Differences in healthcare institution choice among elderly cancer patients: evidence from a cross-sectional study in Shandong Province, China

Abstract

Background

People over 60 years old had higher cancer morbidity and mortality. We aimed to explore the factors associated with the choice of first treatment institution among elderly cancer patients in China and to provide evidence for the development of healthcare policies.

Methods

We designed a cross-sectional study. 9 hospitals in Shandong Province of China were selected as sample institutions and a questionnaire survey was conducted on cancer patients from these hospitals. The chi-square test and binary logistic regression were used to assess the factors that influence elderly patients’ choice of first treatment institution.

Results

537 respondents(97.64%) were included in this study. Of the 537 respondents, 168 (31.28%)chose a county hospital and 369(68.72%) chose a provincial or municipal hospital for their first treatment institution. The high level of medical care is the main reason for the choice of the first treatment institution for cancer patients. Patients in the 60–69 (p = 0.002, OR = 3.392) and 70–79 (p = 0.015, OR = 2.571) age groups were more likely to choose provincial and municipal hospitals for their first treatment institution. Those in the central region(p = 0.001, OR = 0.204), with stage I cancer(p = 0.001, OR = 0.168),annual household income ¥10,000-¥50,000(p = 0.005, OR = 0.327), and unmarried(p = 0.005, OR = 0.344) patients were more likely to choose county hospitals.

Conclusions

Elderly cancer patients had a significantly higher proportion of attending high-level hospitals. Improving the standard of care in county hospitals could effectively guide patients to seek care and reduce the pressure on high-level hospitals.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12885-026-16512-2.

Background

Population aging is a distinct global trend of our time [1]. China’s seventh population census in 2020 indicated that China’s population aged 60 and above reached 264 million, accounting for 18.70% of the total population [2]. In February 2024, the National Cancer Center(NCC) released China’s latest cancer report, which mentioned that the elderly population over 60 years old has higher cancer morbidity and mortality than other people [3]. In 2022, there were 2.79 million incidences of cancer and 1.94 million deaths among the elderly in China, accounting for 55.8% and 68.2% of the cancer morbidity and mortality rate in the whole population [4].

With this background, the healthcare seeking behavior of the elderly after suffering from cancer has become an essential area of investigation. Healthcare seeking behavior refers to the fact that when individuals feel discomfort or experience certain symptoms of illness, they seek medical care considering their conditions, financial status, and type of healthcare institution [5]. The utilization of health services by elderly people is impacted by various factors, in particular, socio-demographic, cultural, economic, and regional accessibility of health resources [6]. American scholar Ronald. M. Anderson established a comprehensive Behavioral Model of Health Services Use (BMHSU), or “Anderson Model” for short, in 1968, which is used to identify factors that affect healthcare utilization. Since its inception, the Anderson model has undergone continuous evolution and refinement. Its development can be broadly divided into three stages: The first stage consists of the initial model, which focuses primarily on healthcare utilization behavior at the household level. It categorizes influencing factors into three main types, propensity characteristics, enabling resources, and needs, and tends to provide a static description; The second stage is the revised model, in which Anderson et al. expanded the scope of research from the household to the individual level, emphasized the importance of health behaviors, introduced outcome variables, and added an “environment” dimension covering the healthcare system and the external environment [7]; The third stage is the behavioral model, which evolved into a six-tiered integrated framework of “environment-disposition-capacity-need-behavior-outcome.” It incorporates new variables such as policy perception, institutional trust, and psychosocial factors, placing greater emphasis on individuals’ perceptions and experiences of healthcare services. It highlights the dynamic interaction among the environment, personal characteristics, and behavior, making it suitable for analyzing healthcare utilization behaviors across different populations [8–11]. As a classic theoretical framework for analyzing health service utilization behavior, the latest Anderson model demonstrates high adaptability and explanatory power within China’s healthcare system. Its logic of dynamic interaction among the macro-environment, individual characteristics, healthcare behavior, and health outcomes aligns with the institutional features of China’s healthcare system, which is dominated by public healthcare, features universal health insurance coverage, yet suffers from uneven resource allocation and a pronounced urban-rural dichotomy. Numerous domestic empirical studies have confirmed that this model provides a reasonable explanation for Chinese residents’ behaviors regarding healthcare facility selection, chronic disease management, and adherence to tiered diagnosis and treatment, and is particularly applicable to analyzing healthcare decision-making among vulnerable groups such as the elderly, those with chronic diseases, and low-income individuals [12, 13].Nonetheless, there has been less previous research on the healthcare seeking behavior of elderly cancer patients, which suggests that we have a great deal of work to do in the future.

