How Similar Are Drug Data and Disease Self-report? Estimating the Prevalence of Chronic Diseases in Less Developed Settings.

Abstract

Drug data has been used to estimate the prevalence of chronic diseases. Disease registries and annual surveys are lacking, especially in less-developed regions. At the same time, insurance drug data and self-reports of medications are easily accessible and inexpensive. We aim to investigate the similarity of prevalence estimation between self-report data of some chronic diseases and drug data in a less developed setting in southwestern Iran. Baseline data from the Pars Cohort Study (PCS) was re-analyzed. The use of disease-related drugs were compared against self-report of each disease (hypertension [HTN], diabetes mellitus [DM], heart disease, stroke, chronic obstructive pulmonary disease [COPD], sleep disorder, anxiety, depression, gastroesophageal reflux disease [GERD], irritable bowel syndrome [IBS], and functional constipation [FC]). We used sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and the Jaccard similarity index. The top five similarities were observed in DM (54%), HTN (53%), heart disease (32%), COPD (30%), and GERD (15%). The similarity between drug use and self-report was found to be low in IBS (2%), stroke (5%), depression (9%), sleep disorders (10%), and anxiety disorders (11%). Self-reports of diseases and the drug data show a different picture of most diseases' prevalence in our setting. It seems that drug data alone cannot estimate the prevalence of diseases in settings similar to ours. We recommend using drug data in combination with self-report data for epidemiological investigation in the less-developed setting.