The growth of private hospitals and their health workforce in China: a comparison with public hospitals

Transcription

1 Published by Oxford University Press in association with The London School of Hygiene and Tropical Medicine ß The Author 2013; all rights reserved. Advance Access publication 17 January 2013 Health Policy and Planning 2014;29:30 41 doi: /heapol/czs130 The growth of private hospitals and their health workforce in China: a comparison with public hospitals Chengxiang Tang, 1,2 Yucheng Zhang, 3 Lieping Chen 2,4 * and Yongqiang Lin 5 1 School of Economics, Australian School of Business, University of New South Wales, Kensington, Sydney, 2052 NSW, Australia, 2 Department of Health Management and Policy, School of Public Health, Fujian Medical University, Fuzhou, Fujian , People s Republic of China, 3 School of Management, Australian School of Business, University of New South Wales, Kensington, Sydney, 2052 NSW, Australia, 4 Fujian Center for Prevention and Control of Occupational Diseases and Chemical Poisoning, Fuzhou, Fujian, , People s Republic of China and 5 Health Department of Fujian Province, Fuzhou, Fujian , People s Republic of China *Corresponding author. School of Public Health, Fujian Medical University, 88 Jiaotong Road, Fuzhou, Fujian , People s Republic of China. clpqsc@126.com Accepted 16 November 2012 Background: China significantly opened its healthcare market through a series of marketopening policies in This study aims to explore the direct consequences of these policies the growth of private hospitals, their workforce characteristics compared with public hospitals in China and the source of their healthcare workforce. Methods: Results: Conclusion: Keywords First, we performed a segmented regression analysis of a longitudinal data series of the number of hospitals in China between 1990 and 2009 to examine the before and after effects of the market-opening policy on private hospitals. Then, to highlight the workforce differences between private and public hospitals, provincial survey data collected in 2009 were compared with data from a second database collected in 2002 to detect the mobility of medical staff from the public to the private hospitals. The number of private hospitals rapidly increased after 2001, and the yearly growth rate increased from 19 to 205, represented primarily by an increase in specialty hospitals. Approximately 22.03% of the physicians in private hospitals are over the age of 60, whereas this proportion in public hospitals is only 2.97%. In 2008, at least 4.1% of the staff working in private hospitals had previously worked in local public hospitals in The broad expansion of private hospitals since 2001 is most likely the result of an unbiased market policy environment for private hospitals. Moreover, specific features of the hospital physician relationship in China may account for the unbalanced age distribution feature among doctors and the mobility of the healthcare workforce in private hospitals. Market-opening policy, private hospital, human resources for health, hospital physician relationships, China 30

2 GROWTH OF PRIVATE HOSPITALS IN CHINA 31 KEY MESSAGES A higher year-to-year growth rate in specialty hospitals has been detected since 2001, when a set of market opening policies were implemented, resulting in a rapid increase in private hospitals in China. When measured, the age distribution of physicians and nurses between public and private hospitals was found to be unbalanced. To a degree, the unique hospital physician relationships found in China can explain the unbalanced age feature and mobility of the healthcare workforce in private hospitals. Introduction China has been experiencing rapid economic and social development since the early 1980s, especially in the private sector. The re-legalization of private medical practice in urban areas began in 1985 after being banned for nearly 30 years; the ban was lifted after criticisms of the Chinese healthcare system s inefficiency and poor quality. A major component of the reform is expanding the healthcare industry to the private sector to increase the number of competitors in this industry. Before reform, all the hospitals in China were public, and there were different healthcare systems in urban and rural areas. In urban areas, a three-tiered system was used that included county hospitals, district hospitals and city hospitals. Similarly, a three-tiered rural health structure was composed of village clinics, township health centres and county hospitals (Eggleston et al. 2008). Under planned economic institutions, the price of health care was firmly in the control of the central government, although both central and local governments would provide funds for public health services. With the reform of China s economy, this system changed in the early 1990s with an increase in the number of private hospitals during this period (State Council 1997). At the same time, the government realized that there should be regulation and an official classification of different types of hospitals. However, private hospitals were still not considered a distinct type of hospital. Although private clinics have spread throughout rural areas and most urban areas since the 1990s, no large-scale growth in private hospitals occurred because of the lack of an appropriate policy environment (Gu and Zhang 2006). The Basic Standards of Healthcare Facilities (Draft) and Key Principles of Distribution of Healthcare Facilities, 1 which were issued in 1994, did not mention how to absorb the private healthcare facilities into the overall government plan (Lim et al. 2002). In , the central government began to categorize all medical facilities into two groups, not-for-profit or for-profit organizations (State Council 2000), which first officially admitted the entry of private capital and the profit motive into the healthcare industry. In addition, a series of policies regarding tax preferences, the management of healthcare personnel, the assessment of private healthcare facilities and healthcare insurance contracts that aimed to emphasize the necessary role of private medical facilities and encourage