Thus, the purpose of this study was to identify the factors that affect the choice of first treatment institution among Chinese older adults once they have been diagnosed with cancer and to capture their current situation and characteristics. The results of the study could provide reliable evidence to meet the challenge of increasing health service needs of elderly cancer patients, especially to understand the reasons for their use of different medical institutions. We are striving to meet the health service needs of elderly cancer patients while proposing targeted measures to guide them to choose their healthcare institutions scientifically.

Methods

Analytical framework

This study adopts the traditional Anderson’s behavioral model created by Ronald. M. Anderson. The model is suitable for studying both the health care service utilization behavior of the population and the whole process of individual health care service utilization behavior, in which the study of the individual mainly involves the choice of medical treatment, medical costs, and disease screening. According to the model, the population’s healthcare utilization behavior is determined by three factors: predisposing factors, enabling factors, and need factors. Predisposing factors refer to characteristics of individuals that predispose them to utilize health care services prior to the onset of illness, including gender, age, marital status, place of residence, employment status, attitudes toward health care services, and level of knowledge [14, 15]. Enabling factors refer to the resources and conditions that promote or hinder access to services, including financial capacity, social support, transportation accessibility, insurance coverage, and the availability of healthcare resources [14, 16] .Need factors are divided into objective and subjective needs, such as clinically diagnosed diseases, the number of chronic conditions, self-rated health status, and symptom perception; these are the direct drivers of service utilization [14, 17]. The following is the analytical framework used in this study (Fig. 1).

Fig. 1.

An analytical framework for the use of healthcare services in elderly cancer patients

Survey subjects

We used a stratified random sampling method. Firstly, Qingdao City, Tai’an City, and Heze City in Shandong province as 3 sample cities are selected based on the level of economic development and geographic location (the geographic location of each city is shown in the Appendix). Secondly, we chose one county with a high rate of outflow and one county with a low rate of outflow as a sample county in each sample city according to the outflow of patients. Finally, we chose one hospital in each of the sample cities and counties, for a total of 3 municipal hospitals and 6 county hospitals. From the oncology wards of these hospitals, cancer patients with confirmed pathological diagnosis and aged ≥ 60 years were randomly selected as study subjects. It is important to note that our selection scope covered all types of cancer.

A questionnaire with face-to-face interviews was used to collect data, which was filled out and returned on-site. The questionnaire was originally designed by the research team to conduct this study(The contents of the questionnaire can be viewed in the supplementary file).Before the official survey, we invited clinicians to provide feedback on the questionnaire and conducted a small-scale pilot study at one hospital to ensure the questionnaire was well-designed. The official survey was conducted based on the modification of the initial survey. The interviewers were uniformly and rigorously trained. We spoke to each elderly person individually and told him/her that we would hold the conversation in confidence and that it would be conducted with the informed consent of the elderly person.

Sample size estimation was conducted for this study, as described below: This study is based on a large-scale cross-sectional survey of cancer patients across all age groups. The parent survey applied Kendall’s sample size estimation guidelines (10 times the number of independent variables) and incorporated a 10% attrition adjustment to account for invalid questionnaires and non-response, resulting in a target sample size of 660 cases. To further enhance the representativeness of the sample, the survey scale was appropriately expanded in accordance with the actual admission capacities of hospitals at various levels. Ultimately, 1,100 questionnaires were distributed, and 1,077 valid questionnaires were returned, resulting in a valid response rate of 97.91%. This study constitutes a subgroup analysis of the elderly population from the aforementioned parent dataset, including only cancer patients aged 60 years and older, after screening, a valid sample of 537 cases was obtained.

Variables and their measurements

The outcome variables of this study were county-level hospitals and provincial/municipal hospitals. This classification was based on the administrative hierarchy and service capacity of medical institutions in China, provincial or municipal hospitals are typically tertiary hospitals with stronger specialized service capabilities, whereas county-level hospitals are mostly secondary or primary hospitals with relatively limited service scope and resource allocation [18]. Among the independent variables, household annual income refers to patients ‘self-reported disposable income after taxes. Cancer staging was derived from patients’ medical records using the TNM staging system. The basic medical insurance for urban and rural residents in Shandong Province includes coverage for major diseases, primarily implemented through the “Urban and Rural Residents Critical Illness Insurance” program. The “first treatment institution” was defined as the medical institution where the patient first received anti-cancer treatment after being diagnosed with cancer.

