Equitable and systemic integration of Digital Health Technologies: challenges and opportunities

Safe and effective Digital Health Technologies (DHTs), such as medical-grade connected sensors and digital health applications, are transforming healthcare on a global scale. However, fragmented policies and inequities in access and adoption hinder their full potential to empower providers, support patients and caregivers, and enhance care. In the EU, while countries like Germany, France, and Belgium have established dedicated frameworks for assessing and reimbursing DHTs, others are at different stages of progress — from early planning to consolidation — or still rely on ad hoc approaches or lack fit-for-purpose pathways altogether1. This uneven landscape contributes to disparities in access and hinders the systemic integration of DHTs across Member States. Beyond access, adoption requires targeted policy interventions, ecosystem efforts and technology management approaches2. As the EU further advances towards a Health Union and digital single market with emerging regulations as the Health Technology Assessment Regulation (HTAR)3, the European Health Data Space (EHDS)4, and the AI Act5, there is urgency to both ensure safety and scale innovation through harmonised policies. This Comment proposes five complementary policy interventions to support equitable access, adoption, and systemic integration of DHTs across Europe and beyond.

Building on the concept of a policy “full-stack”6, we extend our focus beyond regulatory clarity and access, to systemic integration and adoption. We further account for the rising role of Artificial Intelligence (AI) and offer actionable recommendations to bridge existing policy gaps. Our framework incorporates five key components: (1) defining DHTs in the scope of targeted policy interventions; (2) ensuring AI-ready regulatory frameworks and Accelerated Access Pathways; (3) establishing dynamic, fit-for-purpose Health Technology Assessments (HTAs); (4) implementing access and reimbursement best practices; and (5) supporting system and clinical integration. We also advocate for multistakeholder collaboration and cross-country harmonisation to reduce fragmentation and scale innovation. Together, these mutually reinforcing enablers provide actionable pathways for systemic integration of DHTs (Fig. 1).

Fig. 1: A virtuous policy cycle for systemic and equitable integration of Digital Health Technologies.
figure 1

The five mutually reinforcing policy enablers—(1) defining DHTs within fit-for-purpose policy frameworks, (2) ensuring AI-ready regulations and Accelerated Access Pathways, (3) establishing dynamic, fit-for-purpose Health Technology Assessment (HTA) models, (4) implementing fit-for-purpose access and reimbursement models, and (5) supporting clinical integration and adoption through structured policies—form a comprehensive and actionable policy framework to enable systemic and equitable integration of DHTs, scaling impact and promoting harmonisation within and across health systems.

Full size image

Defining DHTs in scope of fit-for-purpose policies

Defining the subset of DHTs requiring (more) targeted policies is key to enabling fit-for-purpose interventions and cross-country harmonisation. While their terminology may differ, national frameworks in Germany, France, and Belgium consistently focus on software applications certified as medical devices under the EU Medical Device Regulation (MDR)7—formally referred to in the MDR as Medical Device Software (MDSW)—with a demonstrable health benefit. To date, these targeted policies have focused exclusively on safe and effective software devices intended for direct use by patients or jointly by patients and healthcare professionals (HCPs). By contrast, those intended solely for HCP use have not yet been included in their scope and remain subject to traditional medical device regulatory and reimbursement pathways.

Germany’s 2019 Digital Healthcare Act8 and Digital Health Applications Ordinance (DiGAV)9 established the first legal basis and criteria for reimbursing Digital Health Applications (“Digitale Gesundheitsanwendungen”, DiGA) via statutory health insurance. Initially limited to Class I and IIa patient-facing or jointly patient- and HCP-facing software certified under the MDR, eligibility expanded under the Digital Act (DigiG) to encompass class IIb devices, including remote monitoring devices, under somewhat more restrictive conditions10,11. France formally recognized Digital Medical Devices (Dispositifs Médicaux Numériques) for therapeutic and remote monitoring purposes as a distinct medical device subgroup in 202312. Belgium’s mHealth initiative, launched in 2018 and revised in 2024, introduced a multi-level validation and reimbursement framework for MDR-certified software applications that allow patients to share health-related information (with or without sensors) with healthcare professionals13. Examples of DHTs currently in scope of existing fit-for-purpose policy frameworks include safe and effective Digital Therapeutics14 (DTx), remote patient monitoring devices15 and digital care support programs16 (Table 1).

