Governance of the risk of operating deficits in basic medical insurance funds in China: based on the data from City H for the period 2021–2025

Abstract

Background

The levels and coverage of the Basic Medical Insurance Fund (BMIF) continue to expand, and its financial position is transitioning from a high-surplus phase to one with increasing deficit risks. Accordingly, research on the governance of deficit risks in the BMIF has become increasingly urgent.

Method

Descriptive statistical analyses were conducted on the number of BMIF participants, the income and spending level of BMIF, the spending structure, the distribution of the fund’s settlement hospitals, and the main types of diseases paid (the top 20). This study investigated the correlations between the operating surplus rate of BMIF and the population aging rate, as well as the surplus rate of the critical illness assistance fund. Simultaneously, the Autoregressive Integrated Moving Average (ARIMA (p, d, q)) model was employed to forecast the operational risks of BMIF.

Results

The number of employees enrolled in the Urban Employee Basic Medical Insurance (UEBMI) scheme shows positive growth, while the number of residents enrolled in the Urban and Rural Residents Basic Medical Insurance (URBMI) shows negative growth. There is a good and increasing trend in the current balance of the Social Pooling Fund (SPF), but a decreasing trend for Individual Medical Savings Accounts (MSAs), and a deficit in Critical Illness Medical Insurance (CIMI); The number of patients seeking medical treatment in allopatry is ever-increasing yearly, and the top ten medical institutions receiving the most medical visits are all high-quality hospitals in the province or top hospitals in China. Furthermore, a phenomenon of palliative chemotherapy in hospital treatments is observed, along with the disorderly outflow of patients for routine post-operative examinations. The surplus rate of the UEBMI is negatively correlated with the population aging rate (r = −0.0185), positively correlated with the surplus rate of the CIMI (r = 0.285), and positively correlated with COVID-19 (r = 0.621). In contrast, the surplus rate of the URBMI is negatively correlated with COVID-19 (r = −0.775). The BMIF operational risk prediction model indicates that the UEBMI in City H is operating stably, whereas the URBMI presents potential risks, with certain regions approaching or falling below the safety threshold.

Conclusion

The BMIF in City H is generally operating stably. However, it also indicates that deficit risks are evolving from “periodic fluctuations” to “structural pressures.” The rigid growth in expenditures driven by population aging and the expansion of major illness protection spending (including MSAs and CIMI) jointly constitute the core drivers of deficit risk. Additionally, structural imbalances in cross-regional medical services and public health emergencies also contribute to the deficit risk of the medical insurance fund. Optimizing the population age structure, promoting public health and the quality of life, containing the prevalence of chronic diseases, and establishing a cooperation mechanism between local medical institutions and those in allopatry have become a recipe for avoiding the risk of deficits in the operation of medical insurance funds.

Introduction

Under the background of implementing the “Healthy China 2030” strategy, the level and population coverage of the Basic Medical Insurance Fund (BMIF) have continued to expand, while its financial operation has gradually shifted from a state of high surplus to one characterized by rising deficit risks. This shift highlights the urgent need for systematic research on the governance of deficit risks in the BMIF.

Current status of research

Most of the earlier studies have focused on three areas: payment system reform, actuarial calculation of contribution rates, and operational performance evaluation.

The main components of the payment system reform include the lump-sum prepayment model, payment by disease, and project payment. The lump-sum prepayment model is intended to achieve the purpose of controlling the rapid growth of total medical care expenditures by motivating hospitals to improve the efficiency of medical services and suppressing the case-average costs [1, 2]. However, whether it can avoid the risk of fund collapse caused by the increase of fund payments due to the actual factors such as population aging and the prevalence of chronic diseases (e.g., diabetes and hypertension) is yet to be tested.

Actuarial research on contribution rates began relatively early, and the corresponding models are well established. By incorporating factors such as population size and structural characteristics, economic indicators, medical expenditures, and compensation levels, payment indicators for the BMIF are determined to ensure the balance between revenues and expenditures of the fund [3–5]. Although such studies consider population structure to some extent, they often fail to adequately account for the impacts of accelerated population aging and policies exempting retired employees from medical insurance contributions on the financial balance of the BMIF. However, the intensification of population aging and the exemption policy for retired employees’ medical insurance contributions may currently constitute key practical factors triggering operational deficit risks. Neglecting to analyze the effects and adaptability of these factors within the payment mechanism limits the evaluation of the effectiveness of the existing payment system in mitigating deficit risks and prevents the provision of reliable data support for the prevention and control of BMIF operational risks.

The operation performance evaluation work has been uninterrupted since the inception of the basic medical insurance system. Zeng analyzed the financial operation of the basic medical insurance for town employees in the context of an aging society using an actuarial model, and concluded that China will have an accumulated deficit in the basic medical insurance fund for town employees in 2036 [6]. Yu analyzed the operational efficiency of the medical insurance fund for urban and rural residents using data envelopment analysis (DEA), and proposed encouraging childbirth to relieve the operational pressure on the medical insurance fund for urban and rural residents [7]. Zhao used DEA to analyze the town employees’ basic medical insurance data of 19 years for Xinjiang Corps and concluded that the accumulated balance of the region’s medical insurance coordination fund was insufficient, and the operating pressure was increasing [8]. It can be seen that the accumulated balance of China’s basic medical insurance has deviated from, or has begun to slightly change from, its previous state of “strong payment capacity and high level of accumulated fund balance”. A timely understanding of the current status of China’s basic medical insurance fund is of practical guidance to maintain the safety of the fund and the sustainable development of the medical insurance system.