Data analysis

Univariate analysis was used in this study to examine the background characteristics of the sample. The chi-square test was applied to analyze the association between healthcare services and different behavioral factors. Subsequently, binary logistic regression was used to analyze the factors associated with elderly people’s choice of different levels of hospitals, with a difference of P ≤ 0.05 considered statistically significant. To assess the robustness of the regression findings, a sensitivity analysis was additionally performed by including all theoretically relevant variables derived from Andersen’s Behavioral Model in the multivariable logistic regression model regardless of univariate statistical significance.

Results

Characteristics of respondents

The univariate analysis in Table 1 includes the frequency distribution of the general characteristics of the 537 study participants categorized by level of medical institution. The results indicated that 68.72% of elderly cancer patients chose provincial and municipal hospitals for their first treatment, while 31.28% chose county hospitals to receive healthcare services. Among them, the percentage of males (68.02%) and females (69.95%) choosing provincial and municipal hospitals was approximately the same. The majority of patients in the age groups of 60–69 years (72.70%) and 70–79 years (68.12%) chose to be treated in provincial and municipal hospitals. There was little difference in the percentage of patients with junior high school education and below and senior high school education and above in terms of their choice of hospitals to be treated. A higher percentage of patients who were already married (70.97%) and those in the East (83.13%) and West (75.98%) regions received care at provincial and municipal hospitals.

Table 1.

Frequency distribution of elderly cancer patients choosing different levels of hospital treatment

Variables	Total n = 537		p-value
	County hospital n (%)	Provincial and municipal hospitals n (%)
Total	168(31.28%)	369(68.72%)
Predisposing factors
Gender
Male	110(31.98%)	234(68.02%)	0.644
Female	58(30.05%)	135(69.95%)
Age
60–69	77(27.30%)	205(72.70%)	0.003
70–79	66(31.88%)	141(68.12%)
≥ 80	25(52.09%)	23(47.92%)
Education level
Junior high school and below	130(32.50%)	270(67.50%)	0.299
High school and above	38(27.74%)	99(72.26%)
Marital status
Widowed/Divorced/Single	24(58.54%)	17(41.46%)	<0.001
Married	144(29.03%)	352(70.97%)
Region
East	28(16.87%)	138(83.13%)	<0.001
Central	97(50.52%)	95(49.48%)
West	43(24.02%)	136(75.98%)
Knowing the early symptoms of Cancer
Yes	34(33.00%)	69(67.00%)	0.675
No	134(30.88%)	300(69.12%)
Enabling factors
Annual household income(CNY)
<10,000	47(35.61%)	85(64.40%)	<0.001
10,000–50,000	87(40.65%)	127(59.35%)
50,000–100,000	22(19.30%)	92(80.70%)
≥ 100,000	12(15.58%)	65(84.42%)
Daily residence
Urban	40(22.35%)	139(77.65%)	0.002
Rural	128(35.75%)	23(64.25%)
Type of medical insurance
Employee basic medical insurance	29(19.3%)	121(80.67%)	<0.001
residents’ basic medical insurance	134(35.45%)	244(64.55%)
Other	5 (55.56%)	4(44.44%)
Need factors
Cancer staging
Ⅰ	25(55.56%)	20(44.44%)	<0.001
Ⅱ	36(41.86%)	50(58.14%)
Ⅲ	40(29.41%)	96(70.59%)
Ⅳ	67(24.81%)	203(75.19%)
Frequency of physical examinations
Once a year	46(24.21%)	144(75.79%)	0.015
Once in two years	10(55.56%)	8(44.44%)
Once in a while	51(34.00%)	99(66.00%)
Never	61(34.08%)	118(65.92%)
Chronic disease
Yes	70(30.43%)	160(69.57%)	0.713
No	98(31.92%)	209(68.08%)
Family history of cancer
Yes	29(25.67%)	84(74.34%)	0.147
No	139(32.78%)	285(67.22%)

The proportion of patients choosing provincial and municipal hospitals for their first treatment increased significantly after their annual household income exceeded ¥50,000. There is a large difference between the proportion of patients living in cities who choose county-level hospitals and provincial and municipal hospitals for medical services. In addition, the proportion of patients with basic medical insurance choosing provincial and municipal hospitals was significantly higher than that of county hospitals. The percentage of patients choosing provincial and municipal hospitals increases as the stage of their cancer increases. The percentage of patients who had a physical examination once every two years who chose county hospitals was higher, which was slightly different from the choice of hospitals in patients with other physical examination frequencies. Regardless of chronic diseases and family history of cancer, most of the patients chose high-level hospitals such as provincial and municipal level after cancer diagnosis. This finding may reflect a tendency for patients diagnosed with serious diseases such as cancer to prefer higher-level hospitals for treatment.

