Primary care quality of primary healthcare institutions in china: current status and influencing factors

Abstract

Objective

This study aims to assess the current state of primary care quality within Primary Healthcare Institutions (PHIs) in China and analyze potential factors that may influence the quality of primary care provided.

Methods

Employing convenience sampling, this research utilized the Chinese Primary Care Assessment Tool (PCAT-C) to gather cross-sectional data on patients’ experiences with primary care during visits to PHIs across the country. A multivariate linear regression model was established to discuss the effects of various factors on the quality of primary care at PHIs.

Results

The study encompassed 2,063 valid cases, with an overall PCAT score of 71.31 (SD = 11). Among the different dimensions, the average score for the Comprehensiveness dimension was the highest at 77.78 ± 15.83, followed by Continuity (75.81 ± 10.68), Family-centered (75.67 ± 17.69), First Contact (69.46 ± 10.36), Coordination (69.39 ± 16.84), and the Community-orientation dimension with the lowest average score at 59.75 ± 16.93. The regression analysis (R² = 0.0437, p < 0.000) indicated that the patient’s area(Eastern, Middle or Western) (p < 0.001), region(Urban or Rural)(p = 0.026), presence of adult children (p = 0.008), working status (p = 0.003), family annual income (p = 0.007), participation in commercial medical insurance (p = 0.001), seeking medical attention for illness was sought (p = 0.014), the number of hospital beds per 10,000 population (p = 0.007), and per capita health expenditure as a percentage of GDP per capita (p = 0.021) significantly impact the quality of primary care at PHIs.

Conclusion

The quality of primary care for PHIs in China is still low and is affected by both micro and macro factors. The quality of primary care at PHIs in China is still relatively low, particularly in the Community-orientation dimension. Apart from being influenced by microscopic factors, the quality of primary care at PHIs is also affected by macro factors such as overall health expenditure and economic status. It is recommended that efforts be made to enhance the equity of the medical and health resource allocation mechanism, thereby promoting balanced regional development.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13690-025-01663-8.

Text box 1. Contributions to the literature
• Provides nationwide data on primary care quality in China’s PHIs, filling the gap of regional research limitations.
• Integrates micro (demographics, insurance) and macro (health resource allocation) factors to analyze influencing factors comprehensively.
• Systematically evaluates the impact of human, material and financial resources on primary care, offering new insights for policy-making.

Introduction

While definitions of primary care may vary slightly among different organizations and scholars worldwide [1–4], they generally share a common essence. Primary care can be summarized as the provision of accessible, comprehensiveness, continuous, and coordinated care to individuals by clinical physicians at primary care institutions, regardless of gender, age, or health status, all within the context of family and community. In China, primary care is specifically defined as offering “comprehensiveness, continuous, and convenient intermittent and preventive health care services to family units within the community“ [5].

The role of primary care in enhancing the performance of health systems has been extensively studied both in China and internationally [6, 7]. A robust primary care system is linked to improved public health, reduced socioeconomic disparities in health outcomes, decreased unnecessary hospitalizations, and a slowdown in the overall growth of health care expenditures [7]. Primary Healthcare Institutions (PHIs), which are the foundational units of China’s three-tier medical system, include community health service centers, street and township health centers, village clinics, and outpatient clinics. These account for over 94% of all medical and health institutions in the country. In China, primary care is primarily provided by general practitioners at community health centers and other local health facilities [8].

Recognizing the value and effectiveness of primary care, many countries, including China, have identified the transformation of primary care as a critical component of health reform. In 2009 [9], China initiated a nationwide systemic health reform, with the government setting clear objectives to strengthen primary care. A series of reforms have been implemented [10], including investments in infrastructure and general practitioner training, and measures to reduce over-prescription by prohibiting primary care providers to profit from drug sales. Despite these efforts, existing health insurance policies and gatekeeper systems have not effectively limited patients from seeking primary care outside of designated primary care institutions. To address this issue, improving the quality of primary care provided by PHIs is considered the most effective solution. Although the government has made significant investments and policy efforts in primary care [5, 11], there remains an urgent need for evidence on the quality of primary care services at PHIs and their influencing factors, with the aim of further improving the standard of primary care at PHIs and fully leveraging their role in a tiered medical treatment system.

While numerous disease-centered or task-oriented studies have assessed primary care quality in China [12–17], such as a cross-sectional study conducted by Jin H and colleagues using the Chinese Primary Care Assessment Tool for Adults at a community health center in Shanghai, there is a lack of national research on the quality of primary care at PHIs and its influencing factors. Enhancing the quality of primary care at PHIs could significantly narrow the gap between patients’ expectations of medical service providers and the actual quality and effectiveness of the medical services provided in China, thereby offering patients more effective treatment options and maximizing the role of PHIs in primary care.

This study aims to understand the quality of primary care at PHIs in China by collecting data on primary care services and health resource allocation from patients visiting PHIs across all provinces (excluding Hong Kong, Macau, and Taiwan) and identify the factors that influence the quality of primary care at PHIs. The goal is to provide scientific guidance for relevant departments to enhance the quality of PHI’s primary care services. The study also holds significance for the implementation of policies such as the two-way referral system, the integration and enhancement of medical and health information systems, and the optimization of PHI service quality.

