Objective:To construct an evaluation system for clinical thinking ability of general practitioners in the treatment of multimorbidity.Methods:This was a cross-sectional study. The draft of evaluation indexes for clinical thinking ability of general practitioners in treatment of multimorbidity was preliminary developed through literature review, collation, analysis and discussion. Nineteen clinical and teaching experts of general practice were selected for consultation via anonymous convenient sampling. From January to June 2022, 2 rounds of expert consultation were conducted using the Delphi method. During the first round of consultation, according to the survey feedback, we modified and improved the evaluation system of general practitioners′ clinical thinking ability for multi-disease co-treatment. During the second round, experts were asked to assess the importance of each index, and to calculate the weight of each index accordingly. Questionnaires were sent to experts via letters. The content of the questionnaires encompasses the basic information of experts, evaluation for various indexes and relevant opinions. The mean value of importance assignment ≥3.5, coefficient of variation ≤0.25 and the full score frequency ≥30% were taken as the criteria. Indexes unsatisfying the criteria were removed, so that the final index system could be constructed.Results:The average age of 19 experts was 50.2 years old, 9 of them were male. A total of 2 rounds of expert consultation were conducted, 19 questionnaires were issued in each round, and 19 effective questionnaires were received afterwards. In the first round of consultation, 10 experts put forward revised opinions, and some indexes were adjusted according to the definition criteria and the discussion of the research group. In the second round of consultation, 3 experts put forward suggestions for modification. According to the definition criteria, no need to delete the indexes. After discussion by the research group, some indexes were adjusted, and finally an evaluation system of clinical thinking ability for multi-disease co-treatment of general practitioners was established, including 4 first-level indexes and 30 second-level indexes. The weights of the 4 first-level indexes in descending order were "overall thinking ability" (38.01%), "diagnostic thinking ability" (33.96%), "evidence-based thinking ability" (14.75%), and "critical thinking ability" (13.28%). Among the 30 secondary indexes, the top 5 were "ability to identify and handle priority emergency incidents" (5.04%), "risk assessment and critical illness identification ability" (4.63%), "emergency referral ability" (4.61%), "communication and expression ability" (4.57%), and "standardized diagnosis and treatment ability" (4.23%).Conclusion:This study successfully constructed an evaluation system for clinical thinking ability of general practitioners in the treatment of multimorbidity.