The goal of this study

The BMIF in Anhui nowadays only involves two categories: “Urban and Rural Residents Basic Medical Insurance (URBMI)” and “Urban Employee Basic Medical Insurance (UEBMI)”. Medical insurance for major diseases belongs to the scope of the medical insurance policy in Anhui, and it is not necessary for the insured to pay additional fees to receive the insurance benefits. In other words, after the insured’s medical expenditures reach a certain amount, they can benefit directly from the reimbursement policy for major diseases. This study was intended to analyze the operational status of China’s BMIF from multiple perspectives, including the number of contributors, fund revenues and expenditures, spending composition, the distribution of designated settlement hospitals and the main types of diseases covered under URBMI and UEBMI. The findings are aimed to provide empirical evidence and policy suggestions for medical institutions and insurance fund administrators to better contain costs of medical services.

Anhui, a province in central China, implements the national strategy of the Yangtze River Economic Belt and takes part in the promotion of the Yangtze River Delta integration and synergistic development process. Anhui has gradually grown to be one of the provinces with the fastest socio-economic development in China. As of the end of 2019, the resident population of Anhui reached 63.66 million, with an urbanization rate of 55.81%; the GDP reached more than 3711 billion CNY, ranking 11th in China [9, 10]. China has developed several nationally recognized models of healthcare reform, among which the “Tianchang Model” of Anhui Province is a representative example, reflecting China’s approach to the development and management of medical insurance funds. According to the United Nations standards, when the population over 65 years old of a country or region reaches 14%, it is called a deeply aging society. Based on this standard, Anhui has entered a deeply aging society [11]. This is also the case in City H of Anhui. City H is a typical northern region in Anhui, with a large population of around 8.17 million and a large volume of medical insurance funds. The effectiveness of its funding and expenditure management is worth analyzing, and its rich management experience in this field is worth learning from by other regions. This study focuses on the risk of operating deficits in the BMIF and employs the Pressure State Response (PSR) model to analyze the operational status. Using indicators such as BMIF enrollment, fund revenues and expenditures, and the distribution of expenditure items, the study examines the key factors influencing operational risks in the BMIF. The objective is to provide empirical evidence and policy-relevant insights to support government efforts in enhancing revenue generation and controlling expenditures.

Theoretical analysis and research hypotheses

Operation of the BMIF and the PSR model

The PSR was initially proposed jointly by the United Nations Environment Programme and the Organization for Economic Cooperation and Development in the 1990s. It is a classic analytical framework in the field of environment and resource management, and has gradually expanded to multiple areas such as social governance and public policy evaluation [12]. Its core logic is to reveal the inherent correlation between external driving factors, system current state, and targeted intervention measures through the causal chain of “pressure state response”. Due to its clear logical structure, well-defined dimensions, and strong adaptability, the PSR framework is effective in decomposing the underlying causes of complex problems and identifying corresponding governance pathways [13].

This study introduces the PSR model as a macro-level analytical framework and employs hypotheses (H) as micro-level testing tools to systematically reveal the causes of deficit risk. Within the PSR framework applied to BMIF operations, population aging is defined as “pressure”, representing a key factor influencing deficit risk. “State” reflects the current level of operational deficit risk in the medical insurance fund under the influence of “pressure” and is operationalized in this study as the balance of BMIF. “Response” refers to the policy measures and regulatory actions adopted by relevant authorities in reaction to changes in deficit risk. Overall, the PSR model captures the dynamic operational process of the BMIF, whereby deficit risk increases or decreases in response to external conditions, prompting corresponding policy responses. This framework provides a systematic analytical perspective for the present study.

Hypothesis

The operating surplus rate of BMIF and the population aging rate

Developed countries such as the United Kingdom, Sweden, and France have been plagued by aging populations in their health insurance systems since the mid to late twentieth century [14]. Ellis et al. used panel data from Australia to find that health care costs are strongly correlated with population aging, and that the rising share of older men in the total population, especially those in poor health, will lead to increasing health care costs [15], and overwhelm health insurance funds. As of November 1, 2020, there were 264.02 million people aged 60 and above in China, accounting for 18.70% of the total population, and 190.64 million people aged 65 and above, accounting for 13.50% of the total population [16]. The higher prevalence of chronic diseases and disability among the elderly population results in more rigid demand for medical services and a higher frequency of healthcare utilization. Studies have shown that the average annual number of medical visits for the elderly is 1.5 to 2 times that of the young population. In terms of cost burden, the average hospitalization cost per visit for individuals aged over 70 is 24,448 CNY, which is 3.3 times that of individuals under 25. This is mainly due to the combination of multiple drugs for chronic diseases (such as the average annual use of 5.2 drugs for cardiovascular disease patients) and the prolonged rehabilitation period [17]. This situation objectively exacerbates the payment pressure on the BMIF and brings significant risks and hidden dangers to the long-term stable operation of the medical insurance fund. Based on this, the Hypothesis 1 is proposed.

H1:

The operating surplus rate of UEBMI and URBMI is negatively correlated with the population aging rate.

The operating surplus rate of BMIF and the major illness protection (including the relief fund for major illness and CIMI)

Major illness insurance primarily covers the eligible medical expenses that individuals are required to bear after reimbursement by basic medical insurance when they incur high medical costs due to serious illnesses. From the perspective of the original policy design, major illness insurance strengthens the medical security safety net for insured individuals through a “secondary compensation” mechanism. However, the singularity of its funding source and the potential for overlapping coverage during operation may generate new operational risks for the fund.

From the perspective of funding supply, major illness insurance is entirely dependent on allocations from the basic medical insurance fund and has not established an independent or diversified financing channel. As a result, the scale and sustainability of major illness insurance funds are directly constrained by the revenue and expenditure balance of the BMIF. From the perspective of expenditure demand, the continuous expansion of coverage and increases in reimbursement rates directly enlarge expenditure levels. Meanwhile, the release of medical demand and the natural growth of medical costs further intensify payment pressure. In addition, factors such as medical service price adjustments and rising costs of pharmaceuticals and medical consumables have led to a sustained increase in total eligible medical expenses for major illness treatment, thereby driving a corresponding rise in major illness insurance expenditures. Based on these considerations, the following Hypothesis 2 is proposed.

