You are here

  • Home
  • Archive
  • Volume 13, Issue 4
  • Examining clinical capability of township healthcare centres for rural health service planning in Sichuan, China: an administrative data analysis

Article Text

Article menu
Public health
Original research
Examining clinical capability of township healthcare centres for rural health service planning in Sichuan, China: an administrative data analysis
  1. Kun Zou1,2,
  2. Zhanqi Duan3,
  3. Ziwu Zhang3,
  4. Jinliang Hu4,
  5. Juying Zhang5,
  6. Jay Pan6,
  7. Chaojie Liu7,
  8. Min Yang8,9
  1. 1West China Research Centre of Rural Health Development, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
  2. 2Department of Pharmacy, Evidence-Based Pharmacy Center, NMPA Key Laboratory for Technical Research on Drug Products In Vitro and In Vivo Correlation, Key Laboratory of Birth Defects and Related Diseases of Women and Children, Sichuan University West China Second University Hospital, Chengdu, China
  3. 3Sichuan Provincial Big Data Center, Chengdu, Sichuan, China
  4. 4Institute of Health Policy and Hospital Management Research, Sichuan Academy of Medical Sciences and Sichuan People's Hospital, Chengdu, Sichuan, China
  5. 5Department of Epidemiology and Biostatistics, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
  6. 6HEOA Group, West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, China
  7. 7School of Psychology and Public Health, La Trobe University, Melbourne, Victoria, Australia
  8. 8West China School of Public Health and West China Fourth Hospital, Sichuan University, Chengdu, Sichuan, China
  9. 9Faculty of Health, Art and Design, Swinbune Technology University, Melbourne, Victoria, Australia
  1. Correspondence to Dr Min Yang; yangmin2013{at}scu.edu.cn

Abstract

Objective This study aimed to examine the clinical capability of township healthcare centres (THCs), the main primary care providers in rural China, as a basis for rural health service planning.

Design Observational study of quantitative analysis using administrative data.

Setting Three counties with low, middle and high social economic development level, respectively, in Sichuan province western China.

Participants 9 THCs and 6 county hospitals (CHs) were purposively selected in the three counties. Summary of electronic medical records of 31 633 admissions from 1 January 2015 to 30 December 2015 of these selected health institutions was obtained from the Health Information Centre of Sichuan province.

Main outcome measures Six indicators in scope of inpatient services related to diseases and surgeries in the THCs as proxy of clinical capability, were compared against national standard of capability building of THCs, among counties, and between THCs and CHs of each county.

Results The clinical capability of THCs was suboptimal against the national standard, though that of the middle-developed county was better than that in the rich and the poor counties. THCs mainly provided services of infectious or inflammatory diseases, of respiratory and digestive systems, but lacked clinical services related to injuries, poisoning, pregnancy, childbirth and surgeries. A large proportion of the top 20 diseases of inpatients were potentially avoidable hospitalisations (PAHs) and were overlapped between THCs and CHs.

Conclusions The clinical capability of THCs was generally suboptimal against national standard. It may be affected by the economics, population size, facilities, workforce and the share of services of THCs in local health systems. Identification of absent services and PAHs may help to identify development priorities of local THCs. Clarification of the roles of THCs and CHs in the tiered rural health system in China is warranted to develop a better integrated health system.

  • PUBLIC HEALTH
  • Organisation of health services
  • Health policy
  • HEALTH SERVICES ADMINISTRATION & MANAGEMENT

Data availability statement

No data are available.

http://creativecommons.org/licenses/by-nc/4.0/

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

https://doi.org/10.1136/bmjopen-2022-067028

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Strengths and limitations of this study

    • This study uses regional objective electronic medical record (EMR) data to portrait clinical capacity of typical township healthcare centres (THCs) with low, middle and high economic level in rural China.

    • The clinical capacity of THCs was compared against national standards, between counties with typical social-economic and geographic differences, and with local county hospitals (CHs) for understanding their strengths and limitations.

    • This study looks into essential but absent clinical services and potentially avoidable hospitalisations in THCs, and overlapped hospital services between THCs and CHs, which has been rarely investigated previously.

    • This study used 2015 data and more recent data is to be used for future research.

    • EMR was only available for inpatients, future research may include outpatient data to allow a more comprehensive portrait of clinical capability of THCs.

    Introduction

    Primary healthcare (PHC) is vital to provide essential, accessible, equitable and affordable care towards universal health coverage and health-related sustainable development goals.1 PHC institutions are also the foundation of the three-tiered healthcare system in China, with secondary hospitals (at county-level) and tertiary hospitals (at city-level or provincial-level) at the end.2 At the bottom of the three-tiered healthcare system, the township healthcare centre (THC) is the main PHC provider for the 650 million rural residents in China,3–5 with its main workforce of licensed general practitioners and registered nurses.6 They are responsible for the national essential public health services, primary outpatient services and inpatient services for common illness for rural residents. Although THCs of 35 762 in total in 2020 provided 1.1 billion episodes of healthcare including 34 million episodes of inpatient services, accounted for 15% and 14% of that provided by all health institutions in China,7 the share of inpatient treated in THCs in the national healthcare system of China was declining from 26% in 2010 to 17% in 2016.7

