Digital health applications (DiGA), both in the sense of the Digital Care Act of 2019 as patient-centric applications and in the sense of the use of telemedical solutions on the part of providers (e.g. teleconsultations, electronic files), are increasingly recognized as real clinical interventions (Kowatsch et al 2019). They have the potential to supplement and improve care and to make processes more effective (Eysenbach 2001; World Health Organization (WHO) 2015). Nevertheless, innovative digital applications in Germany, as in many other Western healthcare systems, are often not established in standard care (Huang et al. 2017). The dissemination of an innovation within an existing system is described in science with the term diffusion and the associated theory of diffusion of innovations (Rogers 2003). In practice, people tend to talk about the phenomenon of inability to scale up. Above all, this phenomenon underlines that it is a central problem for digital innovation projects to leave the test phase in the project context and to expand the sphere of action to include standard care (Häckl 2010; Gersch and Rüsike 2011).

>> Within statutory health insurance, with the entry into force of the law for better care through digitalization and innovation (Digitale-Versorgung-gesetz DVG) on December 19, 2019, the first prerequisites for the reimbursement of DiGA by standard care were created, thereby making market entry much easier . The prerequisite for reimbursement by statutory health insurance companies is the inclusion of DiGA in the DiGA directory, which is maintained by the Federal Institute for Drugs and Medical Devices (BfArM).
In order to actively promote the diffusion and scalability of digital innovations in the context of the German healthcare system, this article formulates concrete demands for future steps on the part of politics and self-administration. These are intended to help lower existing barriers to the development, implementation and accompanying evaluation of DiGA in the German healthcare system.
Requirements for the successful development, implementation and evaluation of digital health applications – a selection
Overview of the catalog of requirements
The demands are based on the scientific results of the young research group Care4Saxony, which examined the potential of DiGA within the three-year project period (07/2017-09/2020). In this article, selected requirements are presented against the background of existing approaches to evaluating and implementing digital applications in healthcare, such as the NICE framework of the British National Health Service (NHS).1 The requirements guide the evidence base for digital health applications improve and include the current adjustments to the procedure for their approval in Germany, ie the BfArM Fast Track.2
The demands are summarized in a catalog of demands and each contain a rationale that describes the scientific background and consequences that arise from the respective demand. In addition, each claim is accompanied by a warning about potential problems that may arise as a result of ignoring the claim.3
The subject areas of the list of requirements are summarized in Figure 1. This article presents four selected requirements that focus specifically on the development (1 - standardized approach), implementation (3 - supportive community) and evaluation (5 - evidence through usability and acceptance and 6 - evidence on effectiveness and benefits) of digital health applications . The necessary basic requirement is the appropriate qualification and further development of the digital skills of DiGA users
The focus is on what is required both in the context of medical studies and in the area of further training as well as in the training of medical and nursing professions. Without these skills, widespread acceptance and use of digital health applications will be more difficult to achieve (Brockes et al. 2017).

A uniform diffusion framework enables standardized procedures for implementation and evaluation
Requirement: A standardized and phased approach is required for the introduction and evaluation of DiGA. For this purpose, we propose the four stages “prototype, solution with first pilot test, broadly tested solution and solution on the market” (Broens et al. 2007) (see Fig. 2).
Rationale: Digital applications in healthcare arise in different environments and based on different prerequisites and preparatory work. Both for companies operating in this market environment and for those healthcare organizations that decide on the financing of such solutions (such as the Federal Joint Committee or the health insurance companies within the framework of selective contracts), it is difficult to ensure the conformity of the solutions with existing ones To evaluate regulations or the potential of the solutions (Dittmar and Wohlgemuth, 2009).
Warning: Unstructured development and implementation of digital healthcare applications can lead to critical project delay and a multitude of identical solutions (Lehmann and Bitzer, 2019). Identifying new approaches individually makes little economic sense and contradicts the development process of digital applications.

Regional structures support digital transformation
Demand: Regions and interest groups should support the implementation of DiGA in the healthcare system to the extent possible. By providing resources and administrative management, among other things, the implementation of DiGA and thus the digital transformation of healthcare within a region can be successful.
Rationale: Whether DiGAs are successful depends on various factors, including the motivation of the end users, the technical functionalities and the legal or financial framework (Otto 2019).
Regional and interest groups can play an additional supporting role towards end users within the applicable legal and financial framework (Edwards et al. 2000). We demand that regional communities and communities take on this role
satisfy. The support can be diverse and ranges from the provision of financial, human and technical resources to the organization of educational campaigns about the possibilities and advantages of DiGA (Atkinson et al. 2009). Existing tools such as the “Telemedicine Maturity Model in Communities”4 can explicitly help to identify specific support options for each community.
Community support should always occur when there is a regional need for DiGA. This need can be identified based on regional structural data (on sociodemographics, disease prevalence and care structures, technical and administrative infrastructure) and structured needs of regional actors (Khatun et al. 2015; Harst et al. 2019b).
Warning: Regional (supply) structures, administrative processes and networks should support the implementation of DiGA, otherwise there is a risk of the population's lack of acceptance of the applications and the risk of implementing unsuitable applications.