private investment in the healthcare sector were implemented in 2001 (Sun 2001). Together, these policies constitute the Market-opening Policy. According to a previous study (Eggleston et al. 2008), the use of the number of inpatient beds or outpatient visits is not an appropriate measure of the private market size yearly growth rate because it is a formative indicator that fails to consider the growth in the number of hospitals. Specifically, most private hospitals in China exist on a much smaller scale than the public hospitals; for instance, in China in 2008, the average number of inpatient beds for a private hospital was 42, which is in sharp contrast to the average number of 228 beds for a public hospital. Thus, it is meaningless to measure the yearly growth rate of private hospitals by inpatient bed number or outpatient visits in the Chinese context. However, clear evidence regarding the effects or consequences of a market-opening policy has rarely been provided. In this study, we performed an analysis of routine longitudinal data to examine some of the outcomes of the growth in the number of private hospitals following the implementation of these policies. The increase in the number of private hospitals has attracted a large number of medical staff. What is the situation with respect to the private hospitals and their workforce after recent growth, especially compared with the public sector? Several current studies have explored the appropriate role and features of private healthcare providers in China (Liu et al. 2006, 2009; Huang et al. 2009), but data regarding the profile and basic characteristics of private healthcare providers in hospitals remain limited. This article attempts to address this lack of information by comparing the healthcare workforce of public and private hospitals based on a provincial survey of human resources information for the health sector. The empirical evidence in this study can contribute to the study of the development of private hospitals in transitional and developing countries. This article contains the following sections. The Methods section defines the private hospital, presenting a working concept of private and providing a detailed description of data collection. The results on the structure and features of the workforces of public and private hospitals are presented in the next section. The Discussion section explains the development trends of Chinese private hospitals in comparison with similar experiences from Central and Eastern Europe. The discussion also presents some potential reasons for the unbalanced age pattern found in private hospitals according to a framework of hospital physician relationships and examines the possible effects of having a larger proportion of elderly doctors providing health services. In the final section, we present a summary of the major findings and make suggestions for new policy implications. Methods Data collection Data were collected from two sources. The first source was the national bureau of statistical publications; the second source was a provincial survey of human resources for health care

3 32 HEALTH POLICY AND PLANNING conducted between August and October 2009 as part of an overall investigation of the health system for the purpose of health sector reform in Fujian province. Survey The cross-sectional survey conducted as part of this research was developed in three stages. First, experts and health department officers were gathered to discuss the questionnaire design and the survey arrangement based on findings from a literature review and previous surveys. Human resources working for health organizations in Fujian province until 31 December 2008 were chosen as participants. To ensure complete understanding and correct feedback from different districts questionnaires, the items and options were modified and adjusted following two group discussions and pilot testing in a small area. In the second stage, all local health officers for human resources who were responsible for implementation of the survey in each county were informed by the Health Bureau of Fujian Province. The basic units for the survey were state-owned hospitals, private hospitals, rural township hospitals, maternity and childcare centres, community health centres and the Centers for Disease Control and Prevention and other public health organizations in both urban and rural areas. The questionnaires were downloaded from the website of the provincial Health Bureau and were answered by staff specifically assigned this task in nearly every facility under the supervision of the county health officers. Several trained assistant directors were in charge of explaining and clarifying materials for facility staff when the purpose of the survey or items was unclear; communication was conducted by telephone and . Health offices in every county collected their data and returned it to the provincial Health Bureau via web or express delivery. The third stage was data correction, entry and analysis. Incorrect and missing information were found by quality control measures; 20% of the original questionnaires from each county were scrutinized to check for errors. After returning to the investigation area to rectify any problems, data were re-entered using Microsoft Excel Questionnaire The questionnaire for human resources in healthcare organizations consisted of two sections. Section one consisted of explicit instructions for completing the questionnaire, whereas section two was divided into two subdivisions: the heading and the survey. The heading collected the background information for each health institution, such as the administrative division code, chief of unit, informant name, contact telephone number, organization name, public or private, outpatients and emergency patients in 2008, hospital admissions in 2008, total revenue (1000 yuan) in 2008 and the total value of medical equipment ( yuan) in The survey requested information on demographic and professional practice for human healthcare resources in the facility. These items included staff number, name, birth date, gender, management and support staff, highest level of education, health