Respondents’ choice of different levels of hospitals

The results in Table 2 demonstrate the reasons elderly patients choose county hospitals for their first treatment is the proximity to their homes, accounting for 48.21% of the total. The most significant reason for choosing provincial and municipal level medical services was the high level of medical technology, accounting for 71.00%.

Table 2.

Patients’ choice of hospitals at different levels

Reason	County hospital	Provincial and municipal hospitals
	n (%)	n (%)
High medical technology	66(39.29%)	262(71.00%)
Geographic proximity	81(48.21%)	44(11.92%)
Recommended by relatives or friends	7(4.17%)	26(7.05%)
Constant working in the hospital	8(4.76%)	21(5.69%)
Other	6(3.57%)	16(4.34%)
Total	168(100%)	369(100%)

Regression analysis

Variables derived from Andersen’s Behavioral Model and variables considered clinically relevant were included in the multivariable logistic regression analysis. Additionally, sensitivity analyses were conducted using a fully adjusted model that included all theoretically relevant variables. Table 3 demonstrates the results of the binary logistic regression analysis regarding the choice of first treatment institution for elderly cancer patients. The Hosmer-Lemeshow test indicated good model fit (χ2 = 4.028, p = 0.855).All ORs were estimated with the reference group as the comparator. An OR > 1 indicates a higher likelihood of choosing provincial/municipal hospitals, while an OR < 1 indicates a higher likelihood of choosing county-level hospitals.

Table 3.

Binary logistic regression analysis of factors associated with choosing provincial/municipal higher-level hospitals as the first treatment institution among elderly cancer patients

Variables	OR	95% CI	p-value
Predisposing factors
Age
60–69	3.392	1.577–7.298	0.002
70–79	2.571	1.197–5.520	0.015
≥ 80	Ref
Marital status
Unmarried	0.344	0.164–0.724	0.005
Married	Ref
Region
East	1.022	0.526–1.983	0.950
Central	0.204	0.121–0.344	<0.001
West	Ref
Enabling factors
Annual household income(CNY)
<10,000	0.468	0.201–1.086	0.077
10,000–50,000	0.327	0.150–0.715	0.005
50,000–100,000	0.781	0.330–1.851	0.575
≥ 100,000	Ref
Daily residence
Urban	1.032	0.571–1.868	0.916
Rural	Ref
Type of medical insurance
Employee basic medical insurance	3.746	0.707–19.834	0.120
Urban residents’ basic medical insurance	3.097	0.638–15.029	0.161
Other	Ref
Need factors
Cancer staging
Ⅰ	0.168	0.078–0.358	<0.001
Ⅱ	0.644	0.359–1.157	0.141
Ⅲ	1.160	0.685–1.964	0.581
Ⅳ	Ref
Frequency of physical examination
Once a year	1.522	0.871–2.662	0.141
Once in two years	0.643	0.223–1.856	0.414
Once in a while	0.931	0.545–1.591	0.792
Never	Ref

Predisposing factors

Under the predisposing factor, age was significantly associated with the choice of first treatment institution, patients in the 60–69 (p = 0.002, OR = 3.392) and 70–79 (p = 0.015, OR = 2.571) age groups had a higher proportion of seeking treatment at provincial or municipal hospitals compared to those aged 80 and above. Marital status was also associated with the choice of first treatment institution, Widowed, single, and divorced patients (p = 0.005, OR = 0.344) had a significantly higher proportion of visiting county hospital for the first treatment institution than those who were married. Geographic distribution was similarly associated with the choice of first treatment institution, with patients in the central region (p = 0.001, OR = 0.204) had a significantly higher proportion of choosing county hospital visit than patients in the western region. No statistically significant association was observed between urban or rural residential areas and the choice of healthcare facilities among elderly patients.