Table 1 Non-exhaustive examples of Digital Health Technologies in scope of existing fit-for-purpose policy frameworks in Germany, France, and Belgium
Full size table

At the EU level, research and harmonisation initiatives have begun using the term Digital Medical Devices (DMDs)17,18,19,20,21 to refer to the subset of certified MDSW that is supported by evidence of a demonstrable health benefit and is therefore within the scope of targeted policy interventions. While all DMDs are MDSW, not all MDSW qualify as DMDs, as some—particularly lower-risk class devices—may lack evidence to claim a demonstrable positive care impact (Fig. 2).

Fig. 2: The emerging terminology of Digital Medical Devices in scope of fit-for-purpose policies.
figure 2

Digital Health Technologies (DHTs) that fall within existing formal fit-for-purpose policy frameworks in the EU share key characteristics: (i) they are certified as medical device software (MDSW) under the EU’s Medical Device Regulation (MDR); (ii) they demonstrate a positive health impact; (iii) they are intended for direct use by patients or jointly by patients and healthcare professionals; (iv) they are not intended for primary prevention; (v) they are typically “stand-alone”, not provided with a medicinal product. This subset of both safe and effective MDSW is increasingly referred to in Europe as Digital Medical Devices (DMDs). We recommend expanding the scope of fit-for-purpose policies also to include HCP-facing DMDs, those used alongside pharmaceuticals, and those intended for primary prevention. Note: the sizes of the areas in the Venn diagram are not representative of actual market sizes.

Full size image

Beyond the scope of existing formal frameworks, we recommend three key evolutions. While current targeted policies are limited to patient-facing DMDs, we propose that DMDs intended exclusively for HCPs—such as software-based clinical decision support systems—should also be included. Many of the regulatory, clinical, and implementation challenges—further elaborated in the following sections—are equally relevant to these HCP-facing technologies.

In addition to the current focus on disease management, treatment or monitoring, we argue that targeted policies should encompass DMDs for primary prevention. Increasing access to DMDs that address modifiable risk factors—such as smoking, unhealthy diets, sedentary lifestyles, and hypertension—through widely accessible mobile technologies could support population-level behaviour change and reduce the burden of preventable non-communicable diseases (NCDs)22. This emerging field, at the crossroads between behavioural science and digital health, holds significant potential23. While primary prevention is often underfunded, this shift could be operationalised by integrating prevention-focused DMDs into national NCD strategies, linking their use to incentive mechanisms—as conditional cash transfers or reward systems—and aligning them with EU-level initiatives like the EU4Health Programme24 or the Healthier Together NCD initiative25.

Finally, as DMDs are increasingly co-developed and/or launched alongside pharmaceuticals and traditional medical devices, policy frameworks should evolve beyond the current narrow focus on “stand-alone” digital devices. The U.S. Food and Drug Administration (FDA)’s Prescription Drug Use-Related Software framework begins to address this by allowing software to be included in drug labels when added clinical benefits are demonstrated26.

AI-Ready Regulatory Approaches & Accelerated Access Pathways

DMDs must meet jurisdiction-specific regulations, such as the MDR in the EU, to achieve certification for market placement. However, discrepancies between regulatory and reimbursement requirements often delay patient access. Currently, there are no guarantees that the clinical evidence required for certification will be sufficient for HTA and coverage decisions. To address these gaps and fragmentation, we recommend adopting Accelerated Access Pathways at the EU level, as already proposed for high-risk devices27 and implemented in specific settings in the U.S.28, which support simultaneous regulatory and reimbursement assessments—streamlining certification, reducing delays, and promoting cross-country harmonisation.