professionals of practice, qualification title of health professionals, whether staff had a formal contract (only for public organizations), and whether staff had been recruited after A specific system of nomenclature was used to describe healthcare workers in China that largely follows international definitions. We used these definitions of the health workforce to classify the professional practice information of participants according to four items in the study: (1) Practising health professionals: Western/modern medicine, traditional Chinese medicine (TCM, which also includes integrated traditional and western medicine), dentistry, nursing, public health, pharmacy, laboratory technology, other health technology (clinical radiology technology) and non-health professionals (e.g. information and financial professions). (2) Qualification title of health professionals: title of chief (given after 5 years of assistant qualification), title of assistant chief (given after 5 years of professional in charge title or 2 years of doctoral degree qualification), title of professional in charge (given after 5 years of junior title or 2 years of master s degree qualification), junior professional (medical graduates with a bachelor s degree and an internship of 1 year who passed the licensing examination), junior assistant professional and staff with no title (other workers or graduates who have not passed the licensing examination). These qualifications are derived from a licensing exam system for monitoring the quality of health professionals. (3) Highest level of education: doctoral level education, master s level education, undergraduate college, junior college or higher vocational education, technical secondary school, high school or below. 2 The health professional s response should be based on the highest level of full-time education achieved. Measuring the increase in private hospitals Before presenting any information related to the characteristics of the private sector or any analysis of its development, it is first necessary to clarify the exact meaning and scope of private hospitals in China. Healthcare market providers are usually classified into three categories of ownership: state-owned or public, private for-profit and private not-for-profit. Chinese health policy regarding the definition and regulation of for-profit and not-for-profit health organizations was first drafted and implemented in 2000, after which private for-profit hospitals were allowed to operate legally for the first time. Private not-for-profit hospitals appear to be scarce in China (Liu et al. 2006), and even national statistics neglect this segment in their ownership mix. This study considers any healthcare providers falling outside of the direct control of government to be part of the private sector, regardless of their for-profit or not-for-profit status (Bennett 1992). 3 At first analysis, given that the share of government spending dedicated to total health expenditures has been declining since before the health reform initiatives in 2006 and that the privatization or survival of public hospitals has proven to be a significant issue for the policy agenda (Ma et al. 2008; Yip and Hsiao 2008), very little public investment and few new public hospitals were approved by the government. In this study, hospitals were divided into general-acute hospitals, hospitals specializing in TCM and specialty hospitals based on national

4 GROWTH OF PRIVATE HOSPITALS IN CHINA 33 statistical yearbooks. As most current private investors in China have focused on several kinds of profit-generating specialty hospitals (Eggleston et al. 2010), we assumed that the expansion of private hospitals could be represented by the increase in specialty hospitals. Furthermore, the use of specialty hospitals as a proxy to represent private hospitals has both theoretical and practical support. Theoretically, one of this study s aims was to examine whether the reform of healthcare policy leads to the growth of private hospitals in China. According to policy research, in the past two decades, there have been no policies to enhance the growth of public hospital numbers in terms of specialty hospitals. Thus, when we measure the growth in specialty hospitals, it is equal to measuring the growth in private hospitals, even though some private investors enter the general hospital market Appendix 1. Before the policy reform of 2001, there was no official practical definition of a for-profit hospital that was the equivalent of private in China. In publications from the National Bureau of Statistics, we could not find information about private hospitals before 2003, but if we only use data since 2003, we would miss the important years following the launch of reforms. It is not rigorous to admit that specialty hospitals can perfectly represent private hospitals, but it is valid to use the growth of specialty hospitals to represent the growth of the private hospitals. For these reasons, this study uses specialty hospital growth as a proxy for private hospital growth. For the second section of the questionnaire, a dummy variable was created to categorize the private and public hospitals. This variable was derived from a binomial item included in the survey questionnaire that all organizations were required to complete. Analysis We used a segmented regression to analyse the development of private hospitals (Lagarde 2011). This analysis controlled for secular trends to determine whether there was a systematic change in the variables associated with the number of different types of hospitals before and after the implementation of Chinese reform policies. The equation used in the linear regression was: Y t ¼ 0 þ 1 time þ 2 intervention þ 3 post slope þ " t The dependent variable Y t is the number of hospitals at time t. Time is a variable that indicates the year corresponding to the number of hospitals from 1990 to Intervention indicates whether the reform policy was implemented; intervention was coded as 0 (before 2002) or 1 (for 2002 and post-intervention time points) because we considered the effects of learning delays and the time required for the planning, design and