JMIR Publications

ABSTRACT

Despite half a century of dedicated studies, medication adherence remains far from perfect, with many patients not taking their medications as prescribed. The magnitude of this problem is rising, jeopardizing the effectiveness of evidence-based therapies. An important reason for this is the unprecedented demographic change at the beginning of 21st century. Ageing leads to multimorbidity and complex therapeutic regimens that create fertile ground for non-adherence. As this scenario is a global problem, it needs a worldwide answer. Might we find this answer thanks to new opportunities created by the digitization of healthcare? Day by day health-related information is collected in electronic health records, pharmacy dispensing databases, health insurance systems and national health system records. These Big Data repositories offer an unprecedented opportunity to study adherence both retrospectively and prospectively, at the population level, as well as its related factors. If only we had widely accepted standard measures of adherence, we could use this data to inform health research, clinical practice and public health. These standards could also help us to better understand the relationship between adherence and clinical outcomes, and allow for fair benchmarking of effectiveness and cost-effectiveness of adherence-targeting interventions. Unfortunately, despite this obvious need, we are still far from having sound standards of formatting, and analyzing Big Data in order to assess, uniformly present and compare patterns of medication adherence across studies. The aim of this paper is to call for producing such consensus standards, in order to help adherence, and make healthcare systems more effective and sustainable.