Furthermore, as AI—particularly generative AI—is increasingly integrated into DMDs, regulations must strike a balance between enabling innovation through algorithmic adaptiveness and autonomy, and ensuring continuous product safety and performance. Predetermined Change Control Plans, as developed jointly by the FDA, UK’s MHRA, and Health Canada, offer a pathway for safe and efficient iteration of AI-based features without requiring repeated regulatory approval29,30. We propose defining and adopting similar mechanisms across the EU.

Finally, in the EU, aligning the MDR with the AI Act is essential to avoid delayed certification, which could hinder timely availability of AI-driven DMDs across EU markets. We recommend ensuring that notified bodies can effectively perform simultaneous assessments under both regulations for all AI-driven DMDs that require conformity assessment under the MDR and are therefore classified as “high-risk” under the AI Act, as specified in Recital 50.

Fit-for-Purpose and Dynamic Health Technology Assessments

Establishing market access pathways for DMDs requires fit-for-purpose HTA criteria, appropriate for the dynamic software development process and for its unique characteristics31. Germany’s DiGAV outlines requirements for DiGA assessments overseen by the Federal Institute for Drugs and Medical Devices (BfArM). France’s PECAN (Prise en charge anticipée numérique) pathway uses parallel assessments: CNEDiMTS (“Commission Nationale d’Évaluation des Dispositifs Médicaux et des Technologies de Santé”) evaluates clinical benefits and organizational improvements, and the Digital Health Agency (“Agence du Numérique en Santé”) ensures interoperability and security compliance. Belgium’s multi-level framework includes technical, clinical, and economic criteria. England has dedicated evidence standards for DHTs, the Digital Technology Assessment Criteria as a baseline for NHS integration32 and NICE’s Early Value Assessment33 for accelerated evaluations.

HTA criteria for DMDs are formally converging across countries. Table 2 and Table 3 apply the HTA Core Model®—a framework developed by the European Network for Health Technology Assessment34—to highlight converging standards and scalable best practices for cross-country harmonisation. However, fragmentation persists due to diverse local dossier structures, supplementary standards and regulations, and limited evidence transferability.

Table 2 Overview of HTA Requirements Across Established European Frameworks and Convergence Outlook for HTA Core Model® Domains Within the Scope of Rapid Relative Effectiveness Assessment (REA)
Full size table
Table 3 Overview of HTA Requirements Across Established European Frameworks and Convergence Outlook for HTA Core Model® Domains Beyond Rapid Relative Effectiveness Assessment (REA)
Full size table

Policymakers should harmonise technical, legal and safety standards. Joint clinical assessments, mutual recognition agreements, and conditional international evidence recognition based on local Real-World Evidence (RWE) generation are timely measures. Shared metrics to capture health economic, organizational and social outcomes will also be key to valuing DMD’s broader impact. Efforts to harmonise assessments and voluntary cooperation under the HTAR35, will be fundamental. Globally, initiatives to clarify regulatory process across jurisdictions36 represent a valuable resource in the current patchworked policy environment.

Further, current policy frameworks do not systematically adopt “dynamic HTAs”, which are designed to accommodate software updates and would be particularly suitable for assessing DMDs37. Unlike drugs, which are marketed in their approved form indefinitely, DMDs require updates to improve software compatibility, functionality, and security. They also efficiently record Real-World-Data (RWD), enabling RWE generation to dynamically assess care impact, adoption and inform performance-based pricing models38,39. As DMDs increasingly integrate adaptive and generative AI, dynamic HTAs will become even more critical. England has begun updating its evidence standards with expert-based consensus recommendations for AI-driven technologies40, recognizing that traditional methods used to assess the clinical and economic benefit of drugs or conventional medical devices do not sufficiently capture the specificities of these technologies. Among other requirements, the recommendations specify that vendors should provide a clear definition of the level of autonomy with which the AI-system is expected to operate, a post-deployment oversight process, and a post-deployment change management plan prior to deployment. We propose evolving towards dynamic, fit-for-purpose HTAs for DMDs to better address these emerging complexities.