administrative approval of new hospitals, even though the critical policy was implemented in The post-slope was coded as 0 before 2002 and was coded sequentially from 1 thereafter. " t is an error term. The coefficient b0 estimates the baseline level of the number of hospitals at the beginning of 1990; b1 is the growth rate before 2002, which is independent of the policy intervention; b2 estimates the immediate effects after the implementation of the market-opening policy in 2002; and b3 estimates the change in trends after intervention. Before analysis, we excluded the effect of first-order auto-correlation by calculating a Durbin- Watson statistic of This study further analysed the healthcare workforce between 145 private hospitals and 231 public hospitals in Fujian province to present some basic characteristics of the private sector that resulted from the rapid expansion of hospitals. During statistical analysis, cross-tabulation using the chi-square test was employed to assess the significant differences in the groups. In addition, to identify the source of the healthcare workers in the private hospitals (to detect the mobility of medical staff from public to private hospitals), we compared this 2009 database with a simpler cross-sectional database collected in 2002 that reported information on the healthcare workforce at the end of A matching analysis using the same name and birth date was conducted to locate those individuals who had worked for public hospitals in 2001 and then for private hospitals. All data analysis involving description analysis was performed using Stata version 10. Results Increasing private hospitals One of the most important factors that China and many other transitional and developing countries in Central and Eastern Europe and Central Asia (CEE and CA) must face is that the hospital system always dominates their health sector. Three kinds of hospitals accounted for the great majority of outpatient (50.45%) and inpatient services (64.37%) in China, even when considering all institutional types offering healthcare provision: clinics, infirmaries and nursing stations (Table 1). Thus, particular attention should be paid to the rapidly increasing number of hospitals in China. Because of market-opening policies for private investors, changes in the types of hospitals, especially the number of total hospitals and the number of specialty hospitals (which are used as a proxy for private hospitals), can be observed following policy implementation in 2002 (Figure 1). Table 2 presents the segmented regression results for four models using the following dependent variables: the number of general-acute hospitals, TCM hospitals, specialty hospitals and total hospitals. The model for specialty hospitals and all of its coefficients was significant. These results indicate that at the beginning of the observation period, the initial average number of specialty hospitals was There was significant year-to-year change in the mean number of specialty hospitals (19 hospitals per year), although it was a small increase. Immediately after the opening of the market, the estimated effect of the policy resulted in an increase of 311 specialty hospitals. The growth rate of specialty hospitals after the policy intervention was significant (205 hospitals per year). This model indicates the rapid growth of private hospitals after the market-opening policy. Moreover, we can observe a significant increase in the total number of hospitals. Immediately following policy implementation, the estimated mean number of total hospitals increased by 646. There were also significant year-to-year changes in the number of hospitals before and after However, the coefficients regarding policy intervention for the other two models (general-acute hospitals and TCM hospitals) are not

5 34 HEALTH POLICY AND PLANNING Table 1 Number of visits and inpatients in health institutions in China in 2009 Health institutions Total visits (100 million person-times) Total inpatients ( persons) n ¼ 35.3 n ¼ Hospital n (%) (50.45) (64.37) General-acute hospital Hospital specialized in TCM Specialty hospital Sanitarium n (%) 0.02 (0.06) 36 (0.31) Community health centres n (%) 2.57 (7.28) 141 (1.23) Health centres n (%) 8.62 (24.42) 3355 (29.22) Urban health centres Rural township hospitals Outpatient Department n (%) 0.51 (1.44) (0.1) Clinics, health centres, nurse stations n (%) 4.24 (12.01) MCH centre (Station) n (%) 1.36 (3.85) (4.53) Specialized disease prevention and treatment institute n (%) 0.18 (0.51) 27.9 (0.24) Source: China Statistical Yearbook Year All Hospitals General TCM Specialty Figure 1 Growth trend in the number of Chinese hospitals from 1990 to Table 2 Results for the segmented regression analysis of the number of different hospitals from 1990 to 2009 Variables General-acute hospitals TCM hospitals Specialty hospitals Total hospitals Time *** (17.90) 29.79*** (2.81) 19.90*** (4.25) *** (19.88) Inter (203.31) (31.89) *** (48.25) * (225.80) Post (37.57) 4.43 (5.89) *** (8.92) *** (41.73) Constant *** (131.75) *** (20.67) *** (31.27) *** (146.33) Adj. R *P < 0.05, ***P <

6 GROWTH OF PRIVATE HOSPITALS IN CHINA 35 significant. In addition, we cannot observe a significant intervention effect for increases in general-acute hospitals or TCM hospitals. The available data do not permit the direct identification of privately owned hospitals over such a long period, but the expansion of private hospitals from another point of view can be observed. The number of for-profit hospitals in China increased from 1792 in 2002 to 4543 in 2009 (Ministry of Health 2003, 2010), a growth rate of 153.5% in just more than 7 years. According to the definition of private hospital, all growth in for-profit hospitals can be attributed to the private sector because governmental policies forbid the involvement of public organizations in for-profit hospitals. Basic characteristics of the private workforce and their source The survey data covered enrollees from 1512 public health facilities that included public hospitals, community health centres, health inspection institutes, CDC, rural township hospitals and 151 private health