    The declining in inpatient services in the THCs could have several reasons. First, due to the fast urbanisation and transportation development in China, patients were more likely to bypass primary care facilities to obtain perceived high-quality clinical services from higher-level hospitals if they were highly accessible and financially affordable.8 Second, currently there is no mandatory or legitimated gate-keeping role of the government for THCs. For patients with more severe or complex conditions, THCs are only advised to refer them to visit county or tertiary hospitals for more specialised care.3 5 In most cases, patients are free to visit any county or tertiary hospitals without restriction. Third, the clinical capability of THCs is varied by regions, and is much weaker in remote and mountain areas due to lack of resources and especially qualified health workforce than their counterparts.9 The weakness of capability of THCs would cause more patients losing to higher-level hospitals and thus cause a vicious circle, putting the survival of THCs and health of rural residents at risk.2 Consequently, patients with common chronic illness such as hypertension and diabetes that are supposed to be cared by THCs or PHC providers went to seek care in county or tertiary hospitals, which lead to the underuse of THC resources and roaring hospital expenditure.8 10

    There was evidence that clinical capability was the primary reason limiting patient’s choice of primary healthcare,11 and improving the clinical capability including drug availability, medical professional skills and equipment in rural primary care systems can induce potential medical care-seeking, and redirect patient flow from higher level hospitals to primary level in rural China.12 Many extrinsic and intrinsic factors were involved in shaping the clinical capability of THCs. Local economic status, health finance, facilities and health workforce were positively associated with clinical capability and range of clinical services of THCs.13 The serving population was also an important extrinsic factor besides local economics.14 Health policies for reducing maternal and neonatal mortality and ensuring the quality of maternal care in China had promoted hospital delivery in secondary or higher-level hospitals,15 16 which led to terminated maternal services in most THCs.

    Understanding the clinical capability of THCs in regions with different social economic development levels would have important implications in rural health service planning and effectively establish gate-keeping role of THCs. To improve the healthcare services in THCs, the government has published the national guideline of THCs capability building and evaluation.17 However, few studies have explored the clinical capability of THCs, and even fewer had compared that between counties with different socioeconomic and health system contexts.2 18 19 The understanding of the role of THCs in rural health systems in China is inadequate. The county hospital (CH), usually one or two located in the county centre in each county, is the first referral destination of the THCs and the main origin of transferring discharged patients down to THCs for posthospital care.20 To understand the similarity and difference in the clinical capability between the CH and THCs could help health authorities to make guideline for patient centred specific care in the healthcare services, which may consequently lead to better quality and integrated healthcare for rural population.

    This study aimed to fill the gaps by examining the clinical capability of THCs and comparing it with local CHs. Considering that clinical capability by definition is a multiple-dimensional system, we previously reported quality of outpatient services of the sampled THCs,21 and of workload in preventive care in the study THCs.22 Hence, this study only focused on scope of inpatient services as a narrow proxy of clinical capability to reflect the coverage of the curative care of THCs. It is known that types of inpatients clinical services and their complexity may represent the clinical capability of a medical institution. The research questions were (1) what scope of inpatient services the THCs currently provide in terms of disease types and surgery procedures, (2) how much difference in those services provided by THCs and by the local county-level hospitals and (3) how much difference in those services among THCs with different economic status?

    Methods

    This is an observational study based on administrative data from the national inpatient electronic medical records (EMRs) system. The EMRs have been established by the Committee of Health and Family Planning of Sichuan province and operated among all levels of hospitals in Sichuan province since 2014 for administrative management. Each inpatient admission has a summary or homepage of the EMRs completed at discharge. Every hospital submits the nationally standardised summary of EMRs electronically at monthly basis. This study used data from January to December 2015. The reliability of the inpatient EMRs database has been verified, on which a number of studies have been conducted and published.23 24

    Sample county

    Sichuan province lies in the Hu (Heihe–Tengchong) line at southern west China. The Hu line has significant geographic and socioeconomic significance. In the southeast of the line, lived 94% of the population of China with 43% of the Chinese territory, while only 6% population lived in the west of line with 57% of the territory.25 Therefore, Sichuan is a typic province that could reflect China’s demographic, geographic and average socioeconomical variations (online supplemental table 1). According to the 2015 Sichuan Province Health and Family Planning Yearbook, Sichuan had 447 CHs and 5448 primary healthcare facilities, with 4511 (83%) rural THCs and 937 urban community health centres.20

    We purposively selected three counties in Sichuan province considering their socioeconomic and health system characteristics. Longquanyi (LQY) represented the richest region located in the plain of Sichuan basin, close to the provincial capital Chengdu (23 km) with convenient transports by bus or metro (about 40 min), with GDP per capita of about US$17 590 in 2015, two times higher than the national level at 2015 (online supplemental figure 1). By contrast, Ganluo (GL) represented the poorest remote region located in western mountains, 350 km southwest from Chengdu (6 hours by bus or train) and 200 km from its municipal city Xichang (5 hours by bus). Its GDP per capita was about US$1908, only one-fourth of the national average. Fushun (FS) county represented the middle-developed region located in a hilly area, 215 km southeast from Chengdu (3 hours by bus), and 37 km from its municipal city Zigong, with GDP per capita being half of the national average (table 1 and online supplemental figure 1). The three sample counties had a total population of 1.96 million and rural population of 1.35 million (68.9%) in 2015 (online supplemental table 2).