Access, pricing and reimbursement mechanisms

Reimbursement is key to bridging regulatory approval, HTA, and equitable access. England illustrates this challenge: despite clear regulatory and HTA standards, NICE recommendations are not directly linked to reimbursement by local boards, limiting systemic adoption41. While DMD reimbursement models vary across countries—also reflecting differences in health system structure, financing, and governance—common enablers that could be adapted and scaled internationally include: early access pathways through Coverage with Evidence Development (CED) mechanisms, HTA-linked reimbursement, and fit-for-purpose pricing models.

Germany’s DiGA “Fast-Track” grants temporary reimbursement, while additional evidence is generated, for up to 12 months (with possible extensions) for DMDs below risk class IIb; France’s PECAN pathway enables early coverage for therapeutic and telemonitoring DMDs, and Belgium’s “M3-light” stage allows provisional reimbursement. All models require DMDs to meet technical and safety standards as prerequisites, with healthcare benefits assessed following temporary coverage. CED pathways suit DMDs given their short lifecycle, frequent updates, lower risk, and capability to generate RWE via early use.

Linking HTA to centralized national reimbursement decisions, as in Germany and France, ensures more timely, equitable and systemic access to DMDs. Decentralized systems would also benefit from stronger HTA-reimbursement alignment. Italy’s new National HTA Program is advancing in this direction42.

In Germany, manufacturers set an initial DiGA price for a product’s first 12 months (which should typically align with provisional access), followed by evidence-based negotiations with the National Association of Statutory Health Insurance Funds. France’s PECAN pathway sets predefined prices for telemonitoring DMDs (up to €50/patient/month for telemonitoring DMDs with organizational impact, and up to €73.33, €82.50, and €91.67/patient/month for those with clinical impact on quality of life, morbidity, and mortality, respectively; in addition to up to €28/patient/month for HCP remuneration)43 and therapeutic DMDs (€435 for the first three months, then €38.30/month, capped at €780.00 annually)44; manufacturers can then apply for a 5-year listing and negotiate prices under common law. Belgium’s first bundled payment model integrates telemonitoring DMD reimbursement into predefined care pathways, starting with congestive heart failure45. This fixed model includes a one-time €200/patient startup fee (covering prescription, device provision, patient training, and first-month oversight) and monthly fees of €90/patient for months 2–6, then €45/patient/month from month 7 onward. Patients face no cost-sharing, and a €24.92/patient annual fee supports general practitioners’ coordination. The U.S. also reimburses provider services for telemonitoring of both chronic and acute conditions, including separate billing codes for device set-up, data interpretation, and remote treatment management46,47.

With respect to pricing and payment models, we advocate expanding HCP reimbursement across countries and DMD types, acknowledging their pivotal role in enabling appropriate adoption and clinical integration, including for patient-facing technologies. As we recommend including HCP-facing DMDs within the scope of dedicated policies, we also note that while regulatory and HTA pathways may align with those for patient-facing DMDs, reimbursement models themselves often differ. Instead of individual access-based mechanisms—such as prescription or patient activation—HCP-facing DMDs are typically funded through institutional procurement or bundled payments, including Diagnosis-Related Groups, requiring adapted reimbursement approaches.

More broadly, bundled payment models can and should be advanced to facilitate aligned adoption and use incentives, appropriately embedding all types of DMDs into care pathways as care-improving or cost-saving innovations. Pricing and reimbursement models should also incentivize long-term adherence, as DMD usage tends to decrease after the initial months48,49: linking dynamic HTAs to performance-based pricing should support this, as set forth by Germany’s DigiG.