H2:

The operating surplus rate of UEBMI and URBMI is positively correlated with the surplus rate of the major illness protection.

The operating surplus rate of BMIF and COVID-19

In Anhui Province, China, the medical insurance policy for COVID-19 was implemented on April 1, 2023, as the policy adjustment cutoff date. Prior to this date, outpatient services were exempt from deductibles, and the reimbursement rate of the medical insurance fund was no less than 70%. After April 1, 2023, with the adoption of the “Class B, Category B management” policy, coverage for COVID-19 treatment costs was adjusted accordingly. Therefore, it is reasonable to propose Hypothesis 3.

H3:

There is a correlation between the operating surplus rate of the BMIF and COVID-19.

Materials and methods

Data extraction

Data from the H City Medical Insurance Bureau and its official website, and local government statistical yearbook were collected, which include BMIF data, major illness protection data, medical treatment in allopatry data, population aging rate, and policy text data. We also accessed the annual reports on the operation of health insurance funds from the Medical Security Bureau.

For data processing, the safety line of the basic medical insurance fund was set at 90% of the nodal fund revenue with the remaining 10% as a risk guarantee to cover late settlements from out-of-province claims, COVID-19-like contingency risk payments, etc. The population aging rate is replaced by the proportion of people aged 60 and above. The implementation of COVID-19 disease reimbursement for sudden public health emergency medical insurance policies is recorded as 1, and the cancellation policy is recorded as 0. Two researchers independently assessed the data quality, used a double-entry method to ensure error-free data entry, and reached consensus on all data conflicts.

Methods

Microsoft Excel spreadsheet software was used to fit the distribution of the sample data [18], explore the relevant data trends, and then analyze the risk of collapse of the BMIF. A range of variables relating to the operation of the medical insurance fund were analysed, including the number of contributors, the level of income and spending of the fund, expenditure composition, the distribution of the fund’s settlement hospitals, and the main types of diseases (the top 20) involved. SPSS19.0 was utilized to examine the binary correlation between the population aging rate, the operating surplus rate of the major illness protection, policy for COVID-19, and the operating surplus rate of the BMIF [19–21]. Simultaneously, the ARIMA (p, d, q) model is utilized to forecast the operational risks of the BMIF.

Results

Status of participation in BMIF

By the end of the third quarter of 2025, the total number of participants in the basic medical insurance was 8,593,500 in City H, of which 634,300 were employees, and 7,959,200 were residents. Details are shown in Fig. 1.

Fig. 1.

Population and growth rate of basic medical insurance fund (unit in A and C: persons)

As displayed in Fig. 1, the number of participants increased cumulatively by 13.2% for employees, compared with the same period of 2021 (550,000). But the growth rate has been decreasing year by year, from 2.35% in 2021 to 0.67% in the third quarter of 2025, indicating a slowing growth trend. The number of participants in URBMI shows a persistent downward trend, with a notable contraction starting in 2023 and a sharp quarter-on-quarter decline of 7.2% in the first quarter of 2024.

Analysis of income and expenditure of BMIF

The basic medical insurance in China consists of medical insurance for employees and for residents, and their income and expenditures were analyzed separately. At the end of the third quarter of 2025, the total income of the UEBMI (including SPF, Individual Medical Savings Accounts (MSAs), and Critical Illness Medical Insurance (CIMI)) was 2251.67 million CNY, the expenditure was 153.976 million CNY, and the current balance was 711.91 million CNY. Detailed information is presented in Table 1

Table 1.

Income and payment of the UEBMI of City H (unit: million CNY)

Time period		Social Pooling Fund (including maternity insurance)				Individual Medical Savings Accounts				Critical Illness Medical Insurance
		Income	Payment	Balance for the period	Balance rate for the period	Income	Payment	Balance for the period	Balance rate for the period	Income	Payment	Balance for the period	Balance rate for the period
2021	1–3	252.37	186.07	66.30	26.27%	217.66	154.77	62.89	28.89%	26.74	25.89	0.85	3.18%
	1–6	525.17	379.01	146.16	27.83%	444.90	327.63	117.27	26.36%	51.44	34.64	16.80	32.66%
	1–9	787.48	666.99	120.49	15.30%	670.83	504.71	166.12	24.76%	76.33	65.85	10.48	13.73%
	1–12	1,109.91	664.10	445.81	40.17%	936.90	701.11	235.79	25.17%	101.27	95.42	5.85	5.78%
2022	1–3	288.71	224.36	64.35	22.29%	243.06	150.66	92.40	38.02%	25.88	19.90	5.98	23.11%
	1–6	594.58	392.76	201.82	33.94%	468.06	334.26	133.80	28.59%	47.30	18.69	28.61	60.49%
	1–9	975.68	601.93	373.75	38.31%	625.01	527.80	97.21	15.55%	83.51	35.55	47.96	57.43%
	1–12	1,476.91	870.95	605.96	41.03%	804.86	716.46	88.40	10.98%	104.98	92.85	12.13	11.55%
2023	1–3	441.93	230.36	211.57	47.87%	174.23	231.71	−57.48	−32.99%	27.48	33.50	−6.02	−21.91%
	1–6	897.81	567.49	330.32	36.79%	326.46	425.67	−99.21	−30.39%	59.21	61.81	−2.60	−4.39%
	1–9	1,326.87	877.06	449.81	33.90%	507.10	625.20	−118.10	−23.29%	84.33	89.43	−5.10	−6.05%
	1–12	1,902.49	1,157.28	745.21	39.17%	679.66	846.95	−167.29	−24.61%	111.21	126.73	−15.52	−13.96%
2024	1–3	476.46	266.54	209.92	44.06%	166.00	172.28	−6.28	−3.78%	27.82	36.18	−8.36	−30.05%
	1–6	955.65	628.49	327.16	34.23%	347.34	362.34	−15.00	−4.32%	55.77	65.18	−9.41	−16.87%
	1–9	1,474.56	974.69	499.87	33.90%	532.40	549.61	−17.21	−3.23%	83.94	99.46	−15.52	−18.49%
	1–12	2,055.52	1,247.39	808.13	39.32%	702.95	713.72	−10.77	−1.53%	112.79	139.44	−26.65	−23.63%
2025	1–3	533.34	262.28	271.06	50.82%	186.53	161.67	24.86	13.33%	28.93	38.40	−9.47	−32.73%
	1–6	1,060.82	584.14	476.68	44.94%	376.64	313.24	63.40	16.83%	57.81	58.16	−0.35	−0.61%
	1–9	1,598.12	916.11	682.01	42.68%	566.77	540.92	25.85	4.56%	86.78	82.73	4.05	4.67%

