Currently submitted to: JMIR Medical Informatics
Date Submitted: Jul 23, 2019
(closed for review but you can still tweet)
Navigating through digital health outcome evaluation methodologies: an algorithm based on a scoping review and emerging methodologies.
Digital health interventions are recognized for their potential and are increasingly implemented globally. The evidence base is growing, but currently there are still relatively few studies evaluating improvements of digital health interventions in health outcomes
The objectives of this article were to understand, analyze and map how researchers approach digital health outcome evaluations in different settings through a scoping review and to develop an algorithm, based on these results, to provide a pedagogical overview of methods for evaluating health outcomes of digital health interventions.
For the scoping review PubMed, EMBASE, and the Cochrane Central Register of Controlled Trials (CENTRAL) were scanned using a predefined search strategy to identify articles measuring the impact of digital health interventions on health outcomes. The algorithm was developed based on analysis and insights from the scoping review.
The database search retrieved 3584 citations of which 208 were included. These articles were reviewed in detail and were classified into different categories: level of income (of the country where the study was conducted), system categories, disease addressed by the intervention, and evaluation method. The gap analysis identified additional emerging approaches that were included in the algorithm.
Through analysis of the literature, we were able to demonstrate that digital health outcome studies rely on traditional clinical evaluation designs, even though these interventions are often more complex and depended on the context, culture, and the individual than classical interventions like insulin on its receptor. In addition to the methodologies extracted from studies through the database search we identified study methodologies through desk research, whose design features address some of the shortcomings of traditional clinical methodologies, when applying them to digital health interventions. We integrated all identified methodologies into an algorithm that provides a high-level overview, enables the user to navigate through these methodologies based on the design features and investigator’s priorities, and to facilitate the identification of one or more potential appropriate methodologies.