Evaluation of patient satisfaction in hospitals in China 2003–2013: A national longitudinal analysis of population-based survey data


 Understanding the perception of patients towards the quality of hospitals is critical to policy development and improvement. However, there remains a dearth of literature examining the impact of the new health reform launched by China's government in 2009 on the changes of the patient satisfaction. Thus, the aim of this paper was to evaluate long-term trends in patient satisfaction to healthcare providers in China from 2003 to 2013. The data from three rounds (2003, 2008, and 2013) of the National Health Service Survey. The unsatisfactory rate, hospital satisfaction assessment and the most dissatisfied items for outpatients and inpatients in urban and rural areas in China were considered Percentages changes from 2003 to 2008, 2008 to 2013, and 2003 to 2013 were calculated to descriptively show the trends of patient satisfaction between these 10 years. Chi-square tests regarding the changes of each item between any two years were carried out to evaluate temporal variance. The results reveal a significant overall improvement in the patient satisfaction of both inpatients and outpatients in urban and rural areas in China from 2003 to 2013, especially for the period between 2008 and 2013. However, the percentage of outpatients who dissatisfied with the waiting time for treatment largely increased between 2008 and 2013. In addition, both inpatients and outpatients tended to pay more attention to the service quality and medical expense. With respect to the patient-physician trust relationship, an overall marked improvement could be identified. The findings of this study are unique in presenting the patient satisfaction at the national level in China, which highlight a need to carry out more national actions focusing on the optimization of visiting mechanism, the deduction of medical costs and the improvement of doctors' service and medical quality.

Introduction and unnecessary medical services and charging excessive costs for a given medical service. Moreover, for experience goods, return policies that can reduce the risk of consumers' irrational buying behavior to a large extent are widely provided. That is, the reversibility characteristic is contained in this selling-buying process. However, credence goods, such as medical services, have no return policies, and the trading process is irreversible. For example, surgery for patients cannot be "returned," although it can be officially proved to be inappropriate or completely wrong after the service. These effects enable patients to be more cautious and thoughtful when selecting healthcare providers, especially the high risk-aversion ones and those that frequently cause workplace violence events in hospitals [12]. Improving the patient-physician trust seems to be the only effective way to reduce this kind of risk.
China has the largest population in the world. Healthcare services have been identified as a key issue in its population's well-being [13]. perspective is necessary. However, little research has been conducted to evaluate the quality of healthcare institutions with respect to the patients, and most of these studies focused only on some particular regions [14,15] or particular diseases [16]. Consequently, the perception and expectation of patients of Chinese hospitals are still poorly understood. This limitation is largely due to the scarce data, which are caused by the deficiency in medical data sharing, consolidation, and standardization processes in China and the special sensitivity feature of the medical data.
In 2009, the Chinese government implemented a new health reform to solve the difficulty and expensive medical treatment problems and to improve patient satisfaction [17].
Understanding the performance of this reform from the patients' perspective is important.
In this field, Yan et al. investigated the effects of the new health reform on patient satisfaction by comparing urban and rural areas [18]. Wang et al. identified the factors affecting patients in terms of China's public hospitals after the new health reform [19].
However, only 40 county-level hospitals in two provinces of China and 5 public hospitals in Jiangsu Province of China were considered by these two papers, respectively, and thus these studies could not present a national assessment analysis. In addition, with respect to patient-physician trust, a decreasing trend was identified by previous studies in China's healthcare market before the implementation of the new health reform [20,21], and this trend was mainly caused by the persistent imbalance between the expectation of the public and the poor effects of China's health policy. Consequently, the satisfaction degree of patients seemed to be continuously low in that period of time [20]. However, the number of studies focusing on this trust relationship after the new health reform is relatively few, and most of them paid more attention to the influencing factors [21] and improvement strategies [22] than to the effect of the new health reform on the changes in the patient trust degree. Moreover, there is also a shortage of the use of national survey data in the existing literature. As a result, the current state of this trust relationship is still poorly understood.
The overall aim of this study was to examine patient satisfaction in hospitals in the urban and rural areas in China over the last 10 years and to evaluate the effect of the new health reform on this issue from national and multidimensional perspectives by considering the quality of medical staffs, medical services, treatment environments, and patient-physician trust. The expectation of patients, which was considered as an important factor in the improvement of patient satisfaction [7], in the future development of hospitals in China was also considered.