Methods

Include variables

Extensive literature searches were conducted for this study to gather evidence on factors that may impact the quality of primary care. A variety of factors were identified from the literature as potentially influential to the quality of primary care. Key demographic and socioeconomic factors were taken into account, including age [17, 18], area(Eastern, Middle or Western) [17, 19], region(Urban or Rural) [10, 20], educational level [17, 19], working status [19, 21] and income [19, 22]. To explore a broader range of microscopic factors, gender and the presence of adult children were also considered, providing a more comprehensiveness view of demographic and socioeconomic factors. Additionally, the type of health insurance [22, 23] and seeking medical attention for illness [17, 24, 25] was sought were examined. It was observed that patients with social or commercial insurance tend to perceive a higher quality of primary care. Furthermore, variations in the quality of primary care were noted across different provinces, with examples such as Guangdong having better primary care quality compared to Shanghai [17, 19]. To delve deeper into the intrinsic factors contributing to these differences, the medical resource allocation in each province was also considered. According to the “2022 China Health Statistics Yearbook,” the number of PHIs in China accounts for 95% of the total, and the medical resource allocation in each province is a significant reflection of the medical resource allocation for PHIs in that area. The allocation of medical resources can be measured by three indicators: human factors, material factors, and financial factors (see Table 1). Through expert consultation, this study established a theoretical framework for the influencing factors of primary care quality at PHIs and categorized the influencing factors accordingly (see Fig. 1; Table 1).

Table 1.

An overview of variables

Category	Group	Variable
Microscopic factors	Demographic characteristics	Gender, age, area, region
	Socio-economic characteristics	Educational level, working status, family annual income, presence of adult children
	Health care factors	Basic health insurance type, commercial medical insurance participation, seeking medical attention for illness
Macro factors: Allocation of health resources	Human factors	Number of licensed physicians per 10,000 population, number of registered nurses per 10,000 population
	Material factors	Number of hospital beds per 10,000 population
	Financial factors	Local Health Expenditure as a Percentage of Fiscal Expenditure, per capita health expenditure as a percentage of GDP per capita

Fig. 1.

The conceptual model of influencing factors of primary care quality

Microscopic factors

Microscopic factors encompass demographic characteristics, socioeconomic characteristics, and health care factors. demographic characteristics includes gender, age, area (Eastern, Middle or Western) and region (Urban or Rural). Socioeconomic characteristics consist of educational level, working status, family annual income and the presence of adult children. Health care features include participation in basic health insurance, commercial medical insurance participation and seeking medical attention for illness. These variables will be obtained through a questionnaire survey.

Macro factors

Macro factors pertain to the allocation of health resources at the provincial level, covering three aspects: human resources, material resources, and financial resources. In terms of human factors, this includes the number of licensed physicians per 10,000 population and the number of registered nurses per 10,000 population. Material factors include the number of hospital beds per 10,000 population. Financial factors encompass local health expenditure as a Percentage of fiscal expenditure and per capita health expenditure as a percentage of GDP per capita. These variables are sourced from the “2022 China Health and Health Statistics Yearbook.”

Population and sample

From July to September 2021, we conducted a cross-sectional survey across 22 provinces, 5 autonomous regions, and 4 municipalities in China (excluding Hong Kong, Macau, and Taiwan due to significant differences in public health systems). The survey included 106 cities, with two PHIs with high patient volume selected per city, totaling 815 institutions. It was planned to collect five questionnaires from each institution. The survey utilized the Chinese Primary Care Assessment Tool (PCAT-C) to gather cross-sectional information on patients’ experiences with primary care during visits to PHIs. Inclusion criteria for participants were as follows: (1) aged 18 or above, (2) received primary care from PHIs, (3) would typically choose the research location for primary care if needed, and (4) read and agreed to participate after reviewing the informed consent document. Exclusion criteria included: (1) being in poor health and unable to complete the survey, and (2) inability to understand the questionnaire. This study has been reviewed and approved by the Ethics Committee of China Pharmaceutical University.

Measurement

The Chinese Primary Care Assessment Tool (PCAT-C, 36 items) was used to assess participants’ experiences with primary care. This tool was designed by Professor Barbara Starfield and Professor Leiyu Shi of the Primary Care Policy Center at Johns Hopkins University and was initially used in the United States [26]. It has since gained international recognition and has been applied in various health systems across different countries, including Canada [27], Spain [28], Brazil [29], South Korea [16] and China [29]. These applications represent the levels of primary care provided in different regions and countries and can assist by providing specific and targeted directions for improvement. The tool includes six core attributes of primary care [26]: first contact, continuity, coordination, comprehensiveness, family-centeredness, and community orientation. A Chinese version developed by the Johns Hopkins research team was tested using adult samples from southern (Guangdong) [8, 30] and western (Tibet) [31] China. The sample testing confirmed the reliability and validity of PCAT-C, with a cumulative variance of 58.91% and a Cronbach’s α of 0.74 [30].

Patients’ experiences were measured using a 4-point Likert scale, ranging from 1 (definitely would not) to 4 (definitely would). Additionally, a neutral response of “uncertain/don’t remember” was used to indicate a lack of familiarity with a particular item. The Likert scale scores were multiplied by 25 and converted to a scale of 0 to 100, where higher scores indicate better performance. The score for each dimension is the average of all converted scale scores within that dimension, and the total score is the average of the seven-dimension scores, reflecting the overall evaluation of patients’ experiences with primary care.