Note: The income data is based on financial statements, and the payment amounts are based on the fund’s insurance business system data

As presented in Table 1, the overall balance of income and expenditure of City H’s UEBMI was maintained from 2021 to 2025, but structural risks increased. The income and expenditure in UEBMI (including SPF, MSAs, and CIMI) of 2024 increased by 25.19% and 43.81% compared with the same period of 2021 (2,148.08 vs. 2871.26 and 1460.63 vs. 2100.55 million CNY). The surplus rate decreased from 32.0% to 20.9% in 2023 and then rebounded to 26.8% in 2024. The significant decline in 2023 is mainly due to substantial losses in MSAs (balance rate −24.61%) and the conversion of CIMI into deficits (−13.96%), which together weaken the growth of the overall fund balance.

We further analyzed the URBMI data and obtained Table 2.

Table 2.

Income and payment of the URBMI of City H (unit: million CNY)

Time period		Income	Payment	Balance for the period	Balance rate for the period
2020	1–9	553550	525319	28231	5.10%
2021	1–9	578721	605442	−2672	−4.62%
2022	1–12	-	-	-	-
2023	1–12	-	-	-	-
2024	1–3	434928	201762	233166	53.61%
	1–6	578349	422544	155805	26.94%
	1–9	674053	632063	41990	6.23%
	1–12	802290	795394	6896	0.86%
2025	1–3	359180	213799	145381	40.48%
	1–6	621947	438435	183512	29.51%
	1–9	784045	625015	159030	20.28%

Note: -,Represents missing data

As presented in Table 2, the income and expenditure in URBMI (including relief fund for major illness) increased by 35.48% and 3.23% compared with the same period of 2021 (7,840.45 vs. 5787.21 and 6250.15 vs. 6054.42 million CNY). The fund balance rate of URBMI was −4.62% in September 2021. The income of the resident medical insurance fund slightly decreased from 802.29 million CNY at the end of 2024 to 784.045 million CNY at the end of the third quarter of 2025, while the expenditure significantly decreased from 795.394 million CNY to 625.015 million CNY, and the surplus rate increased from 0.86% to 20.28%.

According to official data provided by H city, the proportion of expenditure on the relief fund for major illnesses for residents has increased rapidly from 2020 to 2021. From 2022 to 2024, the compliance fees for the urban and rural residents’ relief fund for major illnesses above the deductible line increased from 728 million yuan to 1.194 billion yuan, and the corresponding compensation amount increased from 457 million yuan to 768 million yuan. The reimbursement rate increased from 62.8% to 64.3%. The rapid growth of the relief fund for major illness is significant, with an average annual growth rate of about 30%, which is significantly higher than the overall income growth rate of the resident medical insurance fund. At the same time, the reimbursement rate for compliant medical expenses for extremely poor individuals and low-income recipients remains at around 77%. The deficit risk of H city’s medical insurance fund is mainly affected by major illness expenses in terms of structural composition.

Based on the understanding of the collapse risk of the fund balance, it is necessary to further analyze the underlying mechanism of fund deficits due to chronic diseases. This should include analysis of payment pressure caused by medical treatment in allopatry, average case cost, patient outflow, disease structure, etc.

Situation analysis of medical treatment in allopatry

In the first three quarters of 2021, there were a total of 911,700 hospitalization cases (excluding maternity; the same below) by the insured in City H with an increase rate of 3.21% compared to the same period of 2020 (883,800 cases). The total expenditure and percentage of hospitalization cases in different places were displayed for 2020 and 2021 in Fig. 2.

Fig. 2.

Total expenditure and percentage of hospitalization in different places. Notes: WiCr, within county-regions; OCrWiC, outside county-regions but within the city; OCWiP, outside the city but within Anhui province; OP, out of Anhui province

As shown in Fig. 2, the total hospitalisation expenditure in allopatry increased by 1117.51 million CNY compared to 2020. There were 146,500 out-of-city hospitalizations (OCWiP plus OP) in 2021, accounting for 16.07% of the total number of hospitalizations, with OCWiP referring to outside the city but within Anhui province and OP to out of Anhui province. The number of OP hospitalizations (79,800 cases) increased at the highest rate compared with the same period of 2020 (40,200 cases), reaching 98.39%.

The case-average expenditure of hospitalization was further analyzed. The overall case-average expenditure was around 9140 CNY in 2021, with an increase of 11.89% compared with the same period of 2020 (8,180 CNY), as detailed in Fig. 3.

Fig. 3.