Study design
This study comprised a longitudinal analysis of patient satisfaction in hospitals in China using the National Health Service Survey (NHSS). NHSS is a population-based, nationally representative survey that collects information on health service demand and utilization every five years in China [23]. This study was organized in compliance with the REporting of studies Conducted using Observational Routinely collected Data Statement.

Data source
In total, NHSS underwent five rounds conducted in 1993, 1998, 2003, 2008, and 2013. In terms of patient satisfaction in hospitals, only the third (2003), fourth (2008), and fifth (2013) round of surveys considered the relevant issues. Thus, these three rounds of surveys were applied in this study.
Three rounds (2003, 2008, and 2013) of the NHSS were completed using a multi-stage sampling and face-to-face interviewing method [24]. In the patient satisfaction assessment included in the NHSS questionnaires, both the outpatients and the inpatients, two important components of patients, were considered. In comparison, outpatients and inpatients have considerable differences [25]. Specifically, the former is less dependent on medical care and has less communication with physicians than the latter. Conversely, inpatients generally spend more time in hospitals and depend more on medical care than outpatients. Some existing studies evaluated and compared the patient satisfaction of these two groups [25,26]. Thus, the current study considered the perception of both outpatients and inpatients of hospital satisfaction to show a comprehensive picture. Based on the NHSS data, the two-week outpatients were set as the outpatient group. The structured interview questionnaires were not changed over the three rounds of the NHSS, except for some items that were updated and deleted because of related health policies.
According to relevant health data management regulations, the NHSS can be accessed through the Center for Statistics and Information, National Health and Family Planning Commission of the People's Republic of China.

Measures
Three measures of both outpatients and inpatients were considered in the three rounds of NHSS: unsatisfactory rate, hospital satisfaction assessment, and the most dissatisfying items. The details are shown in Table 1. These three measures reflected the quality of both medical staffs and hospitals and provided a relatively comprehensive assessment to a large extent.
Note that because of the design differences in the NHSS in the different rounds of surveys, the number of options for the hospital satisfaction assessment for the respondents By following Mathews et al. [27], the comparison of each item between any two years was conducted using chi-square tests to evaluate the temporal changes. All statistics were descriptive and were performed by using Stata 12.1. Statistical significance was defined as P < 0.05.

Satisfaction assessment
The satisfaction degree of the outpatients in both urban and rural areas largely improved from a global perspective, especially for the items related to doctors, such as "listing the seriousness of patients' illness." The changes in all items except "interpretation clarity of the treatment" (P value = 0.831) are statistically significant. Specifically, the percentages of the "very good/good" option for these two items within the urban and rural groups increased by 24.55%, 35.53%, 44.40%, and 43.76%, respectively. In terms of "medical treatment environment," a considerable increase was found during the 10 years, especially in 2008-2013, and this increase was more significant in the rural areas. The percentage of rural outpatients who chose "very good/good" increased by 70.73% in 2008-2013. Moreover, the trust degree of outpatients in doctors significantly increased in 2008-2013, especially for those in rural areas. Specifically, the percentage of the option "trust very much" chosen by the outpatients increased by 145.21% and that by the rural outpatients increased by 208.71%. The percentage of the urban and rural outpatients who selected the option "distrust very much" decreased by 80.67% and 94.38%, respectively (Table 3).