A pilot survey of the questionnaire was conducted among patients at three PHIs in Nanjing, Jiangsu Province, China, and the scale was found to have acceptable reliability and validity (Cronbach’s α for the PCAT-C test was 0.783, KMO was 0.865).

Data collection

Undergraduate students with medical backgrounds were selected as research assistants. Prior to the formal survey, they received training on the research background, objectives, etiquette, communication skills with potential participants, methods for handling emergencies, and the use of research software.

Based on the principle of geographical proximity, the number of PHIs was evenly distributed among 53 research assistants, with each person responsible for four PHIs in two cities. If the number of participants in a city did not meet the requirements, the research assistant could consult with the researcher, who would then contact other research assistants and dispatch them to collect additional data.

Potential participants were first introduced to the background, content, and purpose of the survey. Those willing to participate were taken to a quiet place to sign the consent form and complete the questionnaire using an online survey system on a smartphone or tablet. Throughout the survey process, researchers provided no opinions on the questionnaire but only the requirements or instructions for its completion. Two researchers were responsible for reviewing the uploaded data. If any errors or damage were found, the research assistant was immediately notified to make corrections, and a follow-up visit was conducted when necessary.

Data analysis

Continuous variables are expressed as the mean ± standard deviation, and categorical variables are expressed as frequencies (%). We first compiled the demographic, socioeconomic, health care, and health resource allocation data of the participants. Then, independent samples t-tests or one-way ANOVA were employed to inspect the primary care scores of various categorical variables, to make a comparison on whether there were remarkable differences in the primary care scores among different groups, to evaluate the potential influences of various factors on the quality of primary care, and to offer a reference for including variables in the subsequent regression analysis. All analyses were conducted using SPSS 23.0.

The study employed multivariate linear regression to analyze the impact of various factors on the overall primary care score and the scores of the seven dimensions. Initially, the study tested the multicollinearity and heteroscedasticity of the regression results. The results indicated that the average VIF value of the 16 explanatory variables was 2, with all VIF values being less than 5, and the p-value of the White test being less than 0.01, thus indicating no significant multicollinearity or heteroscedasticity among the explanatory variables. Subsequently, the PCAT total score and the scores of each dimension were regressed against the influencing factors using multivariate linear regression. The overall significance test of the linear regression model (overall test) was established (p < 0.001). Furthermore, on the basis of multiple linear regression, we removed the variable pairs that were not significant in the above regression to conduct a robustness test for the regression results. The robustness test results can be seen in the appendix.

Results

This cross-sectional survey included samples from patients at 819 PHIs across 31 provinces, autonomous regions, and municipalities in China. A total of 2,131 individuals were surveyed, and after excluding incomplete questionnaires with errors or omissions, 2,063 valid cases were included in the analysis.

Descriptive statistics

Table 2 presents the demographic, socioeconomic, health care, and health resource allocation profiles of the study participants. Male participants accounted for 43.6% of the total, with an average age of 40.60 ± 16.31 years. The majority of participants were urban residents (63.6%), and 54.3% were employed. The majority had obtained a university degree or higher (41.5%). In terms of health insurance, 64.4% of participants were covered by the Basic Medical Insurance for Urban and Rural Residents, while 30.4% had the Basic Medical Insurance for Urban Employees, and only 27.8% had purchased commercial insurance. Regarding the Seeking medical attention for illness, medical diseases were the most prevalent (26.3%), followed by a much lower prevalence of obstetric and gynecological diseases (7.4%). In terms of health resource allocation, the average number of practicing physicians, registered nurses, and hospital beds per 10,000 population were 25.62 ± 3.57, 35.57 ± 3.27, and 67.97 ± 7.45, respectively.

Table 2.

Descriptive statistics (N = 2063)

Item	Value
Gender, n (%)
Female	1164	56.40
Male	899	43.60
Age in years (mean, SD)	40.60	16.31
Age Group (years)
19 and below	250	12.10
20–40	786	38.10
40–60	753	36.50
60 and above	274	13.30
Area, n (%)
Eastern	889	43.10
Middle	540	26.20
Western	634	30.70
Region, n (%)
Urban	1312	63.60
Rural	751	36.40
Educational level, n (%)
Lower than high/technical school	724	35.10
High/technical school	483	23.40
University or above	856	41.50
Presence of adult children, n (%)
No	630	30.50
Yes	1433	69.50
Working status, n (%)
Employed	1120	54.30
Retired	262	12.70
Students	406	19.70
Unemployed	182	8.80
other	93	4.50
Family annual income (mean ± SD) (LOG(CNY))	11.53	0.83
Basic health insurance type, n (%)
Basic medical insurance for urban employees	628	30.40
Basic medical insurance for urban and rural residents	1329	64.40
not	106	5.10
Commercial medical insurance participation, n (%)
Yes	574	27.80
No	1489	72.20
Seeking medical attention for illness, n (%)
Medical diseases	542	26.30
Surgical diseases	308	14.90
Obstetric and gynecological diseases	152	7.40
Diseases of the eyes, ears, nose and throat	359	17.40
Oral diseases	257	12.50
Other diseases	445	21.60
Allocation of health resources (mean, SD)
Number of licensed physicians per 10,000 population	25.62	3.57
Number of registered nurses per 10,000 population	35.57	3.27
Number of hospital beds per 10,000 population	67.97	7.45
Local Health Expenditure as a Percentage of Fiscal Expenditure (%)	9.94	1.89
Per capita health expenditure as a percentage of GDP per capita (%)	6.61	1.78