Systemic adoption and clinical integration

Regulatory frameworks, HTA criteria, and reimbursement pathways are essential but insufficient to achieve systemic adoption. Germany’s DiGA program exemplifies this gap: only 12% of outpatient doctors and psychotherapists prescribed a DiGA, two-thirds of whom prescribed just one or two, while approximately 20% early adopters account for 60% of DiGA prescriptions50. Such dynamics are typical of emergent technologies but simultaneously highlight the need for comprehensive strategies.

While manufacturers should define appropriate go-to-market strategies, policies must ensure adequate infrastructure, interoperable health information systems, and widespread connectivity. Seamless integration with HCP’s and patients’ Electronic Health Records is needed for adoption and RWD value generation—aligning with the EU Digital Decade51 goals and the EHDS.

Embedding DMDs in clinical guidelines, or expert consensus where evidence is still evolving52, is crucial for appropriate clinical integration. Policies should incentivize evidence generation and the integration of DMDs into care protocols. Expedited assessment and reimbursement should be considered for DMDs that operationalise or digitalise established clinical guidelines, enabling their structured and timely implementation in practice. Engaging digital health experts within medical societies, patient associations and policy forums can drive these processes, positioning DMDs as integral to a digitally-enhanced standard of care.

Despite their potential to improve care, robust economic and social impact data on DMDs remains limited53. Research on organizational optimization, patient and HCP empowerment, equity, cost-effectiveness, and sustainability should be incentivized.

Policies should also embed digital competencies in continuing and undergraduate education for all stakeholders and broad multistakeholder engagement—including physicians, nurses, pharmacists, patient associations, ethical committees, administrators, researchers, manufacturers, policymakers and digital health experts—is vital. Public-private partnerships can further drive scalability and sustainability.

Concluding considerations

The five policy enablers outlined in this article—clarifying DHTs in scope of dedicated policies, establishing AI-ready regulatory frameworks and Accelerated Access Pathways, adopting fit-for-purpose dynamic HTAs, implementing dedicated reimbursement models, and fostering system readiness—are not only essential but also highly complementary and mutually reinforcing. Dedicated reimbursement models for DMDs can incentivise higher-quality evidence generation, which in turn supports dynamic HTAs addressing the iterative nature of software updates and the complexities of AI. Dynamic HTAs can bridge post-market evidence with regulatory oversight, enabling the evolution of AI-ready frameworks that ensure safe, timely, and adaptive approval of emerging innovations. Simultaneously, system readiness initiatives, promoting interoperability, education, and clinical integration, would enable adoption, real-world application, and RWE, generating the data needed to refine policies and generate systemic data-driven value.

However, translating these policy enablers into practice can present challenges. It will require visionary policies, sustained commitment, deliberate and concerted ecosystem efforts, strong cross-country cooperation, multistakeholder engagement, and time and resources. Differences in national priorities, regulatory capacities, healthcare infrastructure, and resource availability may slow coordinated progress. Digital skills development, integration of DMDs within care pathways, and continuous evidence generation are evolving enablers that demand long-term investment and structured incentives. Real-world validation remains ongoing, and unforeseen barriers to systemic and equitable adoption may emerge. Ensuring sustainability for both healthcare systems and manufacturers will also be critical for long-term value generation and scalability. Additionally, further application and evidence generation will be crucial to validate and refine these policy approaches over time.

Finally, harmonizing EU-level regulations—and fully realizing the vision of a European Health Union and a digital single market—has been both a mission and a challenge for years. Nevertheless, the need and opportunity to build on pioneering efforts and lessons learned is timely. Successfully innovating complex healthcare systems demands comprehensive policies that drive aligned and effective multistakeholder action across all levels of the health and innovation ecosystems. By rooting these enablers within a comprehensive policy framework and fostering cross-country harmonization, stakeholders can create a virtuous cycle of innovation, access, and equity—scaling the systemic integration of DHTs while ensuring their broad and equitable benefits across healthcare systems.