facilities, of which 145 were private hospitals and the other six were polyclinics. This study was limited to 376 public and private hospitals covering participants. Table 3 compares the number of hospitals analysed with the provincial statistics yearbook. Table 4 profiles the basic characteristics of the human resources serving the three types of hospitals studied in the survey. Private hospitals were generally much smaller than public hospitals, with the mean number of medical staff in a private hospital being 54 compared with 327 in a public hospital. Thus, on average, there were fewer doctors per private hospital (16) compared with doctors per public hospital (98). The gender ratio was biased towards females, with female healthcare workers dominating the hospital system. Most doctors (81.06% in public and 87.04% in private) practised Western or modern medicine, while a minority (18.94% in public and 12.96% in private) still practised TCM in hospitals. Nursing personnel comprised the major portion of the human resources found in the hospital system. The ratio of nurses in the healthcare workforce was 40.79%. In terms of educational background, technical secondary school (the equivalent of 3 years of professional training) was the highest level of education attained by most human resources in public and private hospitals. The proportions were 42.14% (public) and 45.25% (private). Only 0.59% of public and 0.09% of private healthcare workers had attained a doctoral degree. If education is used as a proxy for capacity and skill, then this finding probably represents an inappropriate level of quality among the health workforce; private hospitals have fewer workers, but many with higher levels of professional qualification, than public hospitals. Table 3 Hospital coverage in Fujian province survey in 2008 Public hospitals Private hospitals Total Coverage rate General-acute hospitals (265) 88.68% TCM hospitals (75) 98.67% Specialty hospitals (71) 94.37% Total (413) 91.04% Most healthcare workers in both the public (61.8%) and private (71.14%) sectors were under the age of 40; this may reflect the rapid expansion of the healthcare workforce in recent years. However, 8.44% of the human resources in private hospitals were over 60, compared with just 1.43% in the public sector. Figure 2 shows the age structure of physicians and nurses by hospital ownership, and it can be seen that the private hospitals have a significantly different age distribution. Private hospitals retain a large proportion of elderly doctors (roughly 22.03% are over the age of 60) and a larger proportion of young nurses (roughly 75.49% are under the age of 30); however, these ratios in the public hospitals are 2.97% and 40.75%, respectively (Table 5). In terms of the source of the elderly doctors, in 2008 at least 4.1% of the medical staff (324/7827) in private hospitals had previously worked in local public hospitals in 2001 (Table 6). The mean age of these inflow workers was 51 by the end of Figure 3 shows the age structure of this mobile group by gender. The majority of the group is composed of male doctors and female nurses. Discussion Using segmented linear regression analysis to examine longitudinal data, we showed that with the implementation of market opening policy in China, private hospitals experienced a rapid development after 2001 as measured by the increase in specialty hospitals. Further analysis of data from a provincial survey indicated that elderly doctors and young nurses made up a substantial portion of the healthcare workforce among the growing number of private hospitals. The growth of private hospitals Similar stories have emerged in many transitional CEE and CA countries during the last two decades. The hospital privatization process in the CEE involves ownership transitions from public to private hospitals, whereas in China, the hospital privatization process mainly involves increasing the number of new entrants. One common consequence of hospital reform shared by both the CEE and China has been a rapid increase in the number of private hospitals following reform. In the Czech Republic, some hospitals and most outpatient services were privatized when control of the state health institutions was transferred to the municipalities (Nemec and Kolisnichenko 2006). By 2007, nearly 1500 private healthcare institutions had been registered in the Ukraine; however, in 1991, there were only a few private entities noted by governmental agencies (Plugaru 2009). Some radical examples can also be observed. The Bulgarian government approved a policy aimed at the full privatization of hospitals in 2008, while all hospitals, except for four state-owned hospitals, have been taken over mainly by healthcare and real estate companies in Georgia. In Poland, the change to private healthcare provision is also in progress (Armstrong 2008). All these countries possess a hospitalcentred delivery system in their health sectors. Public or government-owned hospitals (those directly or indirectly controlled by central and local governments) form the foundation of these health institutions.

7 36 HEALTH POLICY AND PLANNING Table 4 Characteristics of the healthcare workforce between public and private hospitals in Fujian province in 2008 Characteristics Public hospitals Private hospitals Total P-value n ¼ 231 n ¼ 145 n ¼ 376 Mean staff number in a hospital, n Mean doctor number in a hospital, n Mean age on index date, years (SD) (9.80) (13.67) (10.23) <0.000 n ¼ n ¼ 7557 n ¼ Age group, n (%) <30 years (28.26) 345 (49.56) (30.2) < years (33.54) 1631 (21.58) (32.45) 40 years (26.7) 814 (10.77) (25.25) 50 years 7607 (10.07) 729 (9.65) 8336 (10.03) 60 years 1083 (1.43) 638 (8.44) 1721 (2.07) Total (100) 7557 (100) (100) Gender, n (%) Male (34.91) 2823 (36.07) (35.02) <0.05 Female (65.09) 5004 (63.93) (64.98) Total (100) 7827 (100) (100) Education, n (%) Doctor 449 (0.59) 7 (0.09) 456 (0.55) <0.000 Master 2354 (3.12) 60 (0.77) 2414 (2.9) Undergraduate college (22.5) 1339 (17.12) (22) Junior college (or higher vocational education) (20.22) 2101 (26.86) (20.84) Technical secondary school (42.14) 3540 (45.25) (42.43) High school or below 8632 (11.42) 