:

Project Context and main objectives

Introduction: The HIVIND project operates primarily in two HIV high prevalence provinces of India. The project is coordinated from the Division of Global Health, Karolinska Institutet, Stockholm.
This proposal focuses on 2 high prevalence Indian provinces. As the antiretroviral (ART) program is scaled up, adherence is a key issue that needs to be addressed (as it is a key determinant of resistance, which has public health consequences). With limited affordable second-line regimens & restricted laboratory monitoring in low-income settings, optimal adherence to first-line regimens is essential.
The study is a randomized controlled trial of an approach using a contextually relevant intervention (mobile telephones) to influence ART adherence in 600 ART naïve, HIV+ Indian patients eligible for ART, in Karnataka and Tamil Nadu. The conventional existing approach (as in the national guidelines) will be compared with an intervention in which the patient is provided adherence support using a mobile telephone interface. The trial has now been completed. The study besides assessing the effect of intervention on adherence, will also provides data on the proportion of Indian patients failing first line ART, both primary and acquired resistance. A study of factors associated with adherence, in India has been done. In addition the incidence and manifestations of opportunistic infections, immune reconstitution syndrome & adverse drug events have been described. The use of validated low-cost tests that optimize monitoring, are necessary here. Viral load is rarely used to monitor treatment because it is expensive. Instead falling CD4 counts are used. Exavir load, an ELISA based viral load assay has been tested in the cohort.

Project setting:

The basic study design was a randomized control trial of two different approaches to
promoting ART adherence in ART naïve HIV infected Indian patients eligible for ART, in two Indian sites (two provinces), Karnataka and Tamil Nadu. The primary end point was time to the development of virological failure. Secondary end points included adherence.
Karnataka and Tamil Nadu are among India’s high prevalence provinces for HIV prevalence in the general population >1% or >5% among high-risk groups). A similar study site in Vietnam will also be a part of the project, as there is a synergy between these two projects, with learning from the strengths and weaknesses of each site applicable to the other.

Study Areas
Basic socio demographic characteristics of all the two Indian recruiting sites are as given below. The establishment of the clinical platform for the RCT in planned in two sites (in India in the provinces of Karnataka and Tamil Nadu); the Vietnamese site has already begun recruitment. All three sites are high prevalence zones of the HIV epidemic. Karnataka: is India’ ninth largest province with a population of 53 million. It is one of India’s six high HIV/AIDS prevalence states. The province has a population density of 275 per square kilometer. The literacy rate is 66.6%. Karnataka (and south India in general) has a relatively better health status compared to otherIndian provinces. Infant mortality rate stood at 55 per 1000 live births at the time of project commencement. The annual sentinel surveillance system is one of the main sources of data regarding HIV infections in the state. Karnataka is one of the states of India most seriously affected by the HIV/AIDS epidemic. The first case of AIDS in the state was detected in 1988. The state now ranks fifth on the number of reported AIDS cases in the country. The proportion of women in antenatal clinics who test positive for HIV infection is the highest in the country (1.25%), a sign that the epidemic is spreading into the general population and is no longer confined to high-risk groups (infection rate in these groups is 13.6%).
The study site is located in the capital city of Bangalore, at the St. John’s Research
Institute, St. John’s National Academy of Health Sciences, Bangalore.
Tamil Nadu (India): has a population of 62 million, and is India’s most urbanized province. It is also one of the six high prevalence states. HIV in India was first reported from this province. Tamil Nadu has a higher literacy rate of 73.5%. Infant mortality in the province is 49/1000 live births. HIV prevalence in the population ranges from 0.5% (in antenatal clinics) to 9.2% (among high-risk groups). The study will be situated in the YRG Care Research Center, Chennai, India
Both these high HIV prevalence provinces in India have rapidly growing economies and have been at the frontline of the India’s recent economic growth.

Overall Aim
This project aims to study a contextually relevant intervention to promote adherence in ART naïve, HIV+ Indian patients, in a randomized controlled trial design. The intervention, the Mobile phone treatment support design (MTS) will be evaluated against the existing conventional (CT) treatment to assess if failure, adherence and other outcomes are influenced. The CT arm will serve as the control arm. Further, new, simpler and inexpensive ELISA based techniques to assess viral load and resistance will be evaluated in these contexts.

Specific Objectives
a. To create a platform and follow up a study population 600 ART naive, HIV+ Indian patients who are eligible to be initiated on ART. The study population will be randomized into two different arms, MTS and CT. Follow up will include clinical examinations, assessments of adherence, viral load and CD4+ counts at predetermined points in time for a minimum of 24 months or till the development of resistance whichever is earlier.
b. To study the proportion of patients displaying primary resistance in the cohort and to type the mutations in these strains
c. To compare the two arms in terms of proportion of patients and timing (since initiation of ART) of failing treatment (virological failure).
d. To type the resistance (mutations) detected in the cohort.
e. To compare the influence of MTS intervention on adherence to first line ART and to compare then the influence of adherence on virological failure in this setting.
f. To study contextual predictors of failure – socioeconomic, demographic and nutritional
g. To correlate different adherence measures used in the study (3d recall, VAS and mobile telephone based)
h. Further to study quantitatively the association of other factors (socio economic, gender, system related, therapy related, and patient factors) with ART adherence.
i. To explore qualitatively barriers to adherence and the experience of having treatment support via the mobile telephone intervention.
j. To evaluate the ELISA based load tests (ExaVir Load) among Indian patient cohorts. This test will be evaluated against the gold standard (polymerase chain reaction).
k. To describe the incidence and manifestations of opportunistic infections (OI), the incidence and manifestations of immune reconstitution syndrome (IRIS), and manifestations of adverse events
l.To assess the cost effectiveness of two strategies in relation to the outcomes.

Project Results:
Also attached as a pdf attachment as the tables and figures are distorted in this view.