    View this table:
    Table 1

    Health resources and characteristics of inpatient admission of sampled township healthcare centres in the three counties 2015

    Participating institutions

    We purposively selected three THCs from each of the three sampled counties based on their distance of being close, average and remote to the county centre considering the accessibility of health services, and two CHs in each county were selected for comparison with that of THCs (online supplemental figure 1).

    Data extraction

    Under a research agreement with the data management authority, the Health Information Centre of Health Commission of Sichuan Province, the research team obtained the EMRs data electronically for the required period. All patients admitted and diagnosed in the selected THCs and CHs from January to December 2015 were included in the analysis. Overall, the study data had 31 633 admissions, including 6675 to THCs and 24 958 to CHs. The EMRs contained anonymous information of patients’ demographics, admission and discharge status, hospitalisation expense, disease diagnoses coded by the International Classification of Disease (ICD)-10th version, and surgeries coded by the ICD-9-CM.

    Indicators

    Given the fact that patients freely choose any hospital for service use in China, inpatients healthcare services represented the most complicated healthcare that would be offered by a health institution for patients to choose from. For the health system assessment, the use of services by patients has been considered an important indicator assessing capacity of primary healthcare institutions by the health authority in the country.17 A previous study by our team developed a tool to assess clinical care quality of county-level hospitals based on the summary information of EMRs. The tool consisted 25 indicators covering 6 dimensions. The first dimension was clinical capability with types of diseases diagnosed and treated as the key and only indicator.24 For clinical capability of the THCs in this study, we used six clinical service scopes as a proxy of clinical capability indicators based on the national standard of capability building of THCs,17 and on the availability of valid information in the data sources:

    1. Coverage of essential clinical services as listed in the national standard of capability building of THCs,17 a number count of diseases diagnosed by disease domain.

    2. Coverage of diseases by systems defined by ICD-10 chapters, a checklist with the number count of diseases diagnosed by disease systems.

    3. Coverage of surgery by systems defined by ICD-9-CM chapters, a number count of surgery procedures performed by disease systems.

    4. Number of types of diseases defined by ICD-10 codes.

    5. Top 20 diseases and their relative frequency among all services defined by ICD-10 codes. From all diseases diagnosed and treated, list the first 20 ranked by counts and relative frequency.

    6. Top 10 surgeries and their relative frequency among all surgeries defined by ICD-9-CM codes, the same definition to indicator 5 but for surgery services.

    The basic unit of all indicators is a curative service for disease or surgery at each hospitalisation of a patient who could have several units if admitted to and discharged from a hospital several times during the study period. The larger number of indicators in healthcare institutions reflects wider diseases they treated and more surgeries they operated by three levels: disease domain (indicator 1), disease systems (indicators 2 and 3) and specific diseases (indicators 4, 5 and 6).

    Statistical analysis

    Indicators of the THCs and CHs were derived separately. The formers were grouped by county and compared against those of the CHs within each county. The coverage of essential clinical services of the THCs (indicator 1) was presented against the number in the national guideline and compared among three counties using χ2 test. Indicators 2 and 3 of the THCs were compared among counties, and to that of the CHs too descriptively using frequency and percentages. The number of types of diseases and surgeries (indicator 4), ranks of diseases (indicator 5) and surveys (indicator 6) were also described and compared among counties using frequency and percentage. Larger numbers of indicators demonstrated a wider scope of inpatient services which suggested higher clinical capability in THCs. Where ever necessary, χ2 test was used to compare binary indicators between THCs of different counties and between THCs and CHs. One-way analysis of variance was used for continuous variables if they were normally distributed, otherwise Kruskal-Wallis test was used. All analyses were conducted using Stata V.15.0 SE, with two-sided significant level at 0.05.

    Patient and public involvement

    Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.

    Results

    Health resources of sample THCs

    The sampled THCs in LQY, FS and GL served 180 961, 76 326, 65 988 permanent residents, respectively. The number of physicians (including assistant physicians) in the sampled THCs was 79, 57 and 24, the number of registered nurses was 62, 29 and 16, and the number of beds was 148, 119 and 36, respectively, for the three counties in 2017 (table 1).

    Characteristics of inpatients

    Overall, 31 633 admissions were included in the analysis, including 6996 of THCs and 24 958 of CHs. The patient number of THCs for the LQY, FS and GL was 1941, 4042 and 1013, respectively. Generally, inpatients of THCs in the LQY were significantly older, with more men, longer length of stay and higher expenditure per admission (table 1).

    Coverage of essential clinical services in national guideline standard

    The coverage of 66 essential clinical services listed in the national guideline of THCs capability evaluation varied substantially among the three counties. While similarities between the rich-LQY (46, 70%) and the middle-developed FS (47, 71%) county, the coverage rate was much lower in the poor GL county (16, 24%). For clinical services of internal medicine, surgery, ENT, gynaecology/obstetrics and stomatology, the coverage rate ranged from 66.7% to 85.7% in the rich county, 57.1% to 84.6% in the middle county and 10% to 50% in the poor county, which were all significantly lower in the poor county than its counterparts. The most absented services were ENT (1/10), gynaecology/obstetrics (1/7) and surgical services (1/17) (table 2).