776 (9.92) 9408 (11.28) Total (100) 7823 (100) (100) Profession, n (%) Western/modern medicine (24.04) 2028 (26) (24.23) <0.000 Chinese traditional medicine 4100 (5.62) 302 (3.87) 4402 (5.45) Dentistry 922 (1.26) 49 (0.63) 971 (1.2) Nursing (41.43) 2718 (34.84) (40.79) Public health 239 (0.33) 63 (0.81) 302 (0.37) Pharmacy 4782 (6.55) 477 (6.11) 5259 (6.51) Laboratory technology 2459 (3.37) 296 (3.79) 2755 (3.41) Other health technology 3784 (5.19) 378 (4.85) 4162 (5.15) Non-health profession 8914 (12.21) 1490 (19.1) (12.88) Total (100) 7801 (100) (100) Qualification title, n (%) Chief 2079 (2.79) 167 (2.16) 2246 (2.73) <0.000 Assistant chief 6550 (8.78) 499 (6.44) 7049 (8.56) Professional in charge (23.35) 1285 (16.59) (22.71) Junior professional (31.93) 1323 (17.08) (30.53) Junior assistant professional (18.23) 2869 (37.05) (20) Professionals with no title (14.92) 1601 (20.67) (15.46) Total (100) 7744 (100) (100) The expansion of private healthcare facilities, especially private hospitals, may have various effects on equity, health expenditures and access to healthcare services in transitional and developing countries. Admittedly, private healthcare providers, motivated by profit generation, might intensify information asymmetries in the healthcare market and prescribe more drugs and treatments, thus increasing healthcare expenditures (Blumenthal 2001; Xirasagar and Lin 2004); or they may leave remote or poor areas without medical coverage, leading to access inequalities. Despite the possible

8 GROWTH OF PRIVATE HOSPITALS IN CHINA 37 Physician Public Hospitals Private Hospitals Percent Nurse Public Hospitals Private Hospitals Percent Figure 2 Age structure of physicians and nurses by hospital ownership in Fujian province in Table 5 Age structure of physicians and nurses in Fujian province in 2008 Public hospitals Private hospitals Total P-value n % n % n % Physician <30 years 4369 (19.37) 469 (22.17) 4838 (19.61) < years 8644 (38.32) 602 (28.46) 9246 (37.47) 40 years 6286 (27.86) 305 (14.42) 6591 (26.71) 50 years 2591 (11.48) 273 (12.91) 2864 (11.61) 60 years 670 (2.97) 466 (22.03) 1136 (4.6) Total (100) 2115 (100) (100) Nurse <30 years (40.75) 2051 (75.49) (43.62) < years 9725 (32.17) 354 (13.03) (30.59) 40 years 6726 (22.25) 119 (4.38) 6845 (20.78) 50 years 1449 (4.79) 158 (5.82) 1607 (4.88) 60 years 11 (0.04) 35 (1.29) 46 (0.14) Total (100) 2717 (100) (100) disadvantages of private providers, however, empirical research demonstrates their positive effects, including a reduction in the average medical expenditures of public general hospitals (Liu et al. 2009), greater efficiency compared with public hospitals (Nyman and Bricker 1989; Chang et al. 2004), and spillover benefits of medical productivity by for-profit generating hospitals (Kessler and McClellan 2002). In addition, the increasing number of private hospitals in China can improve access to primary and secondary healthcare in many urban and rural areas, as no clear limits between primary and higher levels of care can be distinguished, many hospitals hold large resources for their outpatient departments, and a majority of people are used to visiting hospitals. Although the analysis of longitudinal data detected a shift and an increased growth rate for the number of private hospitals over the long term following the implementation of market-opening

9 38 HEALTH POLICY AND PLANNING Table 6 Specialty distribution of inflow healthcare workers by gender in Fujian province from 2002 to 2008 Male Female Total n % n % n % P-value Western/modern medicine 81 (67.50) 39 (19.12) 120 (37.04) <0.000 Chinese traditional medicine 14 (11.67) 4 (1.96) 18 (5.56) Nursing 1 (0.83) 112 (54.90) 113 (34.88) Public health 0 (0.00) 4 (1.96) 4 (1.23) Pharmacy 4 (3.33) 21 (10.29) 25 (7.72) Laboratory technology 6 (5.00) 9 (4.41) 15 (4.63) Other health technology 8 (6.67) 4 (1.96) 12 (3.70) Non-health profession 6 (5.00) 11 (5.39) 17 (5.25) Total 120 (100.00) 204 (100.00) 324 (100.00) Figure 3 Age structure of inflow healthcare workers by gender in Fujian province from 2002 to policy, the results did not fully imply causality between policy implementation and the expansion of private hospitals. Many other factors could explain the growth of the entire hospital sector. For example, around the year 2000, many public hospitals, including government-owned and state enterprise-owned hospitals, were facing excessive employment, poor financial performance, attempts by doctors and other medical staff to achieve a higher level of remuneration and better working conditions, and attempts by many local governments to reduce their budgetary burdens by encouraging the privatization of some public hospitals. Although performing health sector reform in this manner (ownership conversion) remains a controversial and political issue in China, it could be concluded that the expansion of private hospitals industry resulted from the creation of a set of unbiased market-based policies. Hospital physician relationships in China It is not difficult to identify the sources of elderly doctors and young nurses: the former may come from those doctors retired from other hospitals, especially public hospitals, whereas the latter must represent new workers entering the healthcare market. The question that arises, however, is why this unique age structure arises. The answer may partly be connected with specific Chinese hospital-physician relationships (HPR), a variety of collaborative arrangements made between hospitals and their medical staff (Burns and Muller 2008). In many Western countries, hospitals are traditionally seen as a rent-free workshop for physicians. Hospitals provide the equipment and the staff that enable the physicians to practise medicine; the physicians are not salary-based employees of the hospitals, but they can agree to a patient s admission into the