Main ST results from the HIVIND project
Part 1: The main HIVIND trial:
Trial design; This was a multicentric randomized controlled open-label trial conducted in South India. The rationale, study design and conduct of statistical analysis of the HIV-India (HIVIND) study have been published previously. Patients and the randomization team were aware of the intervention assignment; while research staff assessing patients, laboratory staff, statisticians and authors were blind to the allocation. Allocations were revealed only after the blinded results were analyzed and discussed by all authors.
Ethical approvals for the conduct of the trial were obtained from all participating institutions prior to study initiation. In accordance with national requirements and the principles of the Declaration of Helsinki, written informed consent was obtained from all participants prior to enrollment. Confidentiality was maintained at all levels of data management. An independent data safety and monitoring board (DSMB) reviewed the data and performed a pre-specified blinded interim analysis midway through the trial.
Study setting and participants: Patients with documented HIV infection attending the ambulatory clinics from three sites in two HIV high-prevalence states of India were recruited. The two sites in Karnataka State were part of the National AIDS Control Program, and included St. John’s Medical College Hospital, Bangalore, a missionary teaching hospital with an ART center that is run as a public-private partnership, and Krishna Rajendra Hospital, Mysore, a government-run tertiary-care hospital providing free healthcare, which functioned as a satellite recruiting site to St. Johns. The third site, YRGCARE Medical Centre, Chennai is a private non-governmental center in Tamil Nadu State. Together these sites catered to both urban and rural populations in South India, providing multidisciplinary care, counseling and treatment for over 10,000 HIV-infected patients at the time of study initiation.
Eligible patients included HIV-infected individuals with adequate documentation of their HIV-positive status, aged between 18-60 years, ART-naïve and meeting criteria for initiation of first-line ART as per the 2007 Indian national guidelines[17]. We excluded those who tested positive for HIV-2, were severely ill (Karnofsky score <70), expressed inability to attend all study visits, had no mobile network in their area of residence or had another member of their household already recruited into the same study (to minimize intervention ‘contamination’).
Randomization:Participants were randomly assigned in a 1:1 allocation ratio to the mobile phone intervention arm or control arm. Randomization was performed stratified for sex, in permuted blocks of 4 or 6, using sequentially numbered opaque sealed envelopes. All patients were given a mobile phone in order to obviate any potential effect the possession of the phone might have on adherence; but only those in the intervention arm received phone reminders.
The mobile phone intervention: The main aspect of the intervention was a customized motivational voice call that went out once a week at a time selected by each patient. The patient also chose the gender and language of the voice call. This interactive call asked how the patient was feeling, whether medications were taken as prescribed and required the patient to respond to a question about the previous day’s pill doses, by pressing ‘1’ for yes or ‘2’ for no. If the patient failed to respond to the call, a maximum of three more calls were made over the ensuing 24 hours until a response was obtained. The second aspect of the intervention included a weekly non-interactive neutral pictorial message sent out as a reminder 4 days after the automated call.
The fidelity of the intervention was monitored continuously throughout the trial by checking the mobile phone intervention software reports that were available on a daily basis. In addition, selected research staff also received the intervention and maintained a weekly report for quality control.
Standard care: All participants in the control and intervention arms received standard care, based on national guidelines. This included up to three counseling sessions prior to initiation of ART, routine clinical and laboratory tests at baseline, and follow-up assessments every 6 months. First-line ART drugs included zidovudine, stavudine or tenofovir-based ART, along with lamivudine and either nevirapine or efavirenz, and were dispensed free of cost every 1 to 3 months, as generic fixed-dose combination pills.
Study procedures: Participants returned for study visits at weeks 2, 8, and 12 after ART initiation, and subsequently every 12 weeks until week 96, or till the point of virological failure. All study visits were integrated with routine clinic visits as per the national guidelines. At every study visit medical details were documented. A researcher not involved in routine care and who was blinded to the allocation, measured pill count adherence. Laboratory assessments included CD4 cell count and viral load every 3 months, and other tests as clinically indicated. A quality control manager ensured that data entered in the case report forms were complete and accurate. A double data entry system was employed to minimize errors, which was supervised by a dedicated data manager.
Study outcomes and definitions; The primary outcome in the trial was time to virological failure, defined as plasma viral load >400 copies/ml on two consecutive samples measured at least one month apart, six months after initiating ART. The secondary outcome was patient adherence to ART using pill count expressed as a percentage of pills taken, divided by total pills dispensed, after accounting for returned or lost pills. ‘Optimal adherence’ was defined as mean adherence ≥95% during the follow-up period. Attrition and death rates were included as secondary outcomes.
Sample size calculation: To estimate the necessary numbers of patients in each of the two allocation groups we assumed the risk for virological failure in the control arm to be 10%, (risk rate 0.1) during the two years, while the intervention was expected to reduce this risk to 3% (0.03)[18]. A total sample of 532 patients (266 in each arm) would provide 90% power to detect such a risk difference in a two-sided log-rank test with significance level of 0.05. Expecting an attrition rate of 10%, the trial was planned to have a minimum of 600 patients.
Statistical analysis: Trial analysis was performed using an intention-to-treat principle that included all originally randomized patients. Additional per-protocol analysis was performed excluding patients with major protocol deviations (patients missing/withdrawn consent; 24 and 30, in intervention and control arms respectively) Time-to-event analyses that included Kaplan–Meier survival curves, log rank test and Cox proportional hazards models were used to compare intervention and control groups with respect to virological failure. Cox regression analysis was performed adjusting for site and other socio-demographic confounders. The secondary outcome of mean pill count adherence was dichotomized to a binary variable with outcomes <95% and ≥95%. Crude and adjusted Incidence Rate Ratios (IRR, incidence rate of patients with mean adherence ≥95%) between the two arms were calculated using Poisson regression model. Stratified analyses with regard to primary and secondary outcomes were performed on predefined subgroups based on site, age, presence of transmitted drug resistance, ART regimen type and level of adherence. The consistency of the intervention effect among these subgroups was assessed by formal tests of interaction. Stata software v13 software (Statacorp) was used for all analyses. A p value of <0.05 was considered to denote statistical significance, and all tests were two-sided.
RESULTS
Patients:Between July 2010 and Aug 2011, 1,140 HIV-infected ART-naïve individuals were screened and 509 ineligible participants excluded. Among the enrolled participants, 315 and 316 were randomized to the intervention and standard care arms respectively. Both groups had similar pre-randomization characteristics (Table 1).
Table 1: Baseline social, demographic and clinical characteristics of patients before randomization
Characteristic Mobile phone intervention arm
Standard care arm (control)
Social and demographic characteristics, n (%)
Sex
Females
136 (43.2%)
137 (43.4%)
Age
18-30 yrs
31-40 yrs
41-60 yrs
76 (24.1%)
150 (47.6%)
89 (28.3%)
79 (25.0%)
156 (49.4%)
81 (25.6%)
Education
No formal education
Up to high school
Beyond high school
59 (18.7 %)
189 60.0%)
67 (21.3%)
57 (18.0%)
190 (60.1%)
69 (21.8%)
Literacy
Can read/write in local language
(Kannada, Tamil, Telugu, Hindi)

252 (80.0%)

250 (79.1%)
Residence (urban vs. rural)
Rural
143 (45.4%)
143 (45.3%)
Currently married 217 (68.9%) 218 (69.0%)
Employment
Unemployed/Household duties

116 (36.8%)
117 (37.0%)

Previous experience with mobile telephone usage

263 (83.5%)

260 (82.3%)
Household income
(≤$1000 per year)
229 (72.7%)

237 (75.0%)

Recruiting sites
Bangalore
Mysore
Chennai
81 (25.7%)
81 (25.7%)
153 (48.6%)
77 (24.4%)
83 (26.3%)
156 (49.4%)
Clinical characteristics, n (%)
WHO* clinical staging
3 and 4
175 (55.6%)
170 (53.8%)
CD4 count, n (%)
<250 cells/mm3
230 (73.0%)
217 (68.7%)
Median CD4, cells/mm3 (IQR)* 185 (97, 253) 193 (115, 268)
Baseline viral load, log10 copies/ml (IQR)
5.5 (5.1 6.0)
5.4 (4.9 5.9)
Regimen (305 in Intervention arm, and 308 in Control arm)#
Zidovudine-based
Stavudine-based
Tenofovir-based

136 (44.6%)
34 (11.2%)
135 (44.1%)

133 (43.2%)
38 (12.3%)
137 (44.5%)
Baseline transmitted drug resistance
13/309 (4.2%)
12/308 (3.9%)
* WHO denotes World Health Organization, IQR interquartile range)
# Patients who died or transferred out/withdrew consent: 10 in Intervention arm and 8 in Control arm.