SD: standard deviation, CNY: Chinese Yuan, GDP: Gross domestic product, LOG (): Take the natural logarithm

Primary care attributes

Table 3 delineates a comparative analysis of PCAT scores segmented by participant characteristics. The aggregate PCAT-C score for this survey was recorded at 71. 1 ± 11.00. Within the spectrum of six dimensions, Comprehensiveness emerged with the highest mean score, reaching 77.78 ± 15.83, closely followed by Continuity at 75.81 ± 10.68, Family-centeredness at 75.67 ± 17.69, First Contact at 69.46 ± 10.36, Coordination at 69.39 ± 16.84, and Community-orientation, which scored the lowest with an average of 59.75 ± 16.93. Univariate analysis indicated that parameters such as age, area, region, number of adult children, working status, and the rate of commercial insurance coverage significantly influenced the quality of primary health care (P < 0.05).

Table 3.

Comparison of PCAT scores by participants’ characteristics

Variable	Group	First contact	Continuity	Coordination	Comprehensive	Family-centered	Community-orientation	PCAT-C total score
Gender	Female	69.39(10.46)	75.76(10.53)	69.66(16.99)	77.85(15.84)	75.85(17.45)	59.73(16.84)	71.37(10.97)
	Male	69.54(10.24)	75.86(10.88)	69.04(16.65)	77.7(15.83)	75.43(18.01)	59.78(17.07)	71.23(11.05)
	P Value	0.756	0.831	0.407	0.827	0.591	0.942	0.759
Age Group (years)	29 and below	69.49(9.86)	75.1(10.39)	67.85(16.14)	72.94(16.87)	76.34(18.07)	59.84(17.41)	70.26(10.94)
	20–40	68.45(10.35)	75.01(10.56)	69.09(16.49)	76.07(15.52)	73.39(17.97)	59.62(16.19)	70.27(10.83)
	40–60	70.25(10.44)	76.29(10.78)	70.26(17.26)	79.9(15.84)	76.7(17.29)	59.82(17.06)	72.2(11.18)
	60 and above	70.12(10.42)	77.41(10.84)	69.26(17.29)	81.29(13.99)	78.75(16.92)	59.87(18.27)	72.78(10.74)
	P Value	0.005***	0.004***	0.223	< 0001***	< 0001***	0.994	< 0001***
Area	Eastern	68.82(10.58)	74.52(11.08)	69.25(17.19)	77.32(15.63)	75.1(18.04)	59.24(17.79)	70.71(11.51)
	Middle	70.69(10.13)	77.62(9.93)	71.53(17.82)	78.4(16.63)	77.96(16.4)	61.52(17.01)	72.95(10.8)
	Western	69.31(10.16)	76.07(10.51)	67.77(15.26)	77.9(15.41)	74.51(18.1)	58.96(15.5)	70.75(10.29)
	P Value	0.004***	< 0001***	0.001***	0.447	0.002***	0.018**	< 0001***
Region	Urban	68.37(10.37)	75.17(10.83)	69.11(16.72)	78(16.14)	74.7(18)	59.58(17.41)	70.82(11.27)
	Rural	71.36(10.07)	76.92(10.33)	69.89(17.05)	77.4(15.27)	77.36(17.01)	60.05(16.07)	72.16(10.46)
	P Value	< 0001***	< 0001***	0.314	0.408	0.001***	0.536	0.007***
Education	Lower than high/technical school	70.91(10.39)	76.76(10.74)	69.67(17.29)	78.18(15.83)	77.46(17.62)	59.07(16.81)	72.01(10.96)
	High/technical school	69.73(10.02)	75.6(10.45)	69.42(17.08)	78.38(16)	74.7(17.75)	59.95(16.92)	71.3(11.01)
	University or above	68.07(10.35)	75.11(10.72)	69.14(16.34)	77.11(15.72)	74.7(17.62)	60.21(17.05)	70.72(11.01)
	P Value	< 0001***	0.008***	0.819	0.259	0.003***	0.39	0.068*
Adult children	No	68.29(10.47)	74.58(10.82)	68.57(16.13)	73.8(15.89)	73.86(18.23)	59.76(16.83)	69.81(10.99)
	Yes	69.97(10.27)	76.35(10.58)	69.75(17.14)	79.53(15.49)	76.47(17.39)	59.75(16.99)	71.97(10.95)
	P Value	0.001***	0.001***	0.142	< 0001***	0.002***	0.984	< 0001***
Working status	Employed	69.1(10.32)	75.51(10.6)	69(16.88)	77.83(16.08)	74.35(17.89)	59.11(16.78)	70.82(11.09)
	Retired	69.76(10.82)	76.38(10.82)	70.72(17.34)	81.52(14.25)	78.1(17.57)	61.51(18.59)	73(11.17)
	Students	68.62(10.26)	75(10.45)	69.06(16.16)	73.97(15.95)	75.93(17.6)	60.6(16.56)	70.53(10.7)
	Unemployed	71.99(9.79)	77.89(11.29)	70(16.9)	80.4(14.02)	78.31(16.8)	58.53(16.44)	72.86(10.18)
	other	71.56(10.22)	77.25(10.62)	70.63(17.87)	78.26(16.47)	78.36(16.33)	61.16(16.21)	72.87(11.63)
	P Value	0.001***	0.013**	0.537	< 0001***	0.002***	0.135	0.003***
Basic health insurance type	Basic medical insurance for employees	68.8(10.83)	75.51(11.33)	69.77(17.66)	79.49(16.42)	75.14(18.7)	60.35(18.13)	71.51(11.99)
	Basic medical insurance for urban and rural residents	69.76(10.09)	76.11(10.36)	69.33(16.64)	77.26(15.49)	76.04(17.13)	59.46(16.44)	71.33(10.6)
	No	69.53(10.82)	73.73(10.54)	67.98(14.25)	74.23(15.57)	74.12(18.46)	59.88(15.67)	69.91(9.7)
	P Value	0.157	0.061*	0.584	0.001***	0.376	0.551	0.383
Commercial medical insurance participation	Yes	69.54(10.54)	76.52(10.49)	70.75(17.57)	79.74(16.25)	76.91(18.33)	59.84(18.17)	72.22(11.45)
	No	69.43(10.29)	75.53(10.75)	68.87(16.53)	77.03(15.61)	75.19(17.42)	59.71(16.44)	70.96(10.81)
	P Value	0.828	0.06*	0.023**	< 0001***	0.049**	0.883	0.02**
Seek medical attention for illness	Medical diseases	69.89(10.44)	76.65(10.96)	70.42(16.48)	79.31(15.74)	75.74(18.48)	60.32(17.96)	72.06(11.53)
	Surgical diseases	69.3(10.39)	76.15(11.22)	68.07(16.44)	77.5(15.82)	75.4(18.09)	59.23(16.17)	70.94(10.94)
	Obstetric and gynecological diseases	68.5(12.14)	75.78(10.85)	69.55(19.33)	80.98(13.84)	76.25(17.14)	60.75(17.37)	71.97(11.15)
	Diseases of the eyes, ears, nose and throat	68.72(10.39)	74.76(9.77)	67.91(16.45)	75.91(16.33)	74.08(17.61)	57.69(15.72)	69.85(10.33)
	Oral diseases	70.37(9.56)	75.41(10.07)	70.61(17.08)	77.65(16.25)	76.23(17.06)	59.39(17.18)	71.61(10.59)
	Other diseases	69.43(9.99)	75.62(10.91)	69.49(16.77)	76.61(15.68)	76.53(17.03)	60.94(16.74)	71.44(11.04)
	P Value	0.297	0.179	0.153	0.002***	0.484	0.107	0.075*
	Total	69.46(10.36)	75.81(10.68)	69.39(16.84)	77.78(15.83)	75.67(17.69)	59.75(16.93)	71.31(11)