10 GROWTH OF PRIVATE HOSPITALS IN CHINA 39 hospital and also join hospital committees. However, the common HPR in China depends upon employment and its related combination of salary and bonuses. Following the framework developed by Burns, three types of integration in the continuum of HPR in China can be defined between the public and private hospitals: non-economic, economic and clinical integrations (Burns and Muller 2008). First, non-economic HPR refers to the hard and soft incentives that hospitals use to attract physicians, including well-trained nurses, physicians liaisons and their development, medical equipment and the hospitals efforts to create a transparent and effective organizational culture. Public hospitals mainly rely on an administrative personnel system that holds fixed authorized positions for hospital staff. The staffing decisions are often controlled by overlapping local personnel bureaus and health bureaus. This close relationship between governments and hospital physicians provides the physicians not only with a convenient pathway for technical qualification development but also with a restriction imposed by the administrative authority that is especially obvious for some highly capable senior doctors. These doctors cannot easily move or dual practise even if some private hospitals attempt to attract outstanding doctors by offering higher salaries. Second, economic HPR refers to a hospital s remuneration to the physician for their provision and the performance of clinical services and organizational activities. The remuneration of a physician includes a basic salary, bonus payment and social welfare. A well-known fact in Chinese public hospitals is that a combination of low salary and high bonus perversely encourages physicians to focus on revenue generation or informal payments from drugs and other sources instead of on effective and efficient practice (Pei et al. 2004). Although private hospitals may provide higher salaries and total incomes with more transparent payment systems, physicians in public hospitals most likely remain reluctant to change their employer. On the one hand, they need to adjust to the new economic system; on the other hand, they may have to exert more effort to improve their practice performance. Empirical studies have already provided evidence that the remuneration differences between public and private hospitals show that private hospitals have to pay higher salaries to recruit experts and employees (Zhang 2006). The reasons for this include the higher costs of advertising and purchasing medical equipment. The other side of this coin is that various units in the private hospitals charge much more than public hospitals and must provide a number of benefits to their customers. Finally, clinical HPR encompasses the hospital s systems and structure used to coordinate healthcare services across people, activities, functions and sites. Many private hospitals are new entries into the healthcare industry, and they cannot adequately prepare for the establishment of an institutional system for clinical services. This is also the reason that private hospitals require a large number of human resources from public hospitals, which have established relatively appropriate guidelines, protocols and referral systems. However, physicians from public hospitals are most likely unsatisfied with the limited clinical support structure available in private hospitals. The special HPR between public hospitals and private hospitals may explain the unique age structure in China. Despite the fact that new private hospitals have enrolled many young medical professional students and improved the lifetime welfare of retired doctors, it remains to be seen whether HPR can enhance the quality of care, decrease health expenditures, or improve access to care. Currently, some pilot projects focused on the restructuring of public hospitals have been launched in certain areas, but a rigorous examination of HPR is required to build a more qualified and motivated medical workforce. The possible impacts of elderly doctors Generally, the mandatory retirement age in China is 60 years for males and 55 years for females. The fact that 22.03% of physicians who practise in private hospitals are over the age of 60 is not surprising according to the literature (Xu and Huang 2006; Xue 2008), which demonstrates that private hospitals have hired many elderly doctors. Our study, however, is the first to measure this ratio compared with public hospitals. There is no clear evidence demonstrating a correlation between the quality of health services provided by private health facilities and the significantly higher level of aged doctors working in the private sector. Younger doctors should be more informed about new medical advances but may have less mature clinical judgment because of their relative lack of experience. In contrast, older doctors who have practised medicine over a long period may have refined their clinical judgment but may be uniformed regarding the latest advances in medicine. Thus, can experience counteract the effects of ageing? Eva (2002) suggested that ageing induces cognitive changes in the way that diagnosticians approach clinical cases, and most physicians cannot overcome age-related decreases in performance. Choudhry et al. (2005) documented 62 published studies on physicians length of practice and quality of healthcare, finding that decreasing performance was associated with increasing years in practice for all outcomes assessed. This conclusion was also confirmed by studies of operative mortality and surgeons ages (Waljee et al. 2006). Although most of these studies were conducted in developed countries, particularly in the USA, and very little evidence was collected from developing countries, there might not be significant differences between developed and developing countries in this regard. Eggleston (2010) found that it was not possible to harm overall quality by shifts in ownership type alone. However, this finding regarding the age of physicians may also suggest that it is necessary to pay attention to the high number of physicians more than 60 years of age employed in the private healthcare sector. Conclusions Based on a segmented regression analysis of a longitudinal data series of the number of hospitals from 1990 to 2009 in China, the before and after effects of market opening policy on private hospitals were examined. In addition to the immediate increase in the number of hospitals following policy implementation, the year-to-year changes in the number of specialty hospitals before and after 2001 were significantly different as well, possibly reflecting a rapid increase in the number of private hospitals. The growth of private hospitals could also be observed in some