Most dissatisfying item
The results shown in Table 4  early and then a decreasing later tendency was found for the item "cumbersome procedures." A significant decrease was observed for the items "long waiting time for treatment," "unnecessary services," and "fewer drug varieties" (Table 4).
For the items "low medical service level" and "poor service attitude," a marked increase was found in both the urban and rural areas, especially in 2008-2013. In particular, the percentage of outpatients who chose "low medical service level" and "poor service attitude" as the most dissatisfying item increased by 80.75% and 249.83% in 2008-2013, respectively.
In terms of "unreasonable charges," the population in the urban and rural areas showed different results. The percentage of outpatients in the urban areas who were most dissatisfied with this item decreased by 20.11% in 2008-2013, whereas that of outpatients in the rural areas increased by 106.03% in that period of time.  (Table 5).

Satisfaction of inpatients in hospitals
Moreover, considerable enhancements were found for the items "convenience of querying medical costs" and "convenience of complaint." In terms of "trust degree in the doctor," a significant improvement was found in 2008-2013, especially for rural outpatients.
Particularly, the percentage of the population in the urban and rural areas that chose the option "trust very much" increased by 23.41% and 132.58%, respectively.

Most dissatisfying items
The calculated chi-square test results indicated that the concern of the inpatients regarding the most dissatisfying items was significant during the 10 years (the P values in 2003-2008, 2008-2013, and 2003-2013 for the total, urban, and rural groups are all smaller than 0.01). In the case of "low medical service level" and "poor service attitude," a significant increase was found in these 10 years, and this increasing characteristic was more pronounced in 2008-2013 and in rural areas. In addition, a decreasing early and then increasing later tendency was found in the item "high medical costs," and the items "poor medical equipment" and "cumbersome procedures" showed an inverse changing feature. These findings are similar to those for the outpatients. The percentage of inpatients who considered "unreasonable charges," "long waiting time for treatment," "unnecessary services," or "fewer drug varieties" as the most dissatisfying item decreased by 9.01%, 77.51%, 2.72%, and 47.08%, respectively, in 20013-2013 (Table 6).

Summary of the primary findings
By using the national population-based survey data, this study evaluated the long-term trends in patient satisfaction in hospitals in both urban and rural areas in China. The following findings were obtained: (1) Unlike the widely accepted conclusion that patient satisfaction in China is low, as proposed by previous studies using cross-sectional data sets of some particular areas [13,28], this study provided a more comprehensive comparison analysis using national time- (2) Patient satisfaction with respect to the issues on hospital management, such as treatment environment, visiting process and drug supply, and doctor service quality, significantly increased during these 10 years. However, the percentage of outpatients who were dissatisfied with the waiting time for treatment largely increased in 2008-2013, consistent with some existing studies [29,30].
(3) After the implementation of the new health reform, both outpatients and inpatients began to pay more attention to service quality and medical expense. Specifically, although the overall satisfaction of patients largely improved, the proportion of patients who were most dissatisfied with "low medical service level," "poor service attitude," and "high medical costs" of the total unsatisfied population significantly increased from in