*p < 0.1; **p < 0.05; ***p < 0.01

Regression analysis

All variables within the model are deemed exploratory in their association with primary care quality (see Table 4). Seven regression models are respectively the regression analysis of the total score of primary care quality and the scores of six dimensions with micro factors and macro factors.

Table 4.

Results of the multivariate regression between the variables and individual and total primary care attributes scores

Variable	Group	First contact	Continuity	Coordination	Comprehensive	Family-centered	Community-orientation	PCAT-C total score
Socio-demographic factors
Gender	Female	Reference
	Male	-0.053	0.058	-0.522	0.22	-0.262	0.52	-0.007
	Age in years	-0.015	0.048	-0.012	0.179***	0.073	-0.062	0.035
Area	Eastern	Reference
	Middle	1.294*	3.745***	3.395***	2.122**	5.166***	1.813	2.923***
	Western	1.507*	2.638***	-0.839	-0.828	0.425	-4.053***	-0.192
Region	Urban	Reference
	Rural	2.347***	1.566***	0.908	-0.15	2.337**	0.79	1.3**
Education	Lower than high/technical school	Reference
	High/technical school	-0.076	0.044	0.447	2.512**	-0.884	1.707	0.625
	University or above	-1.057	0.467	0.541	3.585***	-0.052	2.365**	0.975
Adult children	No	Reference
	Yes	1.631*	1.331	1.983	3.642***	3.982**	2.495*	2.511***
Working status	Employed	Reference
	Retired	0.54	-0.284	1.995	0.625	1.59	4.762***	1.538
	Students	1.006	1.764*	2.069	2.691*	6.95***	2.856*	2.889***
	Unemployed	1.048	0.553	1.281	1.721	1.576	0.344	1.087
	other	0.327	0.012	2.089	0.669	1.493	2.931	1.253
Family annual income(LN(CNY))		-1.112***	-0.741**	-0.365	-0.873*	-0.762	-1.48***	-0.769***
Health care factors
Basic health insurance type	Basic medical insurance for employees	Reference
	Basic medical insurance for urban and rural residents	-0.505	-0.084	-1.067	-1.162	-1.123	-2.019**	-0.993
	not	0.121	-1.606	-1.653	-1.463	-2.382	-1.497	-1.413
Commercial medical insurance participation	Yes	Reference
	No	-1.355**	-1.71***	-2.171**	-3.177***	-2.594***	-0.898	-1.984***
Seek medical attention for illness	Medical diseases	Reference
	Surgical diseases	-0.431	-0.325	-2.266	-0.796	0.234	-1.273	-0.809
	Obstetric and gynecological diseases	-1.404	-0.55	-1.115	2.442*	1.49	0.239	0.183
	Diseases of the eyes, ears, nose and throat	-0.812	-1.621**	-2.418**	-2.096*	-1.159	-2.757**	-1.811**
	Oral diseases	0.614	-1.016	0.104	0.092	1.127	-1.399	-0.08
	Other diseases	-0.444	-0.787	-0.762	-1.652	1.328	0.12	-0.366
Allocation of health resources
Number of licensed physicians per 10,000 population		0.253**	0.278**	-0.18	-0.566***	0.067	-0.797***	-0.158
Number of registered nurses per 10,000 population		-0.326***	-0.17	0.085	0.444**	0.202	0.518***	0.126
Number of beds per 10,000 population		0.009	0.031	-0.176***	-0.253***	-0.115	-0.209***	-0.119***
Local Health Expenditure as a Percentage of Fiscal Expenditure (%)		0.262	-0.088	-0.336	-0.392	-0.622*	-0.467*	-0.274
Per capita health expenditure as a percentage of GDP per capita(%)		-0.134	0.084	0.213	1.047***	0.522	1.223***	0.492***