Study outcomes: Primary outcome: There was no observed statistically significant difference in the time to virological failure between the allocation groups (unadjusted hazard ratio (HR) 0.98 95% CI 0.67 to 1.47 p=0.95) (Figure 1). Adjustment in a Cox proportional hazards model for covariates such as site, age, sex, rural residence, education level, previous experience with mobile phones, baseline CD4 count, ART regimen, occurrence of adverse drug reactions and presence of transmitted drug resistance at baseline, did not reveal any significant difference between the intervention and standard care arms (adjusted HR 0.96 95%CI 0.65 to 1.43 p=0.85) (Table 2). A similar lack of significant difference was seen in the per-protocol analysis (adjusted HR 0.96 95%CI 0.64 to 1.43 p=0.89). The rate of virological failure in the intervention and standard care groups were 10.52 (95% CI 8.11 to 14.19) and 10.73 (95% CI 7.95 to 13.92) per 100 person-years respectively. Failure rates across allocation groups remained similar at 6, 12 and 18 months after ART initiation.
Secondary outcomes: Suboptimal adherence rates did not differ significantly between the intervention and standard care groups (crude IRR 1.24 95% CI 0.93 to 1.65 p=0.14). When adjusted for covariates in a Poisson model, no additional change was seen (adjusted IRR 1.24 95% CI 0.94 to 1.63 p=0.13). Comparable results were obtained in the per-protocol analysis (adjusted IRR 1.26 95% CI 0.94 to 1.69 p=0.13). The incidence proportion of patients with suboptimal adherence was 27.0% in the intervention arm and 21.7% in the standard care arm. Both allocation groups had uniform proportions of patients with suboptimal adherence over the course of the trial at 6, 12 and 18 months. Death rates showed that 4.51 (95% CI 2.94 to 6.91) and 5.04 (95% CI 3.35 to 7.58) deaths per 100 person-years occurred in the intervention and standard care arms respectively. Attrition occurred at the rate of 3.43 (95% CI 2.10 to 5.61) and 4.82 (95% CI 3.17 to 7.31) losses to follow-up per 100 person-years in the intervention and standard care arms respectively. Comparison of death rate (adjusted IRR 0.91 95%CI 0.51-1.60 p=0.74) and attrition rate (adjusted IRR 0.59 95 %CI 0.32-1.10 p=0.10) did not differ significantly between allocation groups.
Figure 1: Cumulative hazard rate of virological failure among patients receiving the intervention and standard care



Subgroup analysis: Significant differences in virological failure or adherence rates between allocation groups were consistently absent across subgroups of site, age, level of baseline immunosuppression, ART regimen, presence of transmitted drug resistance and level of mean adherence (Table 3). There were site differences in mean adherence levels (proportion optimally adherent in Bangalore 89%; Mysore 67%; Chennai 73%, p<0.001) but the rate of virological failure was similar across the intervention and control arms at each site.

Part 2: The proportion failing treatment and mutation sequences:
15.5% of patients in the cohort failed first line treatment over the follow up period. Drug resistance mutation profile at baseline and failure. There were total 98 failures in HIVIND trial cohort. The time points and viral loads at failure are given in the periodic report.
Among these 98, six patients had viral load <1000 copies/mL during viral failure confirmatory visit, so viral genetic material was not amplified from these six samples. Among the 92 patients, 36 (39%) did not have any mutations at failure. Rest 56 patients’ mutation profile has been presented. DRM was evaluated using International AIDS Society list, update 2013 1.
Part 3: The relationship between adherence and virolgocal failure
Patients with suboptimal adherence (mean adherence <95%) were significantly more likely to experience virological failure after adjusting for other covariates (adjusted HR 0.27 95%CI 0.18-0.41 p<0.001). The probability of virological failure increased non-linearly with decreasing adherence (Figure 1). In subgroup analysis, adherence significantly impacted virological failure at all levels of education, BMI and ART regimen (described in detail in the periodic report). However in the subgroup of patients >40 years of age, adherence did not have a significant impact on virological failure (HR 0.52 95% CI 0.17-1.63 p=0.263). (table 1S)
Figure 1: Relationship between past month adherence and viral failure

Table 1S: Adherence and failure: effect modifiers with regimen, education level, age, BMI
HR 95% CI P Interaction term
Regimen
ZDV/Stav
Tenofovir
0.26
0.17
0.16 – 0.43
0.08 – 0.34
<0.001
<0.001

0.313
Age
<40
≥40
0.55
0.52
0.32 – 0.96
0.17 – 1.63
0.036
0.263

0.123
Education level
competed high school
0.23
0.19
0.15 – 0.34
0.12 – 0.31
<0.001
<0.001

0.132
BMI
<18.5
≥18.5
0.23
0.25
0.16 – 0.35
0.15 – 0.42
<0.001
<0.001

0.725
Patients with adherence <95% are more likely to fail at all levels of regimen, education and BMI and among those <40 years. However among patients >40 years, adherence does not appear to play a significant role in viral failure.
Part 4: Adherence and its predictors in the HIVIND cohort:
Patient characteristics
A total of 10,304 patients were screened at the three study sites between July 2010 and August 2011, and 631 eligible ART-naïve patients enrolled in the study after providing written informed consent. Patients were excluded if a minimum of 1 month follow-up did not occur; between obtaining consent and one month after enrolment, 7 patients died, 8 were missing and 17 were transferred to a different medical center. The 599 included patients contributed a total of 921 person-years of observation time. Women constituted 42.7% of the patients; men and women were on average 39 (±8) and 34 (±7) years of age at the start of ART, respectively. Mean CD4 count prior to initiating ART was 192 cells/mm3 (±109). Socio-demographic and clinical characteristics of all patients, and those with mean adherence <95%, classified as “sub-optimally adherent patients” are shown in Table 1.
Table 1. Baseline characteristics of all patients and those with suboptimal adherence (mean adherence <95%)
All patients


n = 599 (%) Patients with mean adherence <95%
n=146 (%)
Age
<40 years
≥40 years
437 (72.9)
162 (27.1)
118 (80.8)
28 (19.2)
Sex
Male
Female
343 (57.3)
256 (42.7)
76 (52.1)
70 (47.9)
Education
Less than high school
Completed high school
110 (18.3)
489 (81.7)
38 (26.0)
108 (74.0)
Residence
Urban
Rural
334 (55.8)
265 (44.2)
76(52.1)
70 (47.9)
Disclosure to family or friends
No
Yes
38 (6.3)
561 (93.7)
18 (12.3)
128 (87.7)
Body mass index
< 18.5
≥ 18.5
189 (31.5)
410 (68.5)
56 (38.4)
90 (61.6)
Baseline tuberculosis co-infection
No
Yes
400 (66.8)
199 (33.2)
96 (65.7)
50 (34.3)
Baseline hemoglobin
< 10g/dl
≥ 10g/dl
90 (15.1)
508 (84.9)
25 (17.2)
120 (82.8)
Baseline CD4 count
<100 cells/mm3
≥ 100 cells/mm3
139 (23.2)
460 (76.8)
25 (17.1)
121 (82.9)
ART Regimen
Zidovudine-based
Stavudine-based
Tenofovir-based
263 (43.9)
68 (11.4)
268 (44.7)
67 (45.9)
11 (7.5)
68 (46.6)

Adherence patterns and trends
Overall mean adherence among all patients was 95.4% (SD 9.4). The proportion of patients optimally adherent (mean adherence during study duration ≥95%) was 75.6%. The pattern of adherence among subclasses of patients who experienced virological failure, died, transferred to a different center or were missing is shown in Table 2. The proportion of patients who experienced at least one period of suboptimal adherence during the 2-year observation period was 56.6% (339/599).
Table 2. Adherence among 599 patients observed for 96 weeks after initiating ART
Patient status No. of patients ART exposure time (months) Proportion of patients with adherence ≥95% Mean adherence (% and SD)
Completed 2 years of follow-up 435 24 83.0 97.3 (3.4)
Virological failure 98 9.8 44.9 88.3 (18.0)
Lost to follow-up 26 10.5 57.7 90.1 (12.2)
Died
35 7.7 88.6 96.2 (9.6)
Moved to another center 5 3.8 40.0 85.6 (21.2)
Adherence levels remained high for the majority of patients in this cohort. Over the 2-year period of follow-up, mean adherence increased by 2 percentage points and peaked at 97.8% at week 72, then decreased marginally to 96.1% at the end of 96 weeks. Adherence varied significantly over time (p<0.001). A similar trend was seen when the proportion of patients who were optimally adherent at each time point was considered
Determinants and barriers of optimal adherence
Predictors of optimal adherence in bivariate and multivariate analysis included age 40 years and older (IRR 1.01 95%CI 1.01-1.02) level of education of high school and beyond (IRR 1.02 95%CI 1.01-1.03) drug toxicity-related ART interruption (IRR 0.96 95%CI, 0.95-0.98); greater disclosure status (IRR 1.02 95%CI 1.01-1.03) and sense of satisfaction with one’s own health (IRR 1.01 95%CI 1.00-1.02) felt need for medications to function in daily life (IRR 1.01 95%CI 1.01-1.02) and patient perception of having good access to health-care services (IRR 1.04 95%CI 1.03-1.05) (details in the periodic report). Disease-related factors were not significantly associated with good adherence. Although treatment-related factors such as regimen type and change in regimen were not associated with adherence levels, health provider or patient-initiated interruption of ART due to severe toxicity was significantly associated with zidovudine-based (9.1% interrupted ART) and stavudine-based ART (14.7%) compared to tenofovir-based ART (0.4%) (p<0.001).
There were 194 patients within the first 6 months and 121 patients between 6 and 24 months who reported no difficulty in adhering to ART. The itinerant patterns of reported barriers are well illustrated in Figure 3. ART interruption due to perceived or experienced drug toxicity were prominent early barriers to adherence (14% reported in week 24, versus 2% at weeks 48-96). Forgetfulness (17% versus 30%) and pills not being available (because the patient was away from home at medication time or patient ran out of pills, 29% versus 39%) were more common barriers beyond the first year of ART initiation. Other common adherence impediments were dosing confusion, stigma-related concerns and depression with or without substance abuse.
Adherence and virological failure
A strong inverse relationship between optimal adherence and virological failure was consistently seen in this patient population (IRR 0.55 95%CI 0.44-0.69 p<0.001). As the threshold for defining poor adherence was lowered, a greater proportion of those with poor adherence experienced virological failure. Probability of virological failure was 37% when the optimal adherence threshold was 95%, compared to higher virological failure probabilities of 51.7% and 65.5% when corresponding adherence thresholds were 90% and 80% respectively (Table 4).
Table 4. Probability of virological failure at different thresholds of optimal adherence
Adherence threshold >95% Virological failure
Yes No p
Non-adherent 54 (37.0%) 92 (63.0%)
Adherent 44 (9.7%) 409 (90.3%) <0.001

Adherence threshold >90%
Non-adherent 30 (51.7%) 28 (48.3%)
Adherent 68 (12.6%) 473 (87.4%) <0.001

Adherence threshold >80%
Non-adherent 19 (65.5%) 10 (34.5%)
Adherent 79 (13.9%) 491 (86.1%) <0.001

Part 5- Correlation between different adherence measures used in the HIVIND study
Three indicators of treatment adherence were considered in the study:
Pill count (PC). The pill count measurement is the percentage of pills taken over total pills dispensed accounting for returned or lost pills. Sub-optimal adherence is defined as the percentage below 95%. The percentage of reported suboptimal adherence using this definition was 86%. The percentage of reported 100% adherence was 60%.
Visual analogous scale (VAS). The persons are asked to mark their perceived adherence on a scale from zero to one hundred percent. Sub-optimal adherence is defined as the percentage below 95%. The percentage of reported suboptimal adherence using this definition was 91%. The percentage of reported 100% adherence was 84%.
Four day recall (FDR). Persons are asked about their adherence to during the latest four days. Sub-optimal adherence is defined as the percentage below 95. The percentage of reported suboptimal adherence using this definition was 96%. The percentage of reported 100% adherence was 96% as well.
The correlations between the indicators can be described in different ways. Traditional measures like Pearson and Spearman correlation coefficients are hardly useful because there are large percentages of observations reporting 100% for all three indicators. A way to illustrate the relations between the indicators is to dichotomize the indicator outcomes to 100% or less than 100% and cross-tabulate the indicators pairwise. The results are shown below.

Table of VAS against PC
PC
VAS Less than 100% 100% Total
Less than 100% 15.7 0.8 16.5
100% 24.3 59.2 83.5
Total 40.0 60.0 100

Table of FD against VAS, percent
VAS
FDR Less than 100% 100% Total
Less than 100% 3.2 0.5 3.7
100% 13.3 83.0 96.3
Total 16.5 83.5 100

Table of PC against PDR, percent
FDR
PC Less than 100% 100% Total
Less than 100% 3.4 36.4 40.0
100% 0.3 59.8 60.0
Total 3.7 96.3 100

The percentages of reporting 100% are 60%, 84% and 96% for PC, VAS and FDR respectively. Since the off-diagonal frequencies are extremely unbalanced we can say that comparing VAS and PC means that VAS has roughly 84-60=24 percent units more 100% observations. The corresponding comparison between FDR and VAS is 96-84=12 percent units.

Part 6: Report on Opportunistic Infections, Immune reconstitution inflammatory syndrome and ADRs.
There were 313 and 216 events of OI and IRIS among 148 and 153 patients respectively in the HIVIND cohort
Opportunistic infections
Opportunistic Infection Number of events % of all OIs
Diarrhoea 61 20
Candidiasis 58 18.5
Skin infections/ ulcer 51 16
Tuberculosis 35 12
HERPES 26 8.3
Others (Including PCP pneumonia, cryptococcal meningitis, CMV, other infections) 82 25
OIs resulted in the death of 15 patients. Deaths were from severe cryptosporidial diarrhea (6), tuberculosis (4), cryptococcal meningitis (2) and other infections (2).
Immune reconstitution inflammatory syndrome
IRIS Number of events % of all IRIS
Candidiasis 68 31.5
Tuberculosis 38 18
HERPES 19 8.8
PCP Pneumonia 5 2.4
Skin lesions (including tinea and ulcerative lesions) 59 27
Others 27 12.5
IRIS resulted in the death of 6 patients; 5 from Tuberculosis and 1 from PCP pneumonia.

Adverse drug reactions:90.0% patients experienced at least 1 adverse reaction, and 85 (26.5%) experienced at least 1 severe reaction. The incidence rate was 52 and 15 per 100 person-years for all reactions and severe reactions respectively. The cumulative incidence of zidovudine-induced anemia was 37.1% over 2 years. At 12 and 24 months, the proportion of patients with optimal adherence, undetectable viral load and CD4 counts >350 cells/mm3 were similar between patients who experienced or did not experience severe adverse drug reactions.

Part 7: Qualitative study on patients perceptions of the mobile phone intervention

Qualitative results:
The thematic framework analysis of indepth interviews revealed mixed perceptions regarding the reminders. The IVR calls were preferred over the pictoral non-interactive SMS reminders. However participants suggested that reminders were best sent to those who needed them. Efforts were made to mitigate perceived stigma resulting from disclosure of HIV status if the reminders were accidently intercepted by others. Stigma was also identified as a barrier to the use of mobile phone reminders for ART adherence. Participants did not perceive the reminders as intrusive as they received each component of the reminder once a week.