    View this table:
    Table 2

    Coverage of essential clinical services* against national guideline of capability development of township healthcare centres in the three counties

    Coverage of disease systems

    The capability of THCs in the middle-developed county was stronger than the other two counties in terms of coverage of disease systems. Among the 19 disease systems classified by the ICD-10, THCs in the middle-developed county covered 18 systems except for perinatal care, the rich county covered 13, absent of perinatal, congenital, mental, blood, eye system and pregnancy/childbirth services, while the poor county covered 14, absent of perinatal, congenital, blood, eye system and pregnancy/childbirth (table 3).

    View this table:
    Table 3

    Disease system covered by township healthcare centres in the three counties

    For THCs overall, the top five disease systems with the largest number of admissions were respiratory (50.0%), digestive (14.2%), circulation (7.6%), urinary (7.0%) and musculoskeletal (5.1%) systems, while that of CHs were respiratory (25.8%), digestive (16.9%), circulation (10.3%) systems, injury/poisoning (9.2%) and pregnancy/childbirth (8.4%) (online supplemental table 3).

    Number of disease and surgery types

    THCs in the middle-developed FS county had the largest number of types of diagnosis (430), about three times of that in the rich LQY (136) and the poor GL (101) county. The ratio of number of types of diagnosis of THCs against CHs in the middle-developed county FS (47.8%) was also higher than that of the poor (17.4%) and rich county (15.6%) (table 4).

    View this table:
    Table 4

    Number of types of diseases and surgeries in township healthcare centres and related county hospitals

    Top 20 diseases

    The top 20 diseases accounted for 69.5% of inpatient admissions in THCs and 42.5% in CHs overall. In THCs, most of the top 20 diseases were infectious or inflammatory diseases belonging to respiratory, digestive and urinary systems. Non-communicable diseases of cardiovascular and musculoskeletal systems were also involved but had a smaller proportion compared with infectious diseases, with even smaller proportion of pregnancy and childbirth services (online supplemental table 4).

    In comparison, the top 20 inpatient diseases in CHs included more severe or acute conditions which may require surgeries, such as acute appendicitis (1.6%), cerebral infarction (1.4%), internal haemorrhoids (1.4%), calculus of gallbladder with acute cholecystitis (1.3%), caesarean delivery (1.3%) and age-related cataract (1.0%). There were also more patients with diabetes complications (0.9%) in CHs, which was not among the top 20 diseases in THCs (online supplemental table 5).

    Half of the top 20 diseases were the same between THCs and CHs, mainly infectious or inflammatory diseases, including chronic obstructive pulmonary diseases (COPDs), acute or chronic bronchitis, acute tonsillitis and unspecified upper respiratory infections, and unspecified bacterial pneumonia, chronic gastritis and gastroenteritis and colitis, vertebro-basilar artery syndrome, and full-term uncomplicated delivery (online supplemental tables 4 and 5).

    It was notable that a large proportion of patients were hospitalised due to potentially avoidable or preventable hospitalisations (PAHs),26 such as pneumonia, diabetes, hypertension,COPD, pelvic inflammatory disease, ear, nose and throat infections were among the top 20 diseases of hospitalisation. These hospitalisations were considered potentially preventable because that they could be effectively treated at outpatient clinics or managed by primary healthcare.26 The finding pinpointed areas that warranted improvement and strengthening of primary healthcare (online supplemental tables 4 and 5).

    Coverage of surgeries by system

    There was no surgery in THCs in the rich and poor counties, and only very few in the middle-developed FS county, which were mainly obstetrical (82, 78.1%) and digestive (13, 12.4%) surgeries. In comparison, CHs provided surgeries covering all disease systems, in which the top five were haemic and lymphatic (32.6%), female genital organs (15.4%), obstetrical (11.5%), male genital organs (8.4%) and eye (8.2%) surgeries, followed by digestive (6.0%) and musculoskeletal systems (6.0%) surgeries (online supplemental table 6).

    Number of surgery types

    The number of surgery types in THCs of the middle-developed county was 13 in total, including 4 for female genital organs, 3 for digestive system and 2 for obstetrical procedures. In comparison, CHs in the rich county had the largest number of surgery types (210), followed by the middle (156) and then the poor county (58). Surgical services in CHs of the rich and middle-developed counties covered all disease systems, while surgeries of ear, eye, nose, mouth and pharynx were absented in CHs of the poor county (online supplemental table 7)

    Top 10 surgeries

    The top 10 surgeries accounted for 95.3% of all surgeries in THCs and 54.6% in CHs. The top five surgeries provided by THCs in the middle county were other obstetric operations (43.5%), caesarean section (33.3%), procedures on haemorrhoids (7.4%), repair of hernia (3.7%) and repair of vulva and perineum (1.9%), while that for CHs were caesarean section (11.3%), procedures on haemorrhoids (7.9%), cholecystectomy (7.8%), appendectomy (6.2%) and cataract surgery (5.1%) (online supplemental table 8).