Acceptance, usability and the subjective perception of usefulness determine the effectiveness of DiGA
Requirement: Both the design and usability of an application, as well as the preferences of patients and service providers, should be examined during the development phase and should also be seen as evidence for a sustainably successful implementation (Årsand and Demiris, 2008).
Rationale: In addition to clinical parameters as traditional measures of the effectiveness of an intervention, acceptance or the subjective perception of usefulness and ease of use by end users, i.e. patients or service providers, are also important drivers for long-term use of telemedicine applications (Harst et al. 2019a ). In this respect they differ, at least in part, from pharmacological interventions, although here too a lack of willingness to take them regularly can influence the effect. Outcomes such as health-related quality of life, as provided in the BfArM Fast Track as relevant indicators for the effectiveness of DiGA, are based on this individual perception of the usefulness of an application (Bonn et al. 2019).
Therefore, there is a need for a standardized user-centered development process for digital applications in healthcare that is compliant with existing ISO standards such as ISO 9241 (“Ergonomics of human-system interaction”). Part of this process should be a continuous evaluation of the acceptance and usability of an application before, during and after its implementation.
Warning: Taking into account the patient's subjective assessment and perception is a relevant change, especially from the perspective of evidence-based medicine (EbM), as cumbersome operation and functionalities that do not correspond to the users' preferences are important in evaluating the effectiveness of telemedicine applications in particular can act as disruptive factors. As a result, the effect of the DiGA may be distorted despite the intervention components being evidence-based.

The development and consensus of Core Outcome Sets (COS) in telemedicine studies facilitate the evaluation of DiGA
Requirement: Following on from the requirement described above, core outcome sets are still required to ensure the collection of patient-relevant endpoints and to increase the comparability of study results. Therefore, in the evaluation of DiGA, in addition to endpoints for measuring clinical effectiveness, Patient Reported Outcomes (PRO for short) and endpoints for mapping patient experiences (Patient Reported Experiences, PRE for short) should also be collected in a regular and standardized manner.
Rationale: Digital health applications are often designed as complex and/or multimodal interventions, making causal assessment of effectiveness difficult (Yasmin et al. 2016). There is currently no “best practice approach” to evaluate digital health applications (McKay et al. 2018; Zanaboni et al. 2018). Comparing the effectiveness of digital health applications is made even more difficult by the use of heterogeneous outcomes.
Evidence syntheses show that the effectiveness of digital health applications depends on patient characteristics in terms of their clinical health status, but also on their technology acceptance and individual needs for social support (Harst et al. 2019a; Timpel et al. 2020). Such care, which is geared towards the respective benefits for the patient, follows the principles of “value-based healthcare” (Porter 2010).
The standardized use of relevant outcomes has the potential to sustainably improve the evidence base for digital applications in healthcare. Core outcome sets can contribute to improved comparability of study results by standardizing endpoints (Schmitt et al. 2015; Williamson et al. 2017; Kirkham et al. 2017). This requires the development of new or expansion of existing core outcome sets for the evaluation of digital applications in healthcare in order to generate evidence beyond pure clinical effectiveness.
Warning: An exclusive focus on clinical effectiveness neglects relevant domains of effectiveness, in particular the perception of users and changes in their illness behavior (reality of care). In addition, when collecting heterogeneous outcomes, there is a risk that study results can only be compared to a limited extent, making it more difficult to derive evidence-based recommendations for DiGA.
Summary
The article addresses the challenge of perpetuating DiGA. Using a list of requirements, points were identified that show the need for action in dealing with and evaluating DiGA. For example, it was found that user acceptance and their subjective perception of benefits are demonstrably just as important for the development and diffusion of digital health applications as the collection of primarily medical endpoints, which corresponds to the specific design of the BfArM Fast Track.
The demands therefore also provide assistance and guidance for the sustainable development of digital applications in the healthcare sector. For this purpose, research needs in the area of core outcome sets, regional maturity of communities and the patient-centered benefit assessment of digital health applications are described, which can contribute to improving the evidence base. The diffusion framework for DiGA shown in Fig. 2 summarizes all of these requirements and visualizes the ideal
typical development and diffusion process and is intended to help existing and future DiGA to be implemented sustainably into standard care. <<

1. NICE stands for the National Institute for Health and Care Excellence
2. https://www.bfarm.de/SharedDocs/Downloads/DE/Service/Beratungsverfahren/DiGA-Leitfaden.pdf
3. The full catalog of demands in its long version can be found online at http://care4saxony.de/?page_id=2731

4. http://care4saxony.de/?p=3324