11 40 HEALTH POLICY AND PLANNING countries in CEE and CA that possessed health systems similar to that of China. We also found a large percentage of elderly physicians practising medicine in private hospitals, which was significantly different from public hospitals. At least 4.1% of the medical staff in private hospitals was found to come from retirees or employees previously employed in local public hospitals, and many young healthcare workers were new labourers in the healthcare industry. We argue that the specific HPRs found in China may contribute greatly to the imbalanced age structure found between those working in public and private hospitals. Moreover, the potential effects of elderly physicians on quality were not clear in China, although some evidence regarding its negative effects have been documented. Currently, billions of RMB yuan from central governmental budgets are invested in the restructuring of public hospitals and health sector reform in China. Given the nature of the constantly changing healthcare sector, continuous investments are necessary to achieve high-quality and cost-effective care, which is especially important for new private hospitals. Further research regarding the development effects of private hospitals on the entire health system is required, as is improved workforce planning. Acknowledgements We give special thanks to two anonymous Referees, Rosalind McCollum, James Leeper, and Liu Xiaoyun for their review and comments. We are grateful for Pan Baojun and Chen Miaomiao, who actively participated in the design and advised in the implementation of the study. Thanks to Zhang Meng and Chen Qiaochai, who contributed to the data inputs and corrections. Funding The funding for this study was provided by Health Department of Fujian Province. Conflicts of interest statement. None declared. Endnotes 1 Both policies were formulated under the direction of the Regulations on Administration of Healthcare Facilities, which was also issued in 1994 and aimed to improve the management of healthcare facilities. The policy also rescinded other similar administration policies that had been approved in Many doctors have been trained in the last three decades in China. A large percentage of these doctors received education below the college level; most of these doctors never entered into any formal residency programme. 3 Similar to reform of the healthcare system, China initiated reforms of state-owned enterprise in the 1990s as well, under the guidance of then Prime Minister Zhu Rongji (Steinfeld 2000). The main content of this reform was the privatization of small and weak state-owned enterprises and the marketization of large and advanced state-owned enterprises (Yusuf et al. 2006). During this process, most state-owned hospitals were privatized, and thus, we treat it as one part of the growth of private hospitals. References Armstrong K Eastern Promise. Online at: (Accessed 1 March 2011). Bennett S Promoting the private sector: a review of developing country trends. Health Policy and Planning 7: Blumenthal D Controlling health care expenditures. The New England Journal of Medicine 344: Burns LR, Muller RW Hospital-physician collaboration: landscape of economic integration and impact on clinical integration. The Milbank Quarterly 86: Chang H, Cheng M, Das S Hospital ownership and operating efficiency: evidence from Taiwan. European Journal of Operational Research 159: Choudhry NK, Fletcher RH, Soumerai SB Systematic review: the relationship between clinical experience and quality of health care. Annals of Internal Medicine 142: Eggleston K, Ling L, Qingyue M, Lindelow M, Wagstaff A Health service delivery in China: a literature review. Health Economics 17: Eggleston K, Lu M, Li C et al Comparing public and private hospitals in China: evidence from Guangdong. BMC Health Services Research 10: 76. Eva KW The aging physician: changes in cognitive processing and their impact on medical practice. Academic Medicine 77: S1 6. Gu E, Zhang J Health care regime change in urban China: unmanaged marketization and reluctant privatization. Pacific Affairs 79: Huang C, Liang H, Chu C, Rutherford S, Geng Q The Emerging Role of Private Health Care Provision in China: A Critical Analysis of the Current Health System. Health Policy Program working paper #10. Stanford: Walter H. Shorenstein Asia-Pacific Research Center, Stanford University. Kessler DP, McClellan MB The effects of hospital ownership on medical productivity. RAND Journal of Economics 33: Lagarde M How to do (or not to do)...assessing the impact of a policy change with routine longitudinal data. Health Policy and Planning 27: Lim MK, Yang H, Zhang T et al The role and scope of private medical practice in China. Online at: content/private_medical_in_china.pdf (Accessed 1 March 2011). Liu GG, Li L, Hou X, Xu J, Hyslop D The role of for-profit hospitals in medical expenditures: evidence from aggregate data in China. China Economic Review 20: Liu Y, Berman P, Yip W et al Health care in China: the role of non-government providers. Health Policy 77: Ma J, Lu M, Quan H From a national, centrally planned health system to a system based on the market: lessons from China. Health Affairs 27: Ministry of Health Chinese Health Statistical Digest. Beijing: Ministry of Health. Online at: htmlfiles/zwgkzt/ptjty/200805/35310.htm (Accessed 9 March 2011). Ministry of Health Chinese Health Statistical Yearbook. Beijing: Ministry of Health. Online at: business/htmlfiles/zwgkzt/ptjnj/year2010/helpheader.html (Accessed 9 March 2011). Nemec J, Kolisnichenko N Market-based health care reforms in Central and Eastern Europe: lessons after ten years of change. International Review of Administrative Sciences 72: Nyman JA, Bricker DL Profit incentives and technical efficiency in the production of nursing home care. The Review of Economics and Statistics 71: