As seen in the recent coronavirus pandemic, governments worldwide are expected to manage not only the disease burden of COVID-19 but its impact well beyond the realm of health – the socio- economic, psychological, and welfare impacts have in fact dominated the political discussion. The manifestation of diseases on the population’s well-being, its effect on families and the economy can be more informative for policymaking than merely measuring prevalence or deaths. While one-off studies on specific non-health disease impacts exist, translating findings into policy decisions is difficult because cross-study comparisons cannot be made, and there is no guidance on how to make trade-offs between diseases or across multiple dimensions, e.g. between economic and welfare impacts. The aim of Measuring Experienced Disease Burden (ExpBoD) is to deliver a proof of concept to systematically measure the impact of several diseases across multiple dimensions in Denmark.
Objective: research question and potential hypotheses
Main research question: What is the experienced disease burden among patients with breast cancer, alcohol use disorder, and stroke, respectively? Experienced disease burden is defined as the socio- economic, psychological, healthcare and welfare impacts due to the disease.
Secondary research questions: (1) What are the strengths and weaknesses of existing statistical methods to measuring non-health disease impacts in registry-based studies? (2) How should we facilitate meaningful comparisons of experienced burden across diseases, taking into account the socio-economic and demographic differences across patient groups?
Hypotheses: (1) Even if the levels of disease burden (measured in traditional metrics such as years of life lost or prevalence) are similar, the non-health impacts vary widely; (2) Experienced burden can be systematically measured using existing data sources and compared across diseases; (3) Certain statistical approaches and parameter selection are more suitable for measuring experienced burden. Objectives: (1) Assess statistical methods to identify the most appropriate methodology, (2) Systematically estimate disease impact across dimensions and diseases, conduct sub-group and decomposition analyses, (3) Create a composite index for experienced burden.
Comparative disease burden estimation is a longstanding topic of research in epidemiology, and health policymakers depend on comparative information to prioritize efforts across diseases towards those at greatest disadvantage. Similar comparative estimation of the non-health burden of disease (including economic, social, welfare, healthcare impacts) is useful to understand how different disease burdens affect personal and population level productivity. In this study, we compare the comparative long-term income losses experienced by individuals diagnosed with 20 different diseases across the population in Denmark. We seek to identify who is most affected by the non-health disease burden due to incident disease, and whether different diseases show different gradients across socioeconomic and demographic stratifications.
We adopt a cohort study design using Danish register data to estimate the causal effect of disease incidence on income loss, healthcare utilization, welfare receipt, marital status, and use of prescription drugs for mental illnesses across 20 diseases in Denmark from 2000 to 2018. We match individuals with incident disease burden to a control population by age, sex, region, baseline income, employment status, baseline health, education history, household size, and relationship status and follow outcomes for 10 years following diagnosis or until censorship. For diseases with acute onset such as cancer or stroke, we match in the year before diagnosis, and for diseases that present slowly over time such as depression we match to control 5 years before incident diagnosis. On income, personal and household equivalized income loss over the whole period is estimated for each disease, along with the mortality rate in each cohort. Results are stratified by age, employment status, education, and region in Denmark.