*=p < 0.1; **=p < 0.05; ***=p < 0.01

The determinants of primary care quality in China encompass a spectrum of factors including area, region, the presence of adult children, working status, family annual income, participation in commercial insurance, seeking medical attention for illness, the Number of hospital beds per 10,000 population and per capita health expenditure as a percentage of GDP per capita. Participants with higher PCAT total scores are predominantly found in central China (β = 2.923, p < 0.001) and rural settings (β = 1.3, p = 0.026), and are more likely to be those with adult children (β = 2.511, p = 0.008) or who are currently in education (β = 2.889, p = 0.003). Conversely, those without commercial insurance (β=-1.984, p = 0.001) and those treated for ophthalmological, otolaryngological conditions (β=-1.811, p = 0.014) exhibit markedly lower PCAT total scores. Higher family annual income (β=-0.889, p = 0.007) and a greater availability of hospital beds per 10,000 population (β=-0.119, p = 0.007) are also associated with reduced PCAT total scores. Interestingly, a higher ratio of per capita health expenditure to per capita GDP significantly augments PCAT total scores (β = 0.492, p = 0.021).

Examining the dimensions of primary care, the factors influencing First Contact include area, region, the presence of adult children, family annual income, commercial medical insurance participation, the number of licensed physicians per 10,000 population and the number of registered nurses per 10,000 population. The continuity of care is influenced by area, region, the presence of adult children, family annual income, working status, commercial medical insurance participation and seeking medical attention for illness, along with the number of licensed physicians per 10,000 population. Coordination is shaped by area, commercial medical insurance participation, seeking medical attention for illness and the number of hospital beds per 10,000 population. Comprehensiveness is influenced by age, area, educational level, the presence of adult children, family annual income, working status, commercial medical insurance participation, seeking medical attention for illness, the number of licensed physicians per 10,000 population, the number of registered nurses per 10,000 population, the number of hospital beds per 10,000 population and per capita health expenditure as a percentage of GDP per capita. Family-centeredness is affected by age, area, region, the presence of adult children, working status, and commercial medical insurance participation. Lastly, Community-orientation is impacted by area, educational level, the presence of adult children, family annual income, working status, seeking medical attention for illness, the number of licensed physicians per 10,000 population, the number of hospital beds per 10,000 population and per capita health expenditure as a percentage of GDP per capita.

Discussion

In this cross-sectional study, we utilized the validated the PCAT-C to evaluate the quality of PHIs primary care from the patient’s perspective. Building on these insights, we applied a regression model to dissect the determinants of PHIs primary care quality. Our findings underscore that, beyond the typical considerations of area, region, working status, family annual income, commercial medical insurance participation and seeking medical attention for illness, personal attributes such as the presence of adult children and macro-level indicators like the number of hospital beds per 10,000 population and per capita health expenditure as a percentage of GDP per capita, also exert a significant influence on PHIs primary care quality. Prior research often focused on a disease-centric approach to assess primary care quality, with a narrow sample scope confined to specific cities or regions. These studies typically examined a limited set of factors, often in isolation, lacking a comprehensiveness view. Our nationwide survey provides a representative snapshot of PHIs primary care quality in China, addressing a research gap regarding the impact of both personal and macro-level factors on PHIs primary care quality, and offering a valuable reference for quality enhancement initiatives.

Current status of phis primary care quality

In general, the PCAT scores reported by our respondents are lower than those observed among patients in the United States [32], a discrepancy that may stem from the relative underdevelopment of China’s primary health care system, especially when juxtaposed with those of developed nations. The inclusion of nearly all Chinese provinces in our survey, where there is a marked disparity in medical resource allocation between central and western provinces and a few more developed areas such as Shanghai and Beijing, could also account for the lower PCAT scores. This observation aligns with a study conducted in Shanghai [17]. Notably, the Community-orientation dimension scored the lowest, indicating a significant discrepancy with other dimensions, consistent with findings from a study in Guangdong [33]. This suggests that respondents’ experiences with community-based health services are less than optimal, potentially attributable to a shortfall in PHIs physician teams’ understanding of community health needs and their ability to adapt to the cultural nuances of the communities they serve.

Factors influencing phis primary care quality

Our regression analysis identified nine significant factors affecting PHIs primary care quality: area, region, working status, family annual income, the presence of adult children, commercial medical insurance participation, seeking medical attention for illness, the number of hospital beds per 10,000 population in the province and per capita health expenditure as a percentage of GDP per capita (refer to Fig. 2). While factors such as region, working status, family annual income, commercial medical insurance participation and seeking medical attention for illness align with previous literature, the inclusion of area, the presence of adult children, the number of hospital beds per 10,000 population and per capita health expenditure as a percentage of GDP per capita in our study offer new insights. The following discussion will synthesize these findings with existing literature to provide a more nuanced understanding of these determinants. Additionally, although we concluded that there are urban-rural disparities in primary care quality, potential selection bias may exist. The observed differences could be driven by demographic factors rather than geographical disparities. Therefore, we applied propensity score matching to ensure balanced covariates between urban and rural populations (see Appendix file 1).

Fig. 2.

An influencing model of primary care quality

Adult children enhance patients’ primary care experiences

Our analysis indicates that patients with adult children tend to score higher on the PCAT, which may be due to their generally better health status. This superior health condition likely leads to a more positive primary care experience during PHI visits [18]. A study in Jilin Province’s PHIs corroborates this, showing that patients who receive assistance and support enjoy a more satisfactory experience with PHIs [34]. Under the influence of China’s traditional culture of filial piety, adult children provide extensive intergenerational support, including financial, caregiving, and emotional support. Research has demonstrated that active intergenerational exchanges between the elderly and their adult children foster a reciprocal and equitable exchange of economic support, daily care, and emotional comfort, which in turn improves the elderly’s physical, mental, and social well-being [35, 36], leading to an enhanced primary care experience. A study from the United States echoes these findings [37]. Therefore, it is not surprising that patients with adult children report relatively higher PCAT scores.

Geographical health resource allocation affects primary care experiences

Analysis of geographical influence shows that patients in central China achieve the highest PCAT scores, with those in the eastern and western regions scoring comparatively lower. This is consistent with previous research [38] that has highlighted disparities in health resource allocation and primary care quality across China’s three major geographical areas. A Brazilian study also noted a significant consistency in primary care quality across different areas within a vast and varied country, although it did not explore the reasons for these disparities in depth [39]. Our study goes further by incorporating the health resource allocation of participants’ provinces into the regression analysis, aiming to uncover the intrinsic factors contributing to the variance in primary care quality across areas. Interestingly, the analysis reveals that areas with a higher number of hospital beds per 10,000 population tend to have lower PCAT scores, which is counterintuitive. However, in China, the hospitalization rate in PHIs is typically low [40], as residents often prefer to visit higher-level hospitals directly, bypassing PHIs for both minor and severe illnesses [38, 41]. Furthermore, according to the “2022 China Health Statistics Yearbook,” the proportion of PHIs beds within the total hospital bed count is often low, particularly in cities like Beijing and Shanghai. The expansion of higher-level hospitals as the number of beds per 10,000 population increases exacerbates the disparity in medical resource allocation between these hospitals and PHIs, potentially leading to a less favorable primary care experience for patients at PHIs. Conversely, the analysis shows that a higher ratio of per capita health expenditure to GDP per capita is associated with higher PCAT scores. This ratio is an indicator of the financial commitment to the health sector over a specific period and reflects the extent of government and societal emphasis on health and resident well-being. Evidently, in areas where a larger share of GDP is allocated to health expenditure, there is a greater investment in the health sector and a higher priority placed on resident health [42]. Consequently, individuals in these areas are more likely to access superior primary care services through PHIs, which translates into an improved primary care experience.

In conclusion, our study uncovers significant variations in the quality of primary care across different areas, predominantly influenced by the area’s economic status and the government’s financial investment in health care. This aligns with previous research [43, 44], which also rationalizes the higher quality of primary care experienced by patients in the eastern and central regions compared to those in the west. The highest PCAT scores in central China may be attributed to the most notable improvements in PHIs health resource allocation following the implementation of a tiered diagnosis and treatment system [38], leading to a more perceptible enhancement in the quality of primary care. This has resulted in a slightly better primary care experience in central China compared to other areas. It is imperative for China to consider refining its medical and health resource allocation mechanisms, with an emphasis on directing more resources to economically disadvantaged and remote areas. This approach will gradually improve regional equity, fostering balanced development across the nation.

