By interviewing 273 department managers from 139 county-level hospitals (105 public and 34 private) in Guizhou using the D-WMS instrument, to our knowledge, we are the first to use an internationally validated [24] survey tool to measure management practices in county-level hospitals in China. Our study therefore adds data on hospital management practices in a third LMIC, enabling us to identify which management capacities require improvement and to compare our Chinese data to that of other countries.
Previous Chinese studies have used tools based on the WMS, including the WMS-Hospital (WMS-H) Chinese version [17], the Chinese Hospital Management Survey (CHMS) [4, 18], the Development-CHMS (D-CHMS) [5], and the Hospital Management Practice (HMP) rating scale [6, 25]. These included 17 (HMP [6, 25]), 20 (WMS-H [17] and CHMS [4, 18]) or 21 (D-CHMS [5]) practices, which were similar to those included in the WMS—but modified for the Chinese context—across four dimensions. While comparisons between studies should be undertaken with caution, especially given differences in the scoring systems, how the questions were asked, and the study years, scores from the earlier studies [4–6, 17, 18, 25] were generally higher than in the present study. This is likely because the earlier studies included only tertiary hospitals [17] or a mixture of secondary and tertiary hospitals [4, 5, 18]. After accounting for these differences in hospital types, our scores were generally comparable to those from the studies that mainly used Chinese-adapted tools [4–6, 17, 18]. However, we feel that our international tool (D-WMS) is more useful for comparing results to those from other countries as we used a standardized tool that not specifically adapted to Chinese context.
In our international comparison, the sampled county-level hospitals in southwest China ranked seventh overall among 10 countries examined – below all of the HICs except France and above the other two LMICs (Brazil and India). These results show that management practices in rural China need to be improved to reach the standards found in HICs. However, it should be noted that our hospitals had some important differences relative to those in the other countries [23]: they were regional rather than national; they had fewer staff (mean: 285 versus 558 to 2344); and they were lower-level (no teaching hospitals versus 9–42% teaching hospitals) [12]. As previous studies have shown that management scores are positively associated with hospital size [14, 26] and development level [24], this likely overstated the between-country differences.
However, we found that our Chinese hospitals achieved higher scores than those in the other LMICs, overall and in all of the domains. Our results can also be compared to those from a study in Kerala, India [3], which like us, used the D-WMS. Our data showed better management practices in Guizhou than in Kerala, overall and in each of the four dimensions, with the largest difference between Guizhou and Kerala in target-setting [3]. While most of the countries in our international comparison scored worst in personnel management, Guizhou scored worst in target-setting. Target-setting was the key weakness of hospitals in China, though with large variation across hospitals. Another recent Chinese study, which surveyed 95 county hospitals across rural China, has also reported lower scores for target management than for efficacy, talent, and performance management (39 vs 53–56 on a scale of 0–100) [25]. Poor target management has also been observed in India, in hospitals, retail, and schools [24]. Brazilian hospitals have also performed poorly in target-setting. Those goals stated by our interviewees were rarely set based on internal factors in ways that would enable their use to measure realistic progress from year to year. Disappointingly, quality of care and patient outcomes were generally not priorities. Moreover, hospital goals were typically short- or medium-term rather than long-term, with little prospect of becoming long-term goals. As setting targets is fundamental to management, we suggest that county-level hospitals in China prioritise target-setting. This could involve defining clear and measurable short-, medium-, and long-term clinical, efficiency, financial, and operational goals for individuals and departments. As setting and tracking performance was a part of the Chinese government’s hospital reform, further study addressing the reform’s impacts on target-setting and attainment at hospitals would be valuable.
Our sample scored poorly in ‘retaining talent’, ‘attracting talent’, and ‘rewarding high performers’. As county-level hospitals are crucial in the Chinese healthcare system and talent is critical to organizational performance, we strongly suggest that Chinese hospitals actively improve managerial expertise in attracting, retaining, and rewarding staff. That said, Guizhou ranked fourth in personnel management, behind only the US, Sweden, and Germany. The sampled county-level hospitals’ best scores were for ‘managing talent’ and ‘removing poor performers’. There are currently two talent-recruiting and managing systems in Chinese public hospitals: one is strictly controlled by the local human resources authority, which decides the number and types of staff that the hospitals need; the other is run by hospital staff, who can recruit additional staff if necessary. Hospitals also have more autonomy to dismiss staff within the second system. As the survey did not distinguish between these two systems, it might have overestimated the autonomy and capability to recruit and discharge personnel.
The worst process scores were for ‘implementation’, particularly in personnel management, implying that many advanced management theories/tools were not being used. Interviewees acknowledged that departing staff members’ reasons for leaving were rarely sought and that they lacked criteria to distinguish good performers. This shows that management training must be improved in rural China. We found that most hospital mid-level managers had a clinical background and few had management training. Although having physicians in leadership positions is valuable for hospital performance [20, 27], a mixture of clinical and managerial skills may also have a positive impact on hospital management quality and clinical outcomes [23]. Further, the interplay between management education and policy reform is well-known [29, 30]. Hence, we feel that providing training to physician managers in modern management thinking and techniques could improve hospital management. While monitoring such an impact with a survey such as ours could be labor-intensive and costly, it may help to provide a deeper understanding of hospital management.
The main limitation of our study is that our sample only included county-level hospitals in western China, so results may not be generalizable to other Chinese regions or other hospital types. However, we believe that our findings are likely representative of many other LMICs, and highlight the need to improve management in such countries. While we included all public hospitals in the region, we only sampled a minority of private hospitals, so our results are likely less applicable to private hospitals. Lastly, although management quality has been correlated with hospital outcomes in other countries [2, 3, 14, 31–33], it would still be beneficial to examine the link between these factors in the Chinese context.