    Discussion

    This study had four major findings. First, we found that the clinical capability of the nine studied THCs, irrespective of economic development status of their regions, was suboptimal against the national standard (coverage of essential clinical services ranged from 24% to 70%). Their main scope of clinical services were infectious or inflammatory diseases of respiratory and digestive systems, but lacked services of injury, poisoning, pregnancy, childbirth and surgeries. Second, THCs in both rich and poor economic development regions had lower clinical capability than those in the middle-development region, suggesting the impacts of economic and social conditions on inpatient services with possibly different mechanisms. Third, there was a large proportion of PAHs in the sampled THCs and CHs, such as pneumonia, diabetes, hypertension, chronic obstructive pulmonary disease, pelvic inflammatory disease, ear, nose and throat infections,27 28 which suggested the necessity of strengthening outpatient care of PHC as regard to the management of those diseases. Finally, there was a considerable overlap of inpatient services between the sampled THCs and CHs. Under the current healthcare insurance in China, treatment of the same disease in CHs cost more than in the THCs for both the insurance reimbursement and the out-of-pocket pay of patients.29 The finding indicates the needs and potentials of clarification and improvement of integrated care between the two levels of health institutions in order to reduce the costs of inpatient care.

    It is not surprised to find suboptimal services of the THCs against the national guideline. Health services and resources in western China with weaker health systems are generally lack behind those in Eastern China,30 which would be reflected in clinical services of THCs in western China. Our findings in the variation of capability of inpatient services of THCs were in accordance with findings from previous studies.9 18 19 31 In our study, THCs in the middle-developed county had a relatively higher proportion of health resources, larger share of inpatient services in the local health system and wider capability of services, while THCs in the richer and poorer counties had a relatively smaller proportion of health resource in health systems and smaller share of inpatient services and narrower range of inpatient services. Findings from the study of Zhu et al22 on workloads of basic public health tasks in the same THCs of the same three counties may help to explain the phenomenon. The authors found that the THCs in richer counties being very close to the capital city of Sichuan province could not be able to compete with many higher-level hospitals in providing clinical care. Therefore, the THCs may have focused their resources and services on public health services, which was reflected in their heavy workload in such services. In contrast, THCs in the poor county had the lowest capacity in providing both clinical services and public health services mainly because of their low health resources especially health workforce.22

    The clinical capability of THCs could be also shaped by other factors. In a study around Chengdu Sichuan, Ma and colleagues found that local economic status, financial subsidies, facilities (area of business accommodation and the number of equipment), number of health workforce and health personnel expenditures were positively associated with clinical capability of PHC institutions.13 In a study in An Hui province, Shi and colleagues found that GDP per capita, service population were major external factors influencing health service capacity of township hospitals and human resources were the main intrinsic factor, while the contribution of medical equipment was limited.14 In a study in Sichuan province, Zhu and colleagues found that local economic status and shortage of health workforce were the main factors affected the number and quality of health service of THCs.22 In a qualitative study in Xinjiang province, Kade‘er and colleagues found that factors affected the capability of THCs from the most important to the least were health workforce, financial input, facility and equipment, policy and transport.32 Previous research also found that patients were more likely to bypass primary care facilities to obtain perceived high-quality clinical services from higher-level hospitals if they were highly accessible and financially affordable, the underuse of primary care services may further diminish the capability of THCs in turn.8

    The finding of this study on overlapped ranges of inpatient services between the THCs and the CHs just adds another piece of evidence. Keeping patients in the care of THCs for diseases treatable by both THCs and CHs has a number of important implications in reducing care cost of both patients and insurance agency or government,33 in establishing a practical gate-keeping procedure of the THCs,34 in setting up integrated continuous care plan of patients after discharge35 and in strengthening capability of THCs. However, realising the implications would involve understanding the difference in quality and safety between hospitals for the specific diseases in the two levels, which presents an important research area in near future.21

    As regards disease composition, there were a large proportion of PAHs both in the THCs and CHs.36 This finding was supported by a recent national-wide study in china which found that PAHs due to chronic diseases consisted of a large proportion of hospitalisations—14.09% in primary care institutions and 9.20% in CHs, with COPD, diabetes and hypertension as leading causes.37 38 However, it was a pity that PAHs due to infectious diseases were not measured. Nevertheless, the examination of disease composition from a perspective of PAHs and regional comparisons would help to identify weakness and priorities of improvements of primary care and health systems.26 39 On the other hand, the largely overlapped inpatient services of THCs and CHs provide an opportunity or need of integrated care among the two levels of health institutions, which might improve quality of care and patient health outcomes.40 41

    The study has several limitations. First, the use of disease spectrum and surgeries based on only inpatients medical records did not reflect quality of care nor a full clinical capability of THCs. Further study on outpatient clinical services and care quality of THCs using standard patients approach is guaranteed. Second, the analyses of spectrum of clinical services were based on inpatient medical records and relevant coding, and the accuracy of coding among different counties might influence the results. Third, the study purposely selected representative THCs and CHs for the investigation and only data from January to December 2015 was obtained and analysed, the number of sampled institutions was small, the findings warrant further verification with larger sample size, the timeliness of this study was limited and further study using data of more recent years would be valuable. In addition, results based on descriptive analysis cannot be generalised beyond the sampled counties, although the approach to explore clinical capability of THCs could be extendable. Finally, the examination of PAHs would be more meaningful if administrative data of all health institutions were obtained and the prevalence be compared between counties, which warrants further investigation.