Interpretation of findings
With the development of the economy, the quality of hardware devices, such as medical equipment, can be directly and easily improved in a short period of time. By contrast, how to enhance the medical ability and professional quality of physicians, especially of those in primary hospitals [32,33], and how to optimize the allocation of medical resources from a national perspective remain great challenges for the government. These problems can be viewed as the main cause of the phenomena of "difficulty and high cost of getting medical service" in China to date, and they are continuously cited as the primary solution targets by healthcare policies [34]. Therefore, sustainable and multi-sided modifications need to be considered by the Chinese government from the current status and long-term planning perspective.
According to Yan et al. [35], the healthcare system in China has faced a financial crisis in recent years, and this crisis is caused by the considerable gap between the growth rate of healthcare expenditure and that of the incomes of healthcare funds and budgets. To control the excessive growth of healthcare expenditure, the Chinese government has carried out a series of cost control actions or policies since 2012. However, based on our findings, the performance of these actions is relatively low. This conclusion is consistent with that of Ref. [35], who found that most of the cost control actions implemented by hospitals are lowering the quality of healthcare, leading to patient satisfaction not being considered as high priority. Moreover, Pan et al. found that the expensive medical charges were mainly responsible for patient dissatisfaction [36].
Previous studies also found a significant relationship between the hospital referral mechanism and patient satisfaction [33]. A rational referral mechanism is the basis for scientifically controlling and optimizing the allocation of medical resources and the mobility of patients. In 2009, China launched a two-way transfer mechanism in the new health reform. This referral mode is not mandatory but a voluntary selection behavior for the patients themselves. The only constraint measure carried out by the Chinese government is to implement the health insurance policy that provides the highest reimbursement rate for patients who comply with this mode. However, the fact is that the performance of this transfer mechanism is relatively weak. Zhang et al. stated that the "transferring service at the primary hospitals" and the "admission service at the higherlevel hospitals" are the two main factors causing patient dissatisfaction [38]. These two aspects can sum up the ineffectiveness in the policy implementation process, such as cumbersome and unreasonable procedures.
With respect to the patient-physician relationship, although it improved in 2008-2013, only 21.3% of the outpatients and 19.07% of the inpatients reported that they complete trusted their doctors. Therefore, low public trust between patients and doctors remains a potential problem in China's healthcare system. To solve this problem, enhancing healthcare experiences, clearly defining patients' role in the treatment proposed by healthcare administrators [39], and improving the number and quality of family doctors in community health service centers [40] seem to be effective and practical measures.
To address the gap in medical quality between urban and rural areas, many previous studies found that compared with the healthcare institutions in the rural areas of China, those in the urban areas are equipped with more and better health resources, such as financial, human, and material resources [41]. As a result, the level of patient satisfaction and experience in China's public hospitals seemed to be higher in the urban than in the rural areas [42]. However, we found a relatively contrasting phenomenon. The satisfaction degree of both outpatients and inpatients in the rural areas was confirmed to be higher in urban areas in these 10 years. The main reason for this difference is the different datasets used. However, because of the use of national population-based survey data, our conclusion seems to be more convincing. Since the promulgation of the New Rural Cooperative Medical Scheme (NCMS) in 2009 to improve the health of rural residents, the satisfaction of patients in the rural areas in China has largely increased, especially that of the poor ones [43]. Specifically, the NCMS has substantially advanced healthcare access and application and effectively reduced the medical impoverishment of low-income rural regions. However, this program has failed to decrease the financial burden of outpatient medical costs of poor communities [43], and this issue should be considered by the government in further modifications and reformations.

Policy implications
These findings have two policy implications. First, our findings can be interpreted as the basis for improvement to further advance the healthcare reform in China. More effective strategies need to be conducted to reduce the medical costs and optimize the visiting mechanism for patients, especially for outpatients. Moreover, how to provide a more balanced medical resource allocation should be considered by policymakers to decrease the occurrence of the serious crowding phenomenon in tertiary hospitals. Note that unlike the rapid improvement in hardware devices, the boost in the allocation of medical resources seems to be much difficult in China, the country with the largest population and the third-largest land area in the world. Therefore, long-term, sustainable, and multi-sided modifications need to be considered by the Chinese government.
Second, the results raise the concern about the rebuilding of the patient-physician trust relationship. Under the perception that medical services can be viewed as credence goods, enhancing the trust between patients and physicians is considered to be the only effective way to reduce the risk of insufficient and unnecessary medical services, excessive medical charges, and medical workplace violence. To achieve this goal, the establishment of a rational referral mechanism and a comprehensive family doctor system as well as the improvement of the overall performance of primary healthcare providers should be given attention in further political focuses.

Limitations
This study is primarily limited by the unavailability of patient satisfaction-related data

Declaration of interest
The authors declare no conflict of interest.

Ethical considerations
This study was performed in accordance with the ethical guidelines from the Ethics