Reimbursement policies for digital therapeutics: A bibliometric analysis of research trends, policy frameworks, and evidence gaps

Creative Commons License

Küçükkurt A. C., Çelik Y.

European Health Management Conference 2026, Barcelona, İspanya, 10 - 12 Haziran 2026, (Özet Bildiri)

  • Yayın Türü: Bildiri / Özet Bildiri
  • Basıldığı Şehir: Barcelona
  • Basıldığı Ülke: İspanya
  • Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
  • Acıbadem Mehmet Ali Aydınlar Üniversitesi Adresli: Evet

Özet

Context: Digital therapeutics (DTx) are health technologies delivering evidence-based interventions via software. DTx use and reimbursement discussions are visible in high-burden areas, particularly mental health and musculoskeletal disorders. Germany's Digital Health Act (DVG, 2019) defined "prescription digital health applications" and institutionalized reimbursement through the DiGA framework. France's PECAN approach emerged in 2023–2024 for early reimbursement/access mechanisms.

Bibliometric studies mapping DTx reimbursement literature remain limited. This study aims to reveal the bibliometric profile, thematic evolution, and evidence gaps in DTx reimbursement literature. It seeks to provide a methodological contribution by evaluating AI tools' alignment with human decisions in systematic screening.

Methods: Studies published between January 1, 2015, and December 31, 2025, were searched in Web of Science and Scopus using keywords related to digital therapeutics (DTx) and reimbursement. Inclusion criteria comprised peer-reviewed research articles or reviews addressing DTx reimbursement, pricing, or market access themes. Exclusion criteria included conference abstracts, letters to the editor, and general DTx studies not addressing reimbursement.

Title and abstract screening was conducted in two phases: (1) title screening (n=149) with three-category coding (eligible, ineligible, abstract review needed); (2) abstract screening (n=112) with binary coding (include, exclude). Human researchers' decisions were compared with outputs from three AI tools; Cohen's Kappa, Fleiss' Kappa, sensitivity, and specificity metrics were calculated.

The final dataset (n=70) was analyzed using Bibliometrix/Biblioshiny software. Analyses included annual production trends, Bradford's Law, Lotka's Law, logistic growth model, most productive authors/institutions/countries, citation analysis, keyword co-occurrence network, thematic mapping, and collaboration network analyses.

Results: The literature comprises 70 articles spanning 2019–2025. The annual growth rate was 61.89%; a 350% increase in publications occurred in 2021, driven by DiGA regulations. By country, Germany dominated with 43.7% of production (author contributions: n=118), followed by the USA (17.4%) and South Korea (6.7%). Regarding institutional productivity, Witten/Herdecke University (n=14) was the most productive institution; Journal of Medical Internet Research (n=8) the core journal.

Thematic map analysis revealed that "reimbursement" and "digital therapeutics" clusters were positioned in the "basic themes" category, indicating evidence gaps in the literature concerning central topics.

Collaboration network analysis identified 12 isolated clusters; international collaboration rate remained at 15.71%.

In AI agreement analysis, agreement with human decisions at the abstract screening stage was κ=0.9238 (almost perfect) for AI_1, κ=0.6923 (substantial) for AI_2, and κ=0.4098 (fair) for AI_3. At the title screening stage, all AI tools underutilized the "abstract review needed" category reflecting uncertainty.

Discussion: This study reveals that DTx reimbursement literature is growing but exhibits a geographically concentrated and thematically fragmented structure. Germany's regulatory framework through the DiGA model has shaped the literature, highlighting the interaction between policy design and academic production. Increasing interest in the Asia-Pacific region (particularly China's quantitative rise in 2025) suggests the literature may diversify geographically.

Findings indicate evidence gaps in the field. These include long-term cost-effectiveness evaluations, real-world evidence studies, comparative analysis of reimbursement models, and systematic examination of relationships between patient outcomes and reimbursement decisions. Deviation from Lotka's Law (single-article author ratio: 91.57%) indicates the field's early maturity level.

Rapid DTx technology evolution necessitates that reimbursement and assessment bodies develop dynamic, transparent, and evidence-based decision frameworks. Methodologically, AI-assisted screening demonstrates high reliability at the abstract screening stage (κ=0.9238 for AI_1); however, human oversight maintains its decisive role in classification decisions involving uncertainty.