    The findings in our study have clear implications for further primary healthcare and health systems planning in rural China. First, the clinical service of THCs should be strengthened according to the health needs of local people and the guideline of national standard of capability building of THCs. Second, the clinical services provided in the THCs and CHs should be clearly clarified and coordinated so that the two level of institutions do not compete with each other but form more continuous and integrated healthcare systems in improving the health of rural people. Third, the planning of clinical services of THCs in rural China can be different according to the local socioeconomic, demographic and health system contexts. For THCs in remote and less developed areas with poor/difficult transportation, more clinical services might be planned to meet local health needs and improve the accessibility of healthcare, which might include oral clinic services, simple surgeries, lifesaving emergency services, etc. For THCs in top affluent and developed areas, clinical services might be strengthened for management of chronic diseases to prompt reasonable use of health services, sparing hospital services for more severe and urgent conditions, reducing the overall consumption of health resources and curbing the rapid increase of health expenditure.

    Conclusions

    The clinical capability of selected THCs was suboptimal according to the national standard, which may be affected by multiple factors including health resources, health workforce of THCs, their shares in local health systems and competition from high-level hospitals, the size of serving population and local economics. The findings also highlighted absent area of clinical services and priorities of clinical capability building of THCs. A large proportion of potentially preventable hospitalisations were among top reasons of admissions in both investigated THCs and CHs, indicating necessities and priorities of improvements of rural primary healthcare and county health systems. The large overlap of inpatient services between THCs and CHs offered opportunities and challenges of integrated care between the two levels of health institutions for better care and health outcomes of rural residents.

    Data availability statement

    No data are available.

    Ethics statements

    Patient consent for publication

    Ethics approval

    This was a retrospective observational study using secondary administrative data. For working with three specific counties, this study was approved by the Medical Ethics Committee of Sichuan University (K2017034).