Case-average expenditure of hospitalization in 2020 and 2021. Notes: labels explanations are as in Fig. 2

It was found that the average cost (6,150 CNY) of inpatient stay within county-regions (WiCr) increased slightly in 2021, while all the average costs outside county-regions but within the city (OCrWiC, 10,493 CNY), outside the city within the province (OCWiP, 19,647 CNY), and out-of-province (OP, 19,900 CNY) decreased. The average cost per outgoing patient transfer is lower between January and September of 2021 compared with the same period of 2020, but with more outgoing visits.

Spending on out-of-city medical treatment accounted for 29.72% of the total expenditure, and the expenditure on out-of-city hospitalization amounted to 161.83 million CNY, with an increase of 45.37% compared with 2020 (111.27 million CNY). The distribution of out-of-city hospitalizations is relevant to the direction of the health insurance policy reform, therefore, the outflow of patients who received medical treatment in allopatry was further analyzed.

Analysis of the outgoing flow of patients for medical treatment

Medical expenditure and hospitalization data of the top ten medical institutions that received the majority of outgoing patients were collected from the medical insurance reimbursement system. Details about the medical institutions and expenditure settlement outside the city and Anhui province were shown in Table 3.

Table 3.

Hospitalization and settlement of reception institutions (unit: person-time, million CNY)

CAD	Name of reception institution	Number of persons	Total medical expenses	Direct billing attendance	Direct settlement rate	Average case cost	TR	RR	POPT
OCWP	APH	19,088	440.66	17,400	91.16%	0.023086	303.37	68.84%	14.61%
	FiAHAMU	9964	250.08	8336	83.66%	0.025098	164.80	65.90%	7.63%
	FAHBMC	7933	135.52	7534	94.97%	0.017083	87.66	64.68%	6.07%
	APC-H	4582	53.03	2475	54.02%	0.011575	28.43	53.61%	3.51%
	SAHAMC	3576	74.83	3133	87.61%	0.020926	48.06	64.22%	2.74%
	FAHAUCM	1501	28.43	1321	88.01%	0.018938	16.24	57.14%	1.15%
	APCH	1414	21.43	1295	91.58%	0.015153	13.41	62.61%	1.08%
	ASPH	1105	23.52	979	88.60%	0.021285	15.63	66.47%	0.85%
	FoAHAMU	972	19.82	871	89.61%	0.020391	13.73	69.28%	0.74%
	HPLAJSSF	639	11.34	560	87.64%	0.017747	7.35	64.79%	0.49%
OP	FUCH	1425	29.89	779	54.67%	0.020978	15.13	50.61%	0.95%
	WAHH	1295	68.59	1022	78.92%	0.052965	40.84	59.55%	0.86%
	ZH-FU	1059	30.93	542	51.18%	0.029206	16.36	52.88%	0.70%
	HH-FU	1020	33.73	497	48.73%	0.033069	18.36	54.43%	0.68%
	SFPH	961	27.00	522	54.32%	0.028091	14.76	54.66%	0.64%
	SPH	927	17.85	481	51.89%	0.019256	8.40	47.05%	0.62%
	RH-SJUSM	835	24.14	503	60.24%	0.028914	13.48	55.83%	0.56%
	FHZUMC	691	15.87	277	40.09%	0.022970	8.63	54.38%	0.46%
	PHSJUMC	669	19.59	185	27.65%	0.029281	10.09	51.51%	0.44%
	XH-SJUSM	656	16.49	228	34.76%	0.025144	8.51	51.59%	0.44%

Notes: Maternity and accidental injury data are not included, and the total reimbursement refers to the total of payments from the SPF and the relief fund for major diseases. CAD, Classification of administrative districts; TR, Total reimbursement; RR, Reimbursement rates; POPT, Percentage of person-times, APH, Anhui Provincial Hospital; FiAHAMU,The First Affiliated Hospital of Anhui Medical University; FAHBMC,First Affiliated Hospital of Bengbu Medical College; APC-H,Anhui Provincial Children’s Hospital; SAHAMC,The Second Affiliated Hospital of Anhui Medical University; FAHAUCM,The First Affiliated Hospital of Anhui University of Chinese Medicine; APCH,Anhui Provincial Chest Hospital; ASPH, Anhui Second People’s Hospital; FoAHAMU,The Fourth Affiliated Hospital of Anhui Medical University; HPLAJSSF, 901st Hospital of the People’s Liberation Army Joint Services and Security Forces; FUCH,Fudan University Cancer Hospital; WAHH,Wuhan Asian Heart Hospital; ZH-FU,Zhongshan Hospital, Fudan University; HH-FU, Huashan Hospital, Fudan University; SFPH,Shanghai First People’s Hospital;SPH,Shanghai Pulmonary Hospital; RH-SJUSM,Ruijin Hospital, Shanghai Jiaotong University School of Medicine; FHZUMC,The First Hospital of Zhejiang University Medical College; PHSJUMC,The 9th People’s Hospital of Shanghai Jiaotong University Medical College; XH-SJUSM,Xinhua Hospital, Shanghai Jiaotong University School of Medicine

In the first three quarters of 2021, there were a total of 130,600 hospitalization cases (excluding maternity and accidental injury) outside City H but within Anhui province. The total medical expenses were 2556.97 million CNY with an average inpatient cost of 19,600 CNY and an overall reimbursement rate of 65.46%, which was 1.14% points higher than the level of the same period of 2020 (64.32%). Among the 419 medical institutions where actual settlements took place in Anhui, the top 10 medical institutions in terms of the number of inpatient settlements made a total of 50,800 settlements, accounting for 38.90% of the total number of settlements outside the city but within the province. 150,400 out-of-province hospitalizations were settled with a total medical expenditure of 2,974,010,000 CNY, a case-average cost of 19,800, and an overall reimbursement ratio of 50.18%. Among the 4499 medical institutions where actual settlements took place in China, the top 10 medical institutions in terms of the number of settlements settled 9500 inpatient claims, accounting for 6.34% of the total number of out-of-province inpatient claims.