Mixed perceptions regarding usefulness of the intervention:
Some participants considered the intervention useful while others did not. The intervention served as a reminder despite its biweekly frequency. In addition it was perceived to reflect the concern of the healthcare provider and support the participant. However those capable independently ensuring adequate adherence to medication did not consider the intervention useful.
IVR calls were considered to act as reminders throughout the day and the week despite its weekly frequency. Participants reported that the calls minimized forgetfulness, especially when participants were busy with work or were away from home.
“We will be busy with our work, when we get busy, I feel that the reminder is very important for me take the tablets like this, for my health.”
“Even if the call comes in the afternoon it is helpful, to take the tablet in the evening. When I receive the phone call in the afternoon I will remember the mornings timings in the afternoon.”
The phone calls provided a cue to take medications, encouraging the development of a routine for some. For others, the calls encouraged adherence by making patients contentious of their adherence reports.
Yes it helped me. As soon as I used to get the call, I used to remember that they will ask whether I’ve taken my medication. So that way it was helpful…. Fear, that they’ll ask me about not taking medicines everyday, so that kept me going
Perceived concern of the healthcare provider:
The intervention made the participants feel that they were cared for. Participants also reported the concern and support they felt from the healthcare provider as a result of the phone calls.
“I feel happy that hospital people care for my health when I receive the call”
The participants felt happy to receive the calls. The calls were perceived as being from a friend and not from a machine despite all participants being aware that the calls were automated.
“When I get these computerized phone calls asking me how my health is?? I feel contented… Even if I am ready to pay in lakhs, I don’t think I will get a privilege like this…”
Self-awareness and the need for reminders:
On the other hand those participants who considered themselves adequately adherent to medication did not find the intervention useful.
“…Since I don’t make mistakes, I don’t think I need phone calls. If I had made mistakes, I would feel receiving phone calls would be good…”
Preference for Calls over Messages:
The IVR calls were preferred to the SMS reminder. The calls were preferred due to their interactiveness and technological simplicity. There was a passive acceptance of the intervention however the need for a more personal and bidirectional communication was felt. The anonymity offered by the automated calls was appreciated given that the disease is amenable to stigma with disclosure of information.
Preference:
Participants perceived calls as being extremely important while the SMS was not considered to be of any additional help over phone call. The interacativeness of the phone call was appreciated over the passivity of the SMS.
The call is sufficient, SMS is not necessary, I don’t want the sms… The phone call is better by 75%. If you compare in the phone call they speak, at least to respect what they speak we pick the call at least once. If we pick the call we have to respond whether we have taken the medicines or not. We do not have to respond to the SMSs, the SMS is not necessary
Technological issues:
Many reported technical difficulty with accessing the SMS despite being trained to do so. Participants also perceived technology a barrier to the use of the SMS function by rural residents. One participant reported that neither he nor his 10th standard educated spouse (wife) knew how to view or send an SMS.
Phone call is better, SMS is not required. … for people who are educated SMS will be useful, not for who has a lesser educational qualification like me…”
Perception of the risk of stigma from the intervention:
Participants feared disclosure of their illness and stigma arising thereof as a result of the intervention. Personalising the time the call ensured that the call was attended to only by the participant. The extra effort to ensure that a daily reminder was not attended to by others made weekly reminders preferable.
Participants rarely left the phone unattended on the day of the call out of fear that the call may be attended to by others. Personalizing the time and day of the call enabled participants to choose timings that afforded a certain level of privacy while answering the call.
“No one has picked up the call, Wednesday the phone call comes and Sunday the SMS on these two days I will never leave the phone anywhere, even if there is a problem the phone will be kept with us.”
They also considered that more frequent calls, e.g. daily calls, had a greater risk of leading to disclosure of HIV status, as the chances of leaving the phone unattended would increase.
“I do not think it is good to get the phone call every day because we will not know where we will be at the time and we cannot say the phone will be with us all the time…”
Instances where calls were attended by friends or family resulting in inquiries regarding the calls to participants were reported. When subjected to these inquiries some participants reported calls as either advertisements or calls from mobile network providers. Others reported calls as adherence reminders for chronic disease like diabetes from their healthcare providers.
“.. people around me wanted to know from where the call comes from. I tell them this call is from Aircel/mobile company and will try to escape from that situation.”
“Once or twice daughter picked up the call and asked asked my what should do,? I told “I am taking diabetic medicine press1”. ..why should I tell, why should I disturb them, I haven’t seen the virus, why should I disturb them , that is my philosophy…”
Participants also report that the kept the phone to “their ears only” when in presence of others. One participant reported going out of the house to attend the call, another personalised the call to receive it after the children had left the house.
While most participants did not fear disclosure of HIV status with the SMS, fear of disclosure resulted from friends viewing all their SMSs out of curiosity. Participants chose to delete the SMSs from their inbox to prevent others from viewing them. Two participants perceived the pictoral SMS to be the HIV symbol, one thought that the symbol was traceable to the HIV care center through the internet, the other participant feared that an HIV infected patient receiving the same message browsed his SMSs his HIV status would be disclosed.
…See if you go the website you can find out which hospital it represents, the pictorial diagram can be fond on the internet. It is something with flower-pot, doctor, the hospital logo...
Most patients in the study report having disclosed their disease to immediate family and the people they live with while none disclosed their HIV status to their friends of employers. Stigma from disclosure of their HIV status was the primary reason why participants considered the intervention a potential threat in their daily lives.
Targeted reminders:
Individualizing the frequency of reminders based on patients need was thought necessary based on the levels of adherence and social support available to the patient.
Participants suggested sending reminders poorly adherent patients identified at monthly pill refills.
“…for those irregular with medications you should make the calls every day, those who are regular weekly once is enough.”
“Yes, we can ask them and choice can be given. Some people will say that this should be sent to them every day, they will forget, they may go out, you can send this (reminder) to those who prefer them (reminder) everyday. Some people will say that this (daily reminder) is difficult and it (reminder) can be sent once in a week, for such people this (weekly reminder) would be appropriate and can be sent once a week…”
Patients who resided in rural areas, those with poor social support or those busy with work and hence preoccupied were considered in need of reminders. Patients with illness that affected the memory were also thought to need reminders.
“For people who forget things, this needs to be sent on a daily basis”
Intrusiveness of the intervention:
Most participants did not consider the weekly reminders intrusive. A passive acceptance of the frequency of the reminders was noted. Participants were willing to accept whatever the healthcare facility provided them with even if it meant receiving the calls daily. This attitude enabled participants to put up with technological issues with the reminders e.g. frequent calls on a single day despite responding appropriately or delay in call timings. Participants either switched off the phone or silenced it to prevent persistent calls from attracting the attention of others in the vicinity.
I have not felt irritated, with calls from the hospital or doctors, it may be a bit late sometimes when it is a hospital issue, because the hospital is there for us, for the sake of the hospital if we reserve one day there is no problem
Barriers to responding to the IVRs:
Participants were unable to attend to all the calls. While stigma was one of the barriers to attending calls in public places, technical
Inability to respond:
Not all respondents attended the IVR calls on every occasion. Reasons for not attending to the calls were receiving calls in public places, technical issues such as the phone being switched off or unfamiliarity with the technology when the patients newly received the reminders.
“Sometimes, when I was in temple, I could not take the call, I attend the call during the second or third attempt, even when I missed the two attempts, they will call me on the next day, on Monday”
“Sometimes I would miss the call if I was doing some work. Nobody else in my house attends the call. They will only tell me that I had a call. So, I have missed the calls.”
Part 8: Costing study
Adherence to antiretroviral treatment is critical to maintaining health and good clinical outcomes in HIV infected patients. To address poor treatment adherence, low cost interventions like reminders using mobile communication technology are being tried. While there have a been a number of studies, showing an effect of mobile phone reminders on adherence to ART, there have been no studies reporting the costs of such reminders for a national AIDS program. The costs of mobile phone reminders to support adherence in the context of India’s national AIDS control program (NACP) is reported based on data drawn from the HIVIND trial.
The study was done at two tertiary level teaching hospitals, which implement the national program, in Karnataka province, South India. Costs for a mobile phone reminder application to support adherence, implemented at these sites, i.e. weekly calls, messages or both; were studied based on expenses in the HIVIND trial cohort. Costs were collected based on the concept of avoidable costs specific to the application. Costs assessed were one-time costs and recurrent costs that included fixed and variable costs. Sequential procedure for costing was used. Costs were calculated at national program level, individual ART center level and individual beneficiary level from NACP perspective. Costs assessed were pooled to obtain cost/patient/year. Type of application, number of ART centers and number of patients on ART, were varied in a sensitivity analysis of costs.
The Indian NACP would incur a cost of between 80 and 110 INR (USD 1.29 - 1.77) per patient per year, based on the type of reminder, the number patients on ART and the number functioning ART centers. The total program costs for a scale-up of phone reminders to reach the one million patients, expected to be on treatment by 2017 is estimated to be 0.19% of the total 5 year national program budget.
Cost of mobile phone reminders for ART adherence support in the context of the Indian national program is low, facilitated by low costs of mobile communication in the country. Extending the use of mobile communication applications beyond adherence support under the national program could be done relatively inexpensively.
Part 9: Assessing ExaVir load against PCR in the cohort
The cost and infrastructural limitations of the viral load monitoring in resource limited settings have led to CD4 cell count monitoring as a standard of care, which may be a poor predictor of viral load suppression. In such regions viral load quantification using simple and cost-effective assays by measuring RT-enzyme activity can be an attractive alternative. Thus we aimed to evaluate the performance and cost of Exavir load assay v3 in comparison to Abbott m2000rt real-time PCR assays in a longitudinal cohort of 327 patients before (WK00) and 4 weeks (WK04) after initiation of first line therapy. Plasma viral load was determined by Abbott m2000rt and Exavir load v3 in 629 samples (302 paired samples) and then compared. Over all an excellent correlation of r=0.96 was observed, with a good correlation at WK00 (r = 0.84) and at WK04 (r = 0.77). The Bland-Altman plot for all the samples showed good level of agreement with a mean difference (bias) of 0.22 log10copies/mL. Mean viral load measurements at WK00 differed by 0.18 + 0.36 log10copies/mL, while WK04 differed by 0.26 + 0.33 log10copies/mL. The per test cost of the plasma viral load by Abbott m2000rt and Exavir load assay was $36.4 and $17.3 respectively. The excellent correlation between ExaVir Load v3 and Abbott m2000rt real time PCR assayin our setting suggests that that the implementation of Exavir load assay can be an affordable alternative option for monitoring patients on anti-retroviral therapy in large programs in resource limited settings.
Plasma Viral load by Abbottm2000rt Real time PCR and Exavir Loadv3 RT enzyme activity assay:
Plasma viral load was obtained from 629 samples collected from 327 HIV-1 infected patients, of which 302 were paired (before ART and 4 Weeks after ART initiation) and 25 samples were from single visit by antiretroviral naïve patients. 92.2% of the samples (n = 580) were above the lower detectable range of 2.17log10copies/mL (150 copies/mL) by Abbott m2000rt and 85.7% of the samples (n = 539) were above the lower detectable range of 2.3 log10copies/mL (200 copies/mL) by Exavir Load v3 assay.
The overall mean viral load for Abbott m2000rt real time PCR and Exavir load v3 assay was 4.19+ 1.3 log10copies/mL and 3.98 + 1.3 log10copies/mL respectively, which did not differ by more than 0.21 log10copies/mL. Mean viral load for Abbott m2000rt real time PCR and Exavir load assay before and after ART initiation showed the following values: baseline, Abbott m2000rt: 5.33+0.5 log10copies/mL; Exavir Load assay: 5.07+0.6 log10copies/mL; and at 4 weeks of therapy, Abbott m2000rt: 2.97+0.6 log10copies/mL; Exavir Load assay: 2.79+0.5 log10copies/mL) did not differ for more than 0.26 log10copies/mL and 0.18 log10copies/mL respectively, which are well within the a clinically acceptable difference of 0.5 log10copies/mL.
Agreement between Abbott m2000rt Real time PCR assay andExavir Load RT-enzyme activity assay:
As seen in Table 1, there were 54 samples (8.5%), which were quantifiable by Abbott m2000rt assay but were below the detection limit of the Exavir load v3 assay. The sensitivity and specificity at a lower detection limit of 200copies/mL were observed to be 90.7% and 85.6% respectively. Both the sensitivity and specificity increased with an increased lower detection limit cut off. Over all there was acceptable agreement observed between Abbott m2000rt and Exavir Load assay, with excellent agreement observed at higher values of plasma viral load > 3.0 log10copies/mL (Kappa = 0.76).
Table 1. Agreement between Exavir Load assay and Abbott m2000rt at different plasma viral load levels.