    References

    1. ↵
      1. WHO
      . Ten years in public health 2007-2017: from primary health care to universal coverage –the “affordable dream”; 2017.
    2. ↵
      1. Li X,
      2. Lu J,
      3. Hu S, et al
      . The primary health-care system in china. Lancet 2017;390:2584–94. doi:10.1016/S0140-6736(17)33109-4
      OpenUrlCrossRefPubMed
    3. ↵
      1. Yip WC-M,
      2. Hsiao WC,
      3. Chen W, et al
      . Early appraisal of China’s huge and complex health-care reforms. Lancet 2012;379:833–42. doi:10.1016/S0140-6736(11)61880-1
      OpenUrlCrossRefPubMedWeb of Science
    4. ↵
      1. The National Health Committee
      . Bulletin of health statistics and development in china. Department of Planning dait, 2019.
    5. ↵
      1. Yang L,
      2. Sun L,
      3. Wen L, et al
      . Financing strategies to improve essential public health equalization and its effects in China. Int J Equity Health 2016;15:194. doi:10.1186/s12939-016-0482-x
    6. ↵
      1. Zhao KJ,
      2. Zhao L,
      3. Duan ZQ, et al
      . Development trends of heath technicians of township health centers in sichuan province since the new health care reform [in chinese]. Modern Preventive Medicine 2020;47:3732–6. Available: https://kns-cnki-net-443.webvpn.scu.edu.cn/kcms2/article/abstract?v=3uoqIhG8C44YLTlOAiTRKibYlV5Vjs7i8oRR1PAr7RxjuAJk4dHXoglq__8LEEVN8henf3Y3y-KBc-Xs_dRu2kWazRM0GvKh&uniplatform=NZKPT
      OpenUrl
    7. ↵
      1. NBS
      . National data. National Bureau of Statistics; 2021.
    8. ↵
      1. Liu Y,
      2. Zhong L,
      3. Yuan S, et al
      . Why patients prefer high-level healthcare facilities: a qualitative study using focus groups in rural and Urban China. BMJ Glob Health 2018;3:e000854. doi:10.1136/bmjgh-2018-000854
    9. ↵
      1. Yang L,
      2. Yang YQ,
      3. Li YQ, et al
      . Research of current status of township hospital service capabilities in minority regions of Sichuan Province [in Chinese]. The Chinese Health Service Management 2012;285:210–2. Available: https://www.cnki.com.cn/Article/CJFDTOTAL- ZWSG201203020.htm
      OpenUrl
    10. ↵
      1. Wu D,
      2. Lam TP
      . Underuse of primary care in China: the scale, causes, and solutions. J Am Board Fam Med 2016;29:240–7. doi:10.3122/jabfm.2016.02.150159
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Li X,
      2. Zhang L,
      3. Li Z, et al
      . Patient choice and willingness toward gatekeepers as first-contact medical institutions in Chinese tiered healthcare delivery system: a cross-sectional study. Front Public Health 2021;9:665282. doi:10.3389/fpubh.2021.665282
      OpenUrl
    12. ↵
      1. Liu Y,
      2. Kong Q,
      3. de Bekker-Grob EW
      . Public preferences for health care facilities in rural China: a discrete choice experiment. Soc Sci Med 2019;237:112396. doi:10.1016/j.socscimed.2019.112396
      OpenUrlPubMed
    13. ↵
      1. Ma A,
      2. Wang C,
      3. Zhang L, et al
      . Influencing factors of medical service capabilities in primary health care institutions, Chengdu [in Chinese]. Modern Preventive Medicine 2018;45:1603–7. Available: https://www.cnki.com.cn/Article/CJFDTOTAL- XDYF201809016.htm
      OpenUrl
    14. ↵
      1. Shi YY,
      2. Dai T,
      3. Chen Y, et al
      . Analysis of influencing factors of health service capacity of township hospitals in Anhui [in Chinese]. Med Soc 2013;26:43–7. Available: https://www.cnki.com.cn/Article/CJFDTOTAL-YXSH201306015.htm
      OpenUrl
    15. ↵
      Notice of the National health and family planning Commission on strengthening the safety and security of mothers and infants, department of maternal and child health, National health and family planning Commission No. 42. 2017.
    16. ↵
      Guidance on further strengthening Hospital delivery of pregnant women in rural areas, department of maternal and child health of the Ministry of health No. 12. 2009.
    17. ↵
      1. National Health Commission of the People’s Rupublic of China
      Guidelines for the evaluation of service capacity of township healthcare centers. 2019. Available: http://www.nhc.gov.cn/jws/s2908/201904/523e5775cdba451a81ab2fbc0628d9f0.shtml
    18. ↵
      1. Wang C,
      2. Zhang L,
      3. Yu XS, et al
      . A study on the present situation of the service ability of the primary health care facilities in Chengdu-from the perspective of basic medical services [in Chinese]. Health Economics Resarch 2018;372:48–51. Available: https://kns.cnki.net/kcms2/article/abstract?v=3uoqIhG8C44YLTlOAiTRKibYlV5Vjs7i0-kJR0HYBJ80QN9L51zrP0qia3ncI4s986OvTEddmvWiFeHckKPO8P1RgMhZY_mU&uniplatform=NZKPT
      OpenUrl
    19. ↵
      1. Li Z,
      2. Yang J,
      3. Wu Y, et al
      . Challenges for the surgical capacity building of township hospitals among the central China: a retrospective study. Int J Equity Health 2018;17:55. doi:10.1186/s12939-018-0766-4
    20. ↵
      1. Cheng Z,
      2. Tao H,
      3. Cai M, et al
      . Technical efficiency and productivity of Chinese county hospitals: an exploratory study in Henan province, China. BMJ Open 2015;5:e007267. doi:10.1136/bmjopen-2014-007267
    21. ↵
      1. Fang YA,
      2. Jiang SH,
      3. Jiang P, et al
      . Are rural primary care providers able to competently manage common illnesses? A cross-sectional study in rural Sichuan, Western China. Healthcare (Basel) 2022;10:1750. doi:10.3390/healthcare10091750
    22. ↵
      1. Zhu M-R,
      2. Yang X-B,
      3. Yang M
      . Study of the workload and human resources demand of basic public health services in township hospitals of Sichuan Province. Sichuan Da Xue Xue Bao Yi Xue Ban 2021;52:783–8. doi:10.12182/20210960103
      OpenUrl
    23. ↵
      1. Gao MY,
      2. Cheng LL,
      3. Qiu PY, et al
      . The change analysis of disease spectrum of NCMS inpatients in jiuzhaigou County, Sichuan Province, from 2009 to 2014 [ in Chinese ]. Modern Preventive Medicine 2016;43:4110–7. Available: https://kns.cnki.net/kcms2/article/abstract?v=3uoqIhG8C44YLTlOAiTRKibYlV5Vjs7iAEhECQAQ9aTiC5BjCgn0Rm3UVqNSzBpNcMV-A0bUVPrAjXC7OU_Rujm_NjBtVOex&uniplatform=NZKPT