Distribution of the types and costs of diseases treated out-of-province

The top 20 diseases treated out-of-province and their corresponding medical expenditure data were extracted for patients who were under the jurisdiction of City H, but admitted by the Cancer Hospital of Fudan University, which is located outside Anhui province. Details are displayed in Fig. 4.

Fig. 4.

Top 20 diseases and medical expenditures of patients. Notes: PT, pancreatic tumours; HMTB, history of malignant tumours of the breast; TT, thyroid tumours; PC, palliative chemotherapy; PHMTP, personal history of malignant tumours of the pancreas; MC4M, maintenance chemotherapy for malignancies; C4POM, chemotherapy for post-operative malignancies; the organ names refer to malignant tumors of the corresponding body parts, for example, “rectum” refers to malignant tumour of rectum

There is a disorderly outflow of patients to medical institutions outside City H for palliative chemotherapy, thyroid tumors, cerebral infarction, and follow-up examinations after treatment for malignant tumors among the above-mentioned diseases. At the same time, medical treatment in allopatry covers many common diseases such as lung cancer, stomach cancer, breast cancer, etc. The average cost of medical treatment in allopatry is generally higher than that of local treatment. This phenomenon does not help to balance the gap between the various sickness funds and increases competition between sickness funds [22–25]

From the above analysis, a potential association is observed between the surplus rate of the major illness assistance fund and the financial safety risks of the medical insurance fund. From logical analysis, it can be inferred that there may be a correlation between the aging population rate and the COVID-19 medical insurance policy. We further analyzed the correlation between the population aging rate, the operating surplus rate of the major illness protection, the policy for COVID-19, and the operating surplus rate of the BMIF. The surplus rate of the UEBMI is negatively correlated with the aging population rate (r = −0.0185), positively correlated with the surplus rate of the CIMI (r = 0.285), and positively correlated with COVID-19 (r = 0.621), Hypothesis 1 and 2 are confirmed, Hypothesis 3 is denied. The surplus rate of the URBMI is negatively correlated with COVID-19 (r = −0.775), thereby supporting Hypothesis 3.

Model forecast

ARIMA (p, d, q) was used to fit the fund safety forecast model with the data of City H’ BMIF [26], and obtained the corresponding forecast model results of the safety line of UEBMI, as shown in Fig. 5.

Fig. 5.

Forecasts from AEIMA(2,1,0) of UEBMI. Notes: the red solid line represents the current balance of UEBMI, the gray solid line represents 10% of annual income of UEBMI, the red dashed line represents forecast for balance of UEBMI in the future, the gray dashed line represents forecast of annual income of 10% of UEBMI in the future, and the blue shaded area represents the 95% confidence interval

Figure 5 shows that the current balance of the UEBMI prediction results remains above the 95% confidence interval of the safety line prediction. Meanwhile, the 95% confidence interval of the safety line is partially included in the current balance of the UEBMI prediction confidence interval, accounting for less than 25%.

The corresponding forecast results for the URBMI safety threshold are presented in Fig. 6.

Fig. 6.

Forecasts from AEIMA (2,1,0) of URBMI. Notes: the red solid line represents the current balance of URBMI, the gray solid line represents 10% of annual income of URBMI, the red dashed line represents forecast for balance of URBMI in the future, the gray dashed line represents forecast of annual income of 10% of URBMI in the future, and the blue shaded area represents the 95% confidence interval

Figure 6 shows that the current balance of the URBMI prediction results remains above the safe line prediction value, but the current balance of the URBMI prediction value is lower than some values in the confidence interval of the safe line prediction.

Discussion

Analysis of basic medical insurance enrollment reveals steady growth in employee coverage, while resident enrollment shows a gradual decline. The growth rate for employee enrollment has slowed, whereas resident enrollment growth fluctuates significantly and has even turned negative. Resident enrollment now approaches ten times that of employees, with the combined total of both groups exhibiting a shrinking trend. The positive growth in employee medical insurance enrollment may be attributed to the expansion of China’s higher education system. Against the backdrop of sustained economic growth, a large influx of graduates entering the job market has become a new driving force for enrollment. The recovery of employment following the lifting of COVID-19 lockdowns may also contribute to the increase in employee medical insurance enrollment. Concurrently, it is important to note that H City, as a major labor-exporting region, has seen a significant portion of its workforce enroll in employee medical insurance at their employment locations after leaving the city for work following the end of the COVID-19 pandemic, rather than remaining enrolled in the city’s system. These factors collectively contribute to the slowdown in the growth of employee medical insurance enrollment in H City. Additionally, individuals aged 60 and above in China are exempt from paying medical insurance premiums, and China’s aging population continues to grow, which may partly explain the decline in resident enrollment. Furthermore, as China’s family planning policy has progressed, the population’s fertility preferences have gradually adapted to a one- or two-child family environment. Despite the recent relaxation of the three-child policy, families show limited willingness to have a second or third child due to fears of financial strain and the pursuit of quality of life. This aligns with Zeng’s research findings [7], contributing to the negative growth in resident enrollment. This trend can be understood through indicators linking population policy, cost of living, and family well-being indices, consistent with Anderson’s research findings [14].

