JMIR Publications

ABSTRACT

Background:

The literature on value-based healthcare (VBHC) implementation has expanded rapidly but remains heterogeneous, making synthesis difficult. This study aimed to map the thematic landscape of VBHC implementation and identify recurring domains, cross-cutting patterns, and gaps in the evidence.

Objective:

To map and synthesize the thematic landscape of the value-based healthcare (VBHC) implementation literature using computational thematic synthesis, identifying recurring domains, cross-cutting patterns, and evidence gaps.

Methods:

PubMed was searched without date restrictions and limited to English-language publications. Screening and full-text eligibility assessment were conducted manually in two stages. Eligible full texts were embedded using a text-embedding model (text-embedding-3-large, 3,072 dimensions), projected with UMAP, and clustered with HDBSCAN using consensus across 10 random seeds. Cluster stability was examined using an alternative embedding model, direct-space clustering, and negative controls.

Results:

A total of 8,957 records were screened, and 351 full-text articles were included. Computational synthesis identified 18 themes that grouped into four broader domains: condition-specific VBHC applications, measurement and costing approaches, organisational and workforce models, and health system policy and governance. Commentaries and framework papers were the most common design category (19.4%), while empirical studies were comparatively fewer across most themes. The United States and the Netherlands contributed the largest share of publications. Themes supported by stronger measurement infrastructure, including PROMs, ICHOM outcome sets, and TDABC, were more frequently associated with reported value improvements.

Conclusions:

This evidence-mapping study provides a broad overview of the VBHC implementation literature. Despite wide diffusion of the concept, much of the field remains centred on conceptual discussion rather than empirical evaluation. Future work should prioritise interventional designs and context-sensitive implementation frameworks, particularly in under-represented regions.