Outcome domains and measurement instruments of patient-relevant improvement of structure and processes as a new set of outcomes for evaluating and approving digital health applications: systematic review

Background In October 2020, digital health applications (DiGAs) became part of standard care in Germany. For approval, DiGA manufacturers must demonstrate medical benefit or patient-relevant improvement of structure and processes (PISP). PISP refers to an innovative outcome core area in terms of proof of benefits and reimbursement decisions. These are subdivided into 9 outcome domains, including for example health literacy, facilitating access to care, and coping with illness-related difficulties in everyday life. Their implementation aims at empowering patients, encouraging shared decision-making, and increasing patient-centeredness in healthcare delivery. Given the novelty of PISP, no standardized set of outcomes and outcome measurement instruments currently exists to operationalize the domains. Learning from previous evaluation studies can help operationalize and standardize PISPs for evaluation studies of digital health applications. Therefore, we investigated the outcomes and outcome measurement instruments, used in controlled trials to assess DiGA-compliant applications, published before the Digital Health Applications Ordinance of April 2020. Methods We conducted a systematic review of studies published between 01/2015 and 04/2020, via MEDLINE and Embase, complemented by forward/backward searches. Controlled trials assessing interventions adhering to the definition of DiGA were eligible, if they applied a validated outcome measurement instrument, and if results were presented in German or English. Title-abstract screening, full-text screening, data extraction and narrative synthesis were conducted independently by two researchers. Results Out of 2,671 references identified, 6 studies collecting a total of 48 outcomes were included. 14 outcomes (29.2%) addressed PISP by using 13 different measurement instruments. The outcomes corresponded to 5 of 9 PISP outcome domains with health literacy being the most common (7/14, 50.0%). Conclusions This review provides an overview of the characteristics of PISPs used in previous evaluation studies of DiGA-compliant applications. It shows which outcomes and validated outcome measurement instruments can be used to measure PISP and where knowledge is still lacking. These results serve as a starting point for operationalizing and standardizing PISPs and help to increase the outcome measurement quality of PISPs.