      OpenUrl
    24. ↵
      1. Liu LX,
      2. Duan ZQ,
      3. Pan JP, et al
      . Development of quality of care index system of County level hospitals based on homepage of the medical records and analysis of its applicability [ in Chinese ]. Chinese Journal of Hospital Administration 2017;33:30–4. Available: https://www.researchgate.net/publication/356986956_Development_of_the_quality_of_care_index_system_of_county_level_hospitals_based_on_the_homepage_of_the_medical_records_and_analysis_of_its_applicability
      OpenUrl
    25. ↵
      1. Qi W,
      2. Liu S,
      3. Zhao M, et al
      . China’s different spatial patterns of population growth based on the HU line. J Geogr Sci 2016;26:1611–25. doi:10.1007/s11442-016-1347-3
      OpenUrl
    26. ↵
      1. Bourret R,
      2. Mercier G,
      3. Mercier J, et al
      . Comparison of two methods to report potentially avoidable hospitalizations in France in 2012: a cross-sectional study. BMC Health Serv Res 2015;15:4. doi:10.1186/s12913-014-0661-7
    27. ↵
      1. Purdy S,
      2. Griffin T,
      3. Salisbury C, et al
      . Ambulatory care sensitive conditions: terminology and disease coding need to be more specific to aid policy makers and clinicians. Public Health 2009;123:169–73. doi:10.1016/j.puhe.2008.11.001
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. Gibson OR,
      2. Segal L,
      3. McDermott RA
      . A systematic review of evidence on the association between hospitalisation for chronic disease related ambulatory care sensitive conditions and primary health care resourcing. BMC Health Serv Res 2013;13:336. doi:10.1186/1472-6963-13-336
    29. ↵
      1. National Health Commission of the People’s Rupublic of China
      . Statistical Communiqué on the development of China’s health systems and services in 2020. 2020.
    30. ↵
      1. Dong E,
      2. Xu J,
      3. Sun X, et al
      . Differences in regional distribution and inequality in health-resource allocation on institutions, beds, and workforce: a longitudinal study in china. Arch Public Health 2021;79:78. doi:10.1186/s13690-021-00597-1
    31. ↵
      1. Li HH,
      2. Wang L,
      3. Li YP, et al
      . A comparative study on top 15 inpatient diseases in pilot township health centers in Western China from 2008 to 2010 [in Chinese]. Chin J Evid-Based Med 2011;11:622.
      OpenUrl
    32. ↵
      1. Kade‘er G,
      2. Lang AJ,
      3. Ke SQ, et al
      . Qualitative research of influencing factors of public health service ability promotion in Xinjiang township hospitals [in Chinese]. Med Soc 2017;30:13–6. Available: https://www.cnki.com.cn/Article/CJFDTOTAL-YXSH201704004.htm
      OpenUrl
    33. ↵
      1. Zhou M,
      2. Liao J,
      3. Hu N, et al
      . Association between primary healthcare and medical expenditures in A context of hospital-oriented healthcare system in China: a national panel dataset, 2012-2016. Int J Environ Res Public Health 2020;17:6917. doi:10.3390/ijerph17186917
    34. ↵
      1. Sripa P,
      2. Hayhoe B,
      3. Garg P, et al
      . Impact of GP gatekeeping on quality of care, and health outcomes, use, and expenditure: a systematic review. Br J Gen Pract 2019;69:e294–303. doi:10.3399/bjgp19X702209
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Mitchell GK,
      2. Burridge L,
      3. Zhang J, et al
      . Systematic review of integrated models of health care delivered at the primary-secondary interface: how effective is it and what determines effectiveness? Aust J Prim Health 2015;21:391–408. doi:10.1071/PY14172
      OpenUrlPubMed
    36. ↵
      1. Hodgson K,
      2. Deeny SR,
      3. Steventon A
      . Ambulatory care-sensitive conditions: their potential uses and limitations. BMJ Qual Saf 2019;28:429–33. doi:10.1136/bmjqs-2018-008820
      OpenUrlFREE Full Text
    37. ↵
      1. Cai PP,
      2. Zhang YH,
      3. Wan Q, et al
      . Analysis on potentially preventable hospitalization for non-communicable diseases based on the system of health account in China [in Chinese]. Chinese Health Economics 2019;38:13–6. Available: https://www.cnki.com.cn/Article/CJFDTotal-WEIJ201904006.htm
      OpenUrl
    38. ↵
      1. Quan J,
      2. Zhang H,
      3. Pang D, et al
      . Avoidable hospital admissions from diabetes complications in Japan, Singapore, Hong Kong, and communities outside Beijing. Health Aff (Millwood) 2017;36:1896–903. doi:10.1377/hlthaff.2017.0479
      OpenUrl
    39. ↵
      1. Lavoie JG,
      2. Wong ST,
      3. Ibrahim N, et al
      . Underutilized and undertheorized: the use of hospitalization for ambulatory care sensitive conditions for assessing the extent to which primary healthcare services are meeting needs in British Columbia first nation communities. BMC Health Serv Res 2019;19:50. doi:10.1186/s12913-018-3850-y
    40. ↵
      1. Briggs AM,
      2. Valentijn PP,
      3. Thiyagarajan JA, et al
      . Elements of integrated care approaches for older people: a review of reviews. BMJ Open 2018;8:e021194. doi:10.1136/bmjopen-2017-021194
    41. ↵
      1. Baxter S,
      2. Johnson M,
      3. Chambers D, et al
      . The effects of integrated care: a systematic review of UK and international evidence. BMC Health Serv Res 2018;18:350. doi:10.1186/s12913-018-3161-3

    Supplementary materials

    • Supplementary Data

      This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Footnotes

    • Contributors All authors made substantial contribution to this study and proved the final version. MY was involved in conception, design, data collection, writing of the manuscript; ZD, ZZ were involved in data collection and results interpretation; KZ was involved in design, data analysis, writing; JH, JZ, JP, CL were involved in study design and writing.

      MY is responsible for the overall content as the guarantor.

    • Funding China Medical Board Training for Research on the Role of Township Hospitals in the Healthcare System (RoTHiS) (16-252).

    • Competing interests None declared.

    • Patient and public involvement Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.