This section presents how the aforementioned method is used to develop the performance measurement system for evaluating general practitioners’ offices in China.
3.1 Index pool for evaluating general practitioners’ office in China
Based on the literature research method, the titles and abstracts of relevant papers were summarized and categorized according to the purpose and theoretical basis of this research. A total of 44 indices were finally selected.
In the cross-sectional research, a total of 1,917 questionnaires were collected from medical personnel from 21 cities and prefectures in Sichuan Province, in which 1651 valid questionnaires were identified, and the effective rate was 86.12%. Among the respondents, 1103 (66.8%) were female, and 625 (37.9%) were 36 to 45 years old.
Based on the results of the cross-sectional research, a total of 12 indices were identified, including "appearance of the office", "office size", "facilities & equipment", "interior layout", "staffing", "team building", "content of service", "construction of information system", "appointment service", "referral service", "management of health profiles", and "follow-up service", which are incorporated into the index pool.
After the integration of the results of the literature research and cross-sectional research, 56 index items were initially screened out, and 13 index items with repetitive significance were eliminated. Finally, 40 indices were selected to form the performance measurement system. The index pool is sorted out as follows:
1) Structural dimensions (15 indices): "appearance of the office",24 "construction of information system", "team building",25 "sources of funding",26 "satisfaction",24 "naming", "office size", "facilities & equipment",24 "interior layout", "staffing", "regional healthcare information system",27 "mode of operation", "culture building of team",28 "capacity building of team", "government input".
2) Process dimensions (16 indices): "basic health care", "contract service of family doctor",29,30 "public provisioning of health services", "collaborative community-based services",31 "quality of service", "telemedicine services",32 "synergy of government",33 "general medical services", "emergency medical services",34 "health education and advisory services",35 "health management service", "appointment service",36,37 "pharmaceutical delivery", "outpatient services", "medication guidance", and "service for long-term prescriptions" 35.
3) Result dimensions (11 indices): "income", "income of basic health care", "income of public provisioning of health services", "effective contract rate", "compliance rate", "rate of contract renewal", "contract rate for key populations", "rate of hypertension control",38 "rate of diabetes control",38 "satisfaction of medical staff", "client satisfaction".39
3.2 Preliminary performance measurement system
To establish a preliminary performance measurement system, three single-focus group discussions were conducted in this stage. The first and second focus groups were conducted online, and the third group was conducted on-site. There were 27 focus group members, 16 of whom were male (59.26%) and 15 of whom were over 40 years old (55.55%).
The first discussion lasted 90 minutes with 15 participants. Three first-level indices were determined, including "essential requirement", "health services", and "quality assessment". In addition, the second-level indices corresponding to each first-level index were integrated. The second discussion consisted of 11 participants and lasted 73 minutes. After sorting out the data, 12 second-level indices and 35 third-level indices were preliminarily developed by the research team. The third discussion lasted 90 minutes with 10 participants, and indices at all levels were revised. The preliminary performance measurement system is shown in Appendix 1.
3.3 Modified performance measurement system
An expert team was formed to modify the preliminary performance measurement system, which consisted of 13 consultants from different medical institutions in Chengdu, Shanghai, Chongqing, Beijing, and Shijiazhuang, with 10 experts (77.0%) over 40 years old, the longest working time of 45 years, an average of 21.46 years, and 7 experts (53.8%) with a master’s degree or above. After three rounds of discussion through the Delphi method, the performance measurement system of the general practitioner’s office was finally constructed, which contained 10 second-level indices and 37 third-level indices. The 10 second-level indices included "appearance of the office", "construction of information system", "team building", "operational mechanisms", "basic health care", "contract service of family doctor", "collaborative community-based services", "quality of service", "social assessment", and "economic efficiency". The 37 third-level indices are shown in Appendix 2.
The motivation, authority and coordination of the experts were tested after each round of questionnaire return. After calculation, the positive coefficients of the experts in the three rounds were all greater than 0.7, indicating a high positive degree, and the results were reliable. The authority coefficients (Cr) of consultants in the first, second and third rounds were 0.9424, 0.9003 and 0.9499, respectively. The authority coefficients (Cr) were all greater than 0.7, indicating that the authority of consultants was strong and that the results can be accepted. All three rounds of coordination coefficients showed a high degree of coordination of expert opinion after chi-square tests, which indicated the reliability of the results.
3.5 Assessment of credibility
We selected three representative general practitioners’ offices with good primary care facilities and services from three cities (i.e., Chengdu, Bazhong, and Deyang) in Sichuan Province, China, as case offices to conduct the field study and to verify the proposed performance measurement system.
The results of this research showed that the first office scored 958.5 points, the second scored 768.1 points, and the third scored 947.7 points. According to the scores of the three offices, office 1 and office 3 are relatively high, while office 2 is relatively low. Their differences are mainly reflected in indicators such as regional healthcare information system, mode of operation and incentives. According to the analysis, office 2 is located in northeastern Sichuan, where the economic level is relatively low, and information resources are relatively insufficient. Office 1 is in Shuangliu County of Chengdu. The informatization construction and working model of this county are relatively good, which is basically consistent with the results of the performance using the measurement system developed in this research, indicating that this system can better evaluate the construction of general practitioner’s offices in primary care institutions. In summary, the performance measurement system is effective and manoeuvrable.