PVL by Abbott m2000rt
In copies/mL (log10 copies/mL) Agreement Kappa Value Total samples detected by Abbott Percentage of Sample detected by Exavir Load assay
> 200 (2.3) 89.1 0.46 580 90.7
> 400 (2.6) 88.1 0.57 550 93.5
> 1000 (3.0) 89.7 0.76 458 97.6
> 5000 (3.7) 94.4 0.89 344 99.7
> 10,000 (4.0) 96.8 0.94 324 100

Correlation between Abbott m2000rt Real time PCR assay andExavir Loadv3 RT-enzyme activity assay:
Overall there was a positive correlation observed between the plasma viral load values by Abbott m2000rt and Exavir load assay (r = 0.96). A good correlation was noted in ART-naïve samples (r = 0.84) as well as in samples at Week 04 of ART (r = 0.77).
Level of Agreement between Abbott m2000rt Real time PCR assay and Exavir Load v3 RT-enzyme activity assay:
The Bland-Altman plot for all the samples showed good level of agreement with a mean difference (bias) of 0.22 log10copies/mL, with acceptable limits of agreement (-0.45 and +0.89 log10copies/mL). A good level of agreement was also observed separately at baseline [mean difference bias of 0.25; range of acceptable limit of agreement: -0.39 and +0.89 log10copies/mL] and at WK04 [mean difference bias of 0.19; range of acceptable limit of agreement: -0.52 and +0.89 log10copies/mL].
The test results overall showed good agreement with PCR based tests, at different levels of viral load in plasma.

Potential Impact:
This project will have an impact on several levels in relation to ART treatment strategies for low income - low prevalence settings. The design using two randomized groups (MTS and CT) will specifically address the issue mechanisms to promote ART adherence in these two Asian contexts at a time when both India and Vietnam are scaling up their national access to ART programs. (India is doing this through the government National AIDS Control Organization and Vietnam through the PEPFAR scheme). Adherence and ART resistance development are known to be inter-related.
Monitoring of HIV using simpler ELISA based tests could possibly make it practical and affordable to do regular assessment of patients in low-income settings.
On the level of overall public health, the critical role of adherence in the treatment of HIV has been demonstrated in clinical trials and clinical care settings. Non-adherence to ART has been noted as one of the greatest public health challenges associated with the management of HIV/AIDS. Poor adherence engenders public health concern given the potential for the rapid development of medication-resistant strains of HIV. This may have serious consequences due to potential transmission of drug-resistant HIV by non-adherent patients.
The results of this rigorously conducted trial have been negative, which is an interesting finding, the potential for the rapid development of medication-resistant strains of HIV. This may have serious consequences due to potential transmission of drug-resistant HIV by non-adherent patients.
The results of this rigorously conducted trial have been negative, which is an interesting finding, given that there have been some positive trials (but also some negative ones) from countries in Sub Saharan Africa. The WHO has recently come up with a recommendation on the use of mhealth technologies in supporting patients on ARV. Our trial shows that in India, this strategy did not work, and so the WHO recommendations need to be interpreted here with some thoughtfulness. The trial in India did not show any effect on either adherence or time to treatment failure.
We explore adherence barriers, to see how ART could be made more sustainable. HIVIND is also an innovative project in terms of its research design and that there has been practically no information on this research area (ART and adherence interventions) from India (or Asia). The Cochrane database of systematic reviews recently reviewed studies on patient support and education for promoting adherence to ART. The 10-year review (1996-2005) included only studies, which used a RCT to assess the intervention and measured adherence at a minimum of 6 weeks. Only 19 studies met these inclusion criteria, not one of these had been conducted in Asia. The HIVIND trial has filled a gap in knowledge here.
With regard to monitoring, conventionally, immunological failure in measured in these setting as opposed to virological failure for reasons of cost of measuring load more routinely. The hypothesis is that using a cheaper load test to may facilitate earlier detection of failure than waiting for immunological failure to appear (CD4 decrease). It is between virological and immunological failure, there is multiplication of resistant virus in the patient. This has implications for individual and public health. If the routine use of a cheaper load test that is cost effective, can facilitate the detection of failure earlier (when viral load and resistance are lower), it will allow second line drugs to introduced earlier and prevent wild type virus from being replaced by resistant virus. With limited affordable second-line regimens in India and limited laboratory monitoring, optimal adherence to appropriate first-line regimens is an essential strategy to ensure treatment success in India and in other resource-limited areas. The testing of the low cost Exavir load in this setting showed good correlation with PCR and is a potential test that could be utilized in the program once the decision to introduce viral load monitoring routinely is made.
On the level of the health system and policy, contextual interventions to promote adherence need to be evolved taking into consideration the pluralistic health systems in the two Asian settings, and the strengths available in the health system and society. The pluralism of the health system is represented in the diversity of the Asian partner institutions - one partner being a fully governmental body, the second partner being a Christian teaching hospital within the framework of a government university and the third being a secular non-governmental research and care organization feeding into state policy. Adherence issues in the context of the impending scale-up of ART is problematic especially in under-resourced health systems where Ministries of Health are still adapting to becoming policy rather than service ministries. All settings scaling-up ART access will need to learn as they go, but it is important to create opportunities for such learnings, and to disseminate them promptly and effectively.
Getting research into policy and practice is a major goal for our project and we see a window of opportunity in the beginning of the scaling-up process in the Asian settings to study ART and adherence in context of the overall health system and provide evidence to adjust ART programmes in order to maximize their beneficial impacts. HIVIND will provide examples from different kinds of health systems in Asia, that are highly relevant to national policy and program makers in the Asian region as well as to global actors such as EC, bilateral European donor agencies, GFATM, PEPFAR and WHO involved in financial assistance and global policy making.

The main dissemination activities and the exploitation of results:
The dissemination of the results has followed two tracks: academic and policy maker (government) dissemination.
The academic dissemination has been achieved by way of a number of scientific conference presentations coming out of the HIVIND project. A number of publications have also resulted, manuscripts of the findings of period 4 have been prepared and are currently under review. A list of conferences at which HIVIND project outcomes have been presented is below, as a list of publications.
There have also been a number of meetings with policy makers in the state AIDS control program, the national AIDS control program and with members of the Indian Council for Medical Research throughout the life of the project, but particularly towards the end of the project. Dissemination of findings were carried out at the end of this period to policy makers and providers at regional and national level in India. The European partners together with the Indian partners presented the findings of the HIVIND project at a meeting involving representatives of the National AIDS Control Organistion, the state AIDS program representative, the WHO South East Asia Regional Organisation representative, and representatives from the Indian Council of Medical Research. A number of health care providers were also present. In addition, partner 6, Cavidi AB had a presentation of the HIVIND results using their low cost viral load test to the National AIDS Control Program in New Delhi, India. International dissemination of the project results are various forums are also scheduled. The HIVIND trial has been accepted for presentation at the International AIDS conference, Melbourne.

