Evaluation of the One ID Anywhere Policy of the Sawanpracharak Hospital, Nakhon Sawan Province

Abstract

Background: The "30-Baht Treatment Anywhere with a Single ID Card" policy, launched in 2024, leverages digital health to challenge equitable access to care. Policy success, however, depends not only on its design but on the implementation capacity of service facilities.

Objective: To assess determinants of policy adoption and implementation through changes in outpatient service utilization at the Sawanpracharak Hospital pilot site.

Methods: A convergent parallel mixed-methods design was employed between May–August 2025. The qualitative strand used purposive sampling to recruit 14 staff from 7 professional groups for individual in-depth interviews via a semi-structured interview guide developed from the consolidated framework for implementation research (CFIR) version 2.0. Thirty-five qualitative of 43 CFIR constructs, with quantitative weekly time-series data for outpatient visits were analyzed for type 2 diabetes (E11) and hypertension (I10–I15) patients, comparing 2024 (36 weeks) against a 2019–2023 (240 weeks) baseline using one-sample t-test, Welch's t-test, Cohen's d, Bootstrap 95% CI (10,000 resamples), and Wilcoxon signed-rank test.

Results: Qualitatively, 71.4% of the assessed constructs (25/35) were facilitators of policy implementation. Key enablers included adaptability (+2), leadership engagement (+2), staff self-efficacy and stage of change (+2), and structured implementation processes including planning, executing, and reflecting (+2). Four structural barriers (–2) were identified: evidence quality, cost, available resources, and workflow compatibility. Quantitatively, outpatient utilization in 2024 increased significantly. Diabetes visits rose +19.8% (from 870 to 1,042 visits/week; Cohen's d = 1.040, p < 0.001) and hypertension visits rose +16.9% (from 1,450 to 1,695 visits/week; Cohen's d = 0.896, p < 0.001), both exhibiting large and robust effect sizes confirmed by sensitivity analyses. However, I10–I15 showed non-significant increase from the 2-year baseline (p = 0.100), suggesting a pre-existing secular trend.

Conclusion: The implementation success reflected the alignment between organizational capability—people and processes—and structural support at the policy level. The four main barriers lay beyond hospital control, the nationwide scale-up required prior fix on reimbursement gaps, operational costs, and compatible digital system–workflow. Nevertheless, the single-site design limited generalization including high digital capacity of the pilot hospital. The quantitative analysis of aggregate-level data could not effectively control for secular trends or external confounders.