Scaling-up through piloting: dual-track provider payment reforms in China’s health system

Abstract

Gaining wide prominence in the global health arena, scaling-up increases the coverage of health innovations emerging from pilots and experimental projects to a larger scale. However, scaling-up in the health sector should not follow a linear 'pilot-diffusion' pathway in order to better facilitate local adaptation and policy refinement. This paper puts forth 'scaling-up through piloting' as a distinctive pathway for the strategic management of scaling-up in the health sector. It analyses the recent development of provider payment reforms in China, focusing particularly on the ongoing pilot programmes, namely diagnosis-related groups (DRGs) and diagnosis-intervention packet (DIP), that are being piloted in a dual-track fashion since 2020. Data were drawn from extensive documentary analysis and 20 in-depth interviews with key stakeholders, including decision-makers and implementers. This paper finds that scaling-up through piloting helps Chinese policymakers minimize the vast uncertainties associated with complex payment reforms and maximize the local adaptability of provider payment innovations. This pathway has forged a phased implementation process, allowing new payment models to be tested, evaluated, compared and adjusted in a full spectrum of local contexts before national rollout. The phased implementation creates a 'slower is faster' effect, helping reduce long-term negative consequences arising from improperly managed scaling-up in a complex system. Error detection and correction and recalibration of new policy tools can support national-level policy refinement in a more robust and dynamic fashion. Several key factors have been identified as crucial for strategic scaling-up: necessary central steering, a pragmatic piloting design, strong technical capacity and effective policy learning mechanisms.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 55 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.