AIDS 2010, XVIII International AIDS Conference, Vienna, Austria July 2010 Vienna, Austria
YRG Science, 2010, Chennai, India Aug 2010 Chennai, India
International Conference on Opportunistic Pathogens, (ICOPA) New Delhi, India Sept 2010 New Delhi, India
French-Indian Inter-academic Symposium on Infectious Diseases, National institute of Immunology, New Delhi, India Dec 2010 New Delhi, India
International Conference on Emerging Frontiers and Challenges in HIV/AIDS Research, Mumbai, India Feb 2011 Mumbai, India
AREVIR-Meeting 2011 (EuResist), Bonn, Germany May 2011 Bonn, Germany
6th IAS Conference on HIV Pathogenesis, Treatment and Prevention, Rome, Italy July 2011 Rome, Italy
10th International Congress on AIDS in Asia and the Pacific, Busan, Korea
Aug 2011 Busan, Korea
AIDS Vaccine 2012 Sept 2012
XIX International AIDS Conference, July 2012 Washington DC, USA
Anti retroviral meeting April 2014 Chennai, India
Emerging Frontiers and Challenges in Management and Control of STIs and HIV (Indian Council of Medical Research) April 2014 Mumbai, India
XX International AIDS Conference (July 2014) Melbourne, Australia

Publications from the HIVIND project:
1. Shet A, Decosta A, Heylen E, Shastri S, Chandy S, Ekstrand M. High rates of adherence and treatment success in a public and public-private HIV clinic in India: potential benefits of standardized national care delivery systems. BMC Health Serv Res. 2011 Oct 17;11(1):277.

2. Shet A, de Costa A.(2011) India calling: harnessing the promise of mobile phones for HIV healthcare. Trop Med Int Health;16(2):214-216.

3. Neogi, U., Prarthana, B.S. Gupta, S., D’souza, G., De Costa, A., Kuttiatt, V.S. Arumugam, K., and Shet, A., (2010) Naturally occurring polymorphisms and primary drug resistance profile among antiretroviral-naïve individuals in Bangalore, India. AIDS Res Hum Retroviruses 26(10): 1097-1101.

4. Neogi, U., Sahoo, P.N. Ravi Kumar, B.N. De Costa, A., Shet, A., (2011) Characterisation of HIV-1 subtype C protease gene: selection of L63P mutation in protease inhibitor-naive Indian patients. AIDS Res Hum Retroviruses. 2011 May 9.

5. Neogi, U., Prarthana, B.S. D’souza, G., De Costa, A., Kuttiatt, V.S. Ranga, U., and Shet, A., (2010) Co-receptor tropism prediction among 1045 Indian HIV-1 subtype C sequences: Therapeutic implications for India. AIDS Res. Ther. 7:24 (21 July 2010)

6. Neogi, U., Prarthana, B.S. Gupta, S., D’souza, G., De Costa, A., Kuttiatt, V.S. Arumugam, K., and Shet, A., (2010) Naturally occurring polymorphisms and primary drug resistance profile among antiretroviral-naïve individuals in Bangalore, India. AIDS Res Hum Retroviruses 26(10): 1097-1101.

7. Shet A, Arumugam K, Rodrigues R, Rajagopalan N, Shubha K, Raj T, D'souza G, De Costa A.(2010) Designing a mobile phone-based intervention to promote adherence to antiretroviral therapy in South India. AIDS Behav;14(3):716-20.

8. De Costa A, Shet A, Kumarasamy N, Ashorn P, Eriksson B, Bogg L, Diwan VK; HIVIND study team (2010) Design of a randomized trial to evaluate the influence of mobile phone reminders on adherence to first line antiretroviral treatment in South India--the HIVIND study protocol. BMC Med Res Methodol.;10:25.

9. Rodrigues RJ, Antony J, Krishnamurthy S, Shet A, De Costa A. 'What do I know? Should I participate?' Considerations on participation in HIV related research among HIV infected adults in Bangalore, South India. PLoS One. 2013;8(2):e53054. [PMID: 23460780]

10. Rodrigues R, Shet A, Antony J, Sidney K, Arumugam K, Krishnamurthy S, D'Souza G, Decosta A. Supporting adherence to antiretroviral therapy with mobile phone reminders: results from a cohort in South India. PLoS One. 2012;7(8):e40723. Epub 2012 Aug 27. [PMID: 22952574]

11. Neogi U, Bontell I, Shet A, De Costa A, Gupta S, Diwan V, Laishram RS, Wanchu A, Ranga U, Banerjea AC, Sönnerborg A. Molecular Epidemiology of HIV-1 Subtypes in India: Origin and Evolutionary History of the Predominant Subtype C. PLoS One. 2012;7(6):e39819. Epub 2012 Jun 29. [PMID: 22768132]

12. Sidney K, Antony J, Rodrigues R, Arumugam K, Krishnamurthy S, D'souza G, De Costa A, Shet A. Supporting patient adherence to antiretrovirals using mobile phone reminders: patient responses from South India. AIDS Care. 2012;24(5):612-7. [PMID: 22150088]

13. Rodrigues R, Bogg L, Shet A, Sunil N, De Costa A. Mobile phones to support adherence to antiretroviral therapy – what would it cost the Indian National AIDS Control Program? J International AIDS Society (Accepted June 2014)

14. Shet A, Antony J, Arumugam K, Kumar Dodderi S, Rodrigues R, De Costa A. Influence of Adverse Drug Reactions on Treatment Success: Prospective Cohort Analysis of HIV-Infected Individuals Initiating First-Line Antiretroviral Therapy in India. PLoS One. 2014 Mar 10;9(3):e91028

Manuscripts drafted/ currently under review

15. Telephone reminders are not effective in influencing treatment success in HIV: evidence from a randomized controlled trial in India (under review in the BMJ)
16. Longitudinal analysis of patterns and predictors of adherence to first-line antiretroviral therapy: evidence of sustainability from an Indian HIV cohort (drafted)
17. Virological efficacy with first-line antiretroviral treatment in India: predictors of viral failure and evidence of viral resuppression (under review in Trans Roy Soc Trop Med)
18. mHealth adherence support intervention for antiretroviral therapy in South India: Participants perceptions from the HIVIND trial

19. Good correlation of HIV Reverse Transcriptase Enzyme Activity Assay with HIV-1 RNA load makes it a useful and cost-effective surrogate for virological monitoring in resource-limited settings – A report from the HIVIND cohort in South India (drafted)



Contact address: ayesha.de.costa@ki.se