An analysis of revenues and expenditures under the UEBMI indicates that the current balance of the Social Pooling Fund (SPF) remains sound and continues to increase, whereas the balance of Medical Savings Accounts (MSAs) remains positive but shows a declining trend, and the Critical Illness Medical Insurance (CIMI) has entered a deficit state. In contrast, the revenue-expenditure performance of the URBMI is more volatile. Expenditures related to the major illness assistance fund have grown at an average annual rate of approximately 30%, resulting in a high risk of year-end settlement deficits. Although the increase of the number of insured employees has helped alleviate pressure on medical insurance expenditures, the slowdown in enrollment growth has constrained income expansion. Meanwhile, the sharp decline in the number of residents participating in the insurance program has weakened the revenue base of the fund. Coupled with population aging and the resulting rise in per capita medical expenses, the URBMI is therefore facing severe deficit pressure. From the perspective of expenditure structure, the major illness assistance fund has a pronounced impact on fund balance. In 2024, total URBMI expenditures amounted to 7.954 billion yuan, of which spending on major illness assistance accounted for nearly 10%. This risk exposure is more prominent than that observed under the UEBMI, where expenditures on CIMI assistance accounted for approximately 6%. Moreover, the deficit of the CIMI is projected to widen from a negative balance in 2023 to −26.65 million yuan in 2024. Although the major illness assistance fund receives fiscal subsidies, its expenditure growth rate is higher, indicating a comparatively greater level of operational risk.

Analysis of the BMIF expenditure structure reveals a sustained annual increase in both hospitalization volumes and corresponding medical expenditures resulting from patient outflows for medical treatment. One of the possible reasons for this is related to the reform of China’s medical insurance system, which has made it easier to file for medical treatment in allopatry, with a high on-the-spot settlement rate. The policy of real-time on-the-spot settlement of medical insurance for medical treatment in allopatry has brought convenience to the public in terms of accessing medical treatment in allopatry. The second reason may be related to the shortage of local health care resources, shortcomings in medical technology, and the lack of treatment experience of local physicians, which encouraged some patients to choose medical treatment in allopatry. Other reasons may stem from the prevalence of relocation of rural residents in China, who tend to seek medical treatment close to where they are employed rather than where they pay for their medical insurance premiums.

The results of the study showed that all the top ten medical institutions for allopatric medical treatment belong to provincial high-quality hospitals or domestic top hospitals in China. It is suggested for the supervisory department of the medical insurance fund that there be a possibility of overusing medical resources for ailments and unnecessary “upgrading” of care to specialist treatment by the public when seeking medical treatment. Analyses of the types of disease treated in allopatry revealed the phenomenon of unnecessary palliative chemotherapy in hospitals and a disorderly outflow of patients for follow-up examinations after oncology treatment that could be handled properly in local areas. This could be because of patients’ dissatisfaction and bad experience with local medical treatment and services. To alleviate the pressure of medical insurance payment formed by the medical treatment in allopatry, the medical insurance administrators could suggest that local hospitals establish cooperative medical relationships with relevant medical institutions in allopatry according to residents’ preference when seeking medical treatment. Possible cooperation modes could be in the form of telemedicine, medical experts’ flexible employment in varying clinics, and professional, technical, and practical guidance in local areas. At the same time, there is an urgent need to establish a fund transfer mechanism for basic medical insurance both within and outside the region in order to ensure the safety of fund coordination [26].

The balance rate of the resident medical insurance fund declined from 5.1% in 2020 to −4.62% in 2021 and then to 20.28% in 2025. The practice of “treating minor illnesses with great care” and “treating common diseases with expert treatment” in the outgoing of patients for medical treatment may add payment pressure to the operation of BMIF before 2021. One of the reasons could possibly be the rising trend of urban and rural health insurance fund contribution standards. The contribution fees required by the URBMI increased from 120 CNY/year/person in 2016, to 220 CNY/year/person in 2018, to 320 CNY/year/person in 2021, and then 400 CNY/year/person in 2024, which, to a large extent, guarantees the sustainable growth of the fund account. Meanwhile, according to the current income status of urban and rural residents in China, there is very limited room for expansion of the contribution rate, which is consistent with the results of Yu’s study [8], and there is little room left for safe operation of fund accounts if the 10% risk guarantee fund were excluded. At the same time, the analysis of disease types in cross-regional medical treatment reveals the prevalence of palliative chemotherapy in hospital-based disease management. There is also a disorderly outflow of patients for follow-up examinations after tumor treatment that could be appropriately provided locally. Such unreasonable phenomena may stem from patients’ dissatisfaction with local medical resources or poor experience with clinical diagnosis and treatment services. Medical insurance authorities could advise local hospitals to establish cooperative medical relationships with relevant medical institutions based on residents’ healthcare-seeking preferences, such as through telemedicine services, flexible expert consultations, and on-site technical guidance by specialists. These measures can help alleviate the pressure of medical insurance payments caused by cross-regional medical treatment. Meanwhile, it is imperative to establish a fund adjustment mechanism for basic medical insurance both within and across administrative regions to safeguard the security of overall fund pooling [16, 27]. Diseases that could have been treated locally within the city, such as those requiring palliative chemotherapy, thyroid tumor treatment, cerebral infarction management, and follow-up examinations after malignant tumor therapy, are being referred to hospitals outside the province or provincial-level hospitals. Coupled with the fact that the case average cost of cross-regional treatment for such diseases is generally higher than that of local treatment, this has become a potential factor exacerbating the deficits of BMIF.

The correlation analysis results show that the balance rate of the employee medical security fund is negatively correlated with the aging rate of the elderly population, positively correlated with the balance rate of the major illness assistance fund, and positively correlated with COVID-19. The surplus rate of the residents’ medical insurance fund is negatively correlated with COVID-19. Physical function gradually deteriorates with age, occupying more medical resources compared to young people. Therefore, there are more opportunities to consume medical insurance funds, and the decrease in the surplus rate of BMIF is a natural result. The high surplus rate of the CIMI occupies resources from other medical insurance funds, thus promoting the increase of the UEBMI surplus rate. A proactive reimbursement policy has been implemented for the COVID-19 disease. During the disease progression, residents’ medical expenses rise more rapidly than their professional income or job titles; the correlation between the balance rate of UEBMI and COVID-19 is exactly opposite to that between the balance rate of URBMI and COVID-19. One possible reason is that the insured population for residents is much larger than that of employees, and the total reimbursement amount is also larger, resulting in differences in data representation. The second reason may stem from differences in cultural level and occupational attributes among employees, leading to varying levels of emphasis on health or overall differences in physical health. Certainly, other reasonable reasons cannot be ruled out.