Differential impact factors across dimensions of primary care quality

The correlation between microscopic factors and the various dimensions of primary care quality observed in this study is in line with existing literature. Our regression analysis has pinpointed the number of licensed physicians per 10,000 population and the number of registered nurses per 10,000 population as influential factors for the First Contact dimension, echoing conclusions from a study in India [45]. Given that primary care providers are the entry point into the health care system, and in PHIs, these roles are mainly filled by licensed physicians, a higher physician-to-population ratio translates into a superior First Contact experience. Conversely, an overabundance of nurses might hinder swift and effective communication with doctors, potentially degrading the First Contact experience as the number of registered nurses per 10,000 population. The sole macro factor identified to influence Continuity is the number of licensed physicians per 10,000 population. Continuity, being closely tied to primary care providers, is about establishing long-term relationships that facilitate mutual understanding and align expectations and needs. The role of licensed physicians in PHIs is thus paramount in ensuring continuous care [46]. The macro factor for Coordination is limited to the number of beds per 10,000 population, as previously discussed. For Comprehensiveness, the macro factors include the number of licensed physicians per 10,000 population, the number of registered nurses per 10,000 population, the number of beds per 10,000 population, and per capita health expenditure as a percentage of GDP per capita. Comprehensiveness necessitates a broad range of services from PHIs, tailored to address all but the rarest health needs within the population. Achieving this requires significant investment in human, material, and financial resources, all of which have a bearing on the Comprehensiveness dimension. Family-centered care, which emphasizes the family’s integral role in the assessment and treatment process, sees minimal impact from macro factors. Instead, the presence of adult children, a microscopic factor, significantly influences Family-centered care, with the highest impact coefficient among all dimensions. Lastly, the macro factors for Community-orientation parallel those for Comprehensiveness, recognizing the broader health care needs within the community and the characteristics that affect them [26]. Like Comprehensiveness, Community-orientation demands substantial resources from PHIs, thus all three areas of resources significantly influence Community-orientation.

Innovation

There are several innovative points in our research. Firstly, the coverage of the investigation in this research is larger and more comprehensive compared to previous studies. The cross-sectional investigation conducted in this research spans 22 provinces, 5 autonomous regions, and 4 municipalities directly under the Central Government, covering almost all provincial administrative regions in the country (except Hong Kong, Macao and Taiwan). Secondly, in contrast to previous disease-centered or task-oriented studies, this research incorporates micro and macro factors into the study of influencing factors of primary health care quality in PHIs for the first time, exploring more comprehensive influencing factors of primary health care. Furthermore, the macro factors in this research include a total of 5 factors in three parts: human resources, material resources and financial resources. The consideration of macro factors is more systematic and representative. These factors have scarcely appeared together in previous studies on the quality of primary health care.

Limitations

Our research has several limitations. The cross-sectional design, while effective for acquiring a large sample size, restricts the ability to establish causality. Longitudinal studies would provide a more robust analysis of cause-and-effect relationships. Furthermore, these survey data rely entirely on patient self-reports and were collected through convenience sampling. In addition to the usual recall and response biases inherent in self-reporting, this approach limits the inclusion of technical quality issues in primary care, while the use of convenience sampling may also restrict the generalizability of the findings.Future development of primary care assessment tools based on clinical data could mitigate recall bias and incorporate technical quality issues. Furthermore, given the economic, cultural, and health service development disparities across regions, our study may not encapsulate the entirety of China, despite an extensive dataset of over two thousand individuals. Nonetheless, our findings are instrumental in informing policy decisions regarding primary care, particularly in the context of an aging population, the Healthy China 2030 initiative, and the tiered diagnosis and treatment system in China.

Conclusion

This study evaluates the primary care experience of patients within China’s PHIs and the interplay between PHIs primary care quality and a spectrum of influencing factors. Our findings indicate that the overall quality of primary care in China’s PHIs is relatively low. Regression analysis has identified nine significant factors affecting PHIs primary care quality: area, region, working status, family annual income, presence of adult children, commercial medical insurance participation, seeking medical attention for illness, the number of hospital beds per 10,000 population and Per capita health expenditure as a percentage of GDP per capita. Consequently, policymakers are urged to implement or adjust strategies to enhance the equity of health resource distribution, aiming to elevate the quality of services provided by PHIs. By improving the dimensions of accessibility, continuity, comprehensiveness, coordination, and family-centeredness, the quality of PHIs primary care can be enhanced, thereby increasing service utilization rates.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

PHIs

Primary Healthcare Institutions

PCAT-C

Chinese Primary Care Assessment Tool

PCAT

Primary Care Assessment Tool

GDP

Gross Domestic Product

Author contributions

KZ, CW, EW and YH made their contributions to the conception and design of the work; KZ, CW and YH made their contributions to the acquisition and analysis of the data; YH and EW made their contributions to the interpretation of data; KZ made contributions to drafting of the work; CW and EW made their contributions to revision of the work.

Funding

None.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

The ethical approval to conduct the pilot survey, pre-test and main survey was granted by the Ethics Committee of China Pharmaceutical University (Project Number: CPU2019015). All methods of the study were carried out in accordance with the ethical standards of the Ethics Committee of China Pharmaceutical University and the principles of CFDA-GCP and the Helsinki Declaration.

Consent for publication

All authors confirm that they have obtained appropriate consent from any identifiable individuals included in the study, and that the manuscript does not contain any information or images requiring publication consent from third parties. No additional consent for publication is applicable in this study.

Competing interests

The authors declare no competing interests.

Written informed consent to participate was signed by all participated.