Abstract
BACKGROUND: General practitioners’ (GPs’), attitudes to and acceptance of GP-CDSS-2.0, and the influence of GP-CDSS-2.0, a clinical decision support system for the anticoagulation treatment of non-valvular atrial fibrillation (NVAF) in primary care, on GPs were also essential to further promote GP-CDSS-2.0 after the study. The aim of the study was to identify the effect of applying GP-CDSS-2.0 in community health centers on GPs’ anticoagulation treatment knowledge-attitude-practice (KAP) for patients with AF, and to understand GPs’ perceptions and using experience of GP-CDSS-2.0. METHODS: This study was a part of the whole cluster randomized controlled trial, targeting GPs, adopting a mixed method, using both quantitative and qualitative studies. It lasted from July 2022 to July 2023, conducted in 14 community health centers in Baoshan District and Jing’an District of Shanghai, China (but one community health center withdrew from the study during primary outcome analysis period). The outcome of the quantitative study was GPs’ anticoagulation treatment KAP for patients with AF, assessed by the KAP questionnaire developed by our team in previous work. Before the end of the follow-up, we interviewed GPs in the 7 community health centers of the intervention group based on the principle of saturation and covering all 7 community health centers. We recorded GPs’ attitudes to and using experience of GP-CDSS-2.0, and analyzed interview documents by thematic analysis. RESULTS: In the quantitative study, we enrolled 285 GPs, with 126 in the intervention group and 159 in the control group. There were no statistical differences in the baseline demographic characteristics of the two groups’ subjects. At the end of the study, the proportion of GPs at a “good” level assessed by the KAP questionnaire in the intervention group and the control group were 60.3% and 45.3%, respectively. The intervention group’s average scores on all three dimensions were higher than those of the control group. The KAP scores in the intervention group were significantly increased during the study (score before training minus score at the end of follow-up)=-10.45[Formula: see text]1.802, p < 0.001), but in the control group not (difference=-2.92[Formula: see text]1.605, p = 0.208). At the end of the study, after adjusting differences at baseline, the average score in the intervention group was at a “good” level and higher than that in the control group (the intervention group = 109.12[Formula: see text]16.413, the control group = 103.23[Formula: see text]18.826, F = 9.944, p = 0.002). In the qualitative study, we interviewed 7 GPs. Forty-four coded items and four themes emerged after analyzing the interview documents. The four themes were as follows: frequency of GP-CDSS-2.0 use and influencing factors (2), user experiences with GP-CDSS-2.0 and comprehensive geriatric assessment (CGA) features (3), suggestions for GP-CDSS-2.0 improvement, and (4) needs and recommendations for AF management in primary care. All seven interviewees expressed that GP-CDSS-2.0 was helpful for their daily practices. CONCLUSIONS: Implementation of GP-CDSS-2.0 in China’s primary care settings effectively maintained GPs’ AF anticoagulation KAP at a “good” level, with GPs in the intervention group expressing favorable attitudes toward the system. TRIAL REGISTRATION: Registration Number: ChiCTR2100052307; date of registration: Nov 22 2021. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12875-026-03205-9.