The risk prediction model for the operation of the BMIF shows that the operation of the H city UEBMI is healthy and stable generally, while a small number of areas with operational risks for the URBMI have fallen below the potential range of the safety threshold. The health management department should be reminded to pay timely attention to factors affecting the operation risk of the medical insurance fund, especially population aging, public health emergency events (such as COVID-19), cross-regional medical treatment management, the high incidence and chronic nature of diseases, and major illnesses that may cause rapid fund depletion.

For policy-related reasons, comprehensive data on disease types in cross-regional medical treatment cannot be acquired, and only the data from 2020 to 2021 were adopted in this research. This may restrict the generalization of the research results and thus stands as a limitation of the study. In the meantime, it is undeniable that the patterns of patients’ medical settlement identified in this study provide certain theoretical significance and empirical reference value for formulating policy recommendations on the governance of operational risks in medical insurance funds.

Conclusion

The aging population, growth in the expenditures of major illness protection, and expansion of the outgoing flow of patients for medical treatment are the core factors driving the risk of deficit in the BMIF. Although UEBMI has strong risk resistance, the expansion of CIMI has weakened its balance stability. The growth rate of claims from the major illness relief fund for URBMI exceeds the growth rate of income, becoming a key trigger for deficit risk. The increase in medical expenses due to an aging population, coupled with the outflow of patients seeking medical treatment and high-cost diseases, has further intensified the payment pressure on BMIF. The risk of deficit in H City’s BMIF has evolved from “periodic fluctuations” to “structural pressure”. Promoting the health of the public population, containing the prevalence of chronic illnesses, and improving the quality of life of the public population have become the key to reducing the pressure on medical insurance funds. Establishing various medical cooperation modes between the local medical institutions and those in allopatry that receive the most patient visits is considered an effective way to reduce the number of residents seeking medical treatment in allopatry.

Abbreviations

BMIF

Basic Medical Insurance Fund

UEBMI

Urban Employee Basic Medical Insurance

URBMI

Urban and Residents Basic Medical Insurance

SPF

Social Pooling Fund

MSAs

Medical Savings Accounts

CIMI

Critical Illness Medical Insurance

DEA

Data envelopment analysis

PSR

Pressure State Response

WiCr

Within county-regions

OCrWiC

Outside county-regions but within the city

OCWiP

Outside the city but within Anhui province

OP

Out of Anhui province

CAD

Classification of administrative districts

TR

Total reimbursement

RR

Reimbursement rates

POPT

Percentage of person-times

APH

Anhui Provincial Hospital

FiAHAMU

The First Affiliated Hospital of Anhui Medical University

FAHBMC

First Affiliated Hospital of Bengbu Medical College

APC-H

Anhui Provincial Children’s Hospital

SAHAMC

The Second Affiliated Hospital of Anhui Medical University

FAHAUCM

The First Affiliated Hospital of Anhui University of Chinese Medicine

APCH

Anhui Provincial Chest Hospital

ASPH

Anhui Second People’s Hospital

FoAHAMU

The Fourth Affiliated Hospital of Anhui Medical University

HPLAJSSF

901st Hospital of the People’s Liberation Army Joint Services and Security Forces

FUCH

Fudan University Cancer Hospital

WAHH

Wuhan Asian Heart Hospital

ZH-FU

Zhongshan Hospital, Fudan University

HH-FU

Huashan Hospital, Fudan University

SFPH

Shanghai First People’s Hospital

SPH

Shanghai Pulmonary Hospital

RH-SJUSM

Ruijin Hospital, Shanghai Jiaotong University School of Medicine

FHZUMC

The First Hospital of Zhejiang University Medical College

PHSJUMC

The 9th People’s Hospital of Shanghai Jiaotong University Medical College

XH-SJUSM

Xinhua Hospital, Shanghai Jiaotong University School of Medicine

PT

Pancreatic tumours

HMTB

History of malignant tumours of the breast

TT

Thyroid tumours

PC

Palliative chemotherapy

PHMTP

Personal history of malignant tumours of the pancreas

MC4M

Maintenance chemotherapy for malignancies

C4POM

Chemotherapy for post-operative malignancies

Author contributions

All listed authors have contributed to the development, writing and revision of this scoping review. JJS, XGZ, SJH, LLX, YC, YS and LPZ was involved in the framework design, review of the articles, analysis of the data, writing and revision of the manuscript. SJH, LLX, YC, HYZ, YS contributed in the literature search and collected data. All authors read and approved the final manuscript.

Funding

This work was supported in part by the National Natural Science Foundation of China (No.72374005), the NSF Center for Basic Science Project (No.72188101), the Natural Science Foundation for the Higher Education Institutions of Anhui Province of China (No. 2023AH050561, No. KJ2021A0266, and No.2022AH051143), and the Innovative Teaching Team Project for Higher Medical Mathematic (No.2024cxtd041).

Data availability

The data used and/or analyzed during the current study is available from the corresponding author on reasonable request.

Declarations

Ethics approval

Our study was approved by the ethics committee of Anhui Medical University (No0.20190463). All methods were carried out in accordance with relevant guidelines and regulations. All experimental protocols were approved by the ethics committee of Anhui Medical University. Because the data does not involve personal privacy or animal experiments, there is no need to sign an informed consent form.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.