Enabling factors

Household annual income was significantly associated with the choice of first treatment institution. Compared to patients with a family annual income exceeding ¥100,000, those with a family annual income between ¥10,000-¥50,000 (p = 0.005, OR = 0.327) had a higher proportion of choosing county medical services. No statistically significant association was observed between the type of medical insurance and patients’ choice of first treatment institution.

Need factors

Cancer staging divides the severity of cancer into different stages and exhibits a significant correlation with the choice of first treatment institution. Compared to patients in stage Ⅳ, those in stage Ⅰ (p = 0.001, OR = 0.168) had a higher proportion of choosing county hospitals for treatment. No statistically significant association was observed between the frequency of physical examination before the patients’ cancer diagnosis and their choice of first treatment institution after diagnosis.

Sensitivity analysis using a fully adjusted logistic regression model including all theoretically relevant variables derived from Andersen’s Behavioral Model showed that the main findings remained largely consistent (Supplementary Table). The associations of age, marital status, region, household income, and cancer stage with the choice of first treatment institution remained stable in both direction and magnitude after adjustment for additional covariates. These findings suggest that the results of the primary model were robust.

The flow of older cancer patients to the clinic

Information on multiple visits of cancer patients was collected to analyze the flow of patients at all levels of hospitals (Fig. 2). We found that the largest percentage of patients had been attending county and municipal hospitals with 25%, followed by those who had been attending county hospitals with 24%, 18% of patients had been attending municipal hospitals, and 16% of patients who had moved between county and provincial hospitals.

Fig. 2.

Patient flow through the medical process

Discussion

The objective of this study is to understand the first treatment institution choices and influencing factors for elderly patients undergoing their first anti-cancer treatment after being diagnosed with cancer. This study found that proximity was the most prominent reason for elderly cancer patients to choose county hospitals for their first treatment, and the level of medical technology was the most important reason for patients to choose provincial and municipal hospitals for their first treatment. Nevertheless, as a whole, elderly cancer patients are more likely to choose high-level hospitals for their first treatment. Cancer is an extremely complex type of disease characterized by a long course, intricate treatment methods, and high requirements for medical technology, and it is a typical demand-oriented type of medical treatment, which may be associated with a preference for seeking care at medical institutions with better medical resources after diagnosis [19]. In this study, 68.72% of the patients selected provincial and municipal medical institutions for the first treatment, which is consistent with the findings of a previous study on the flow of hospital choices for cancer patients’ first treatment [20]. It is also consistent with the results of a study on the institutional flow of treatment costs for patients with malignant tumors [21]. In recent years, compared with high-level hospitals, county hospitals have developed relatively slowly and the gap with high-level hospitals has gradually widened. Some scholars investigating the flow of hospitalization of cancer patients found that the proportion of hospitalizations in township and county hospitals has been decreasing annually, while the proportion of hospitalizations in provincial and municipal hospitals has a clear upward trend [22].

Predisposing factors

We found that there were differences in the utilization of health services among older cancer patients in different age groups. Compared to patients aged 80 and above, patients aged 60–69 are 3.392 times more likely to choose provincial or municipal hospitals for their first treatment, while patients aged 70–79 are 2.571 times more likely to do so.We believe that younger elderly patients may have better survival outcomes compared to older elderly patients. A Japanese study also found that cancer-specific survival rates were poorer in patients aged 75 and above compared to those under 75 years of age [23]. Additionally, younger elderly patients may exhibit superior mobility compared to older elderly patients [24]. Consequently, younger elderly cancer patients are more likely to seek care at tertiary hospitals, which could be related to their favorable prognostic factors and superior mobility. It is worth mentioning that we found variations in healthcare-seeking behavior among elderly individuals across different regions. Patients in the central region were more likely to choose county hospitals for their first treatment than those in the western region. The western region, with its advantageous geographical location, economic prosperity, and abundant medical resources. The accessibility of medical resources may be associated with healthcare-seeking behavior among elderly individuals [25–27].

Enabling factors

Furthermore, studies have demonstrated that social and family support may be associated with patients’ healthcare choices [28, 29]. Our research found that patients who are widowed, single, or divorced are more inclined to seek first treatment at county hospitals compared to married patients. A stable marital status may be associated with greater emotional and material support for cancer patients, which could in turn be related to their healthcare choices and prognosis [30, 31]. Income level may be associated with the affordability of medical services and healthcare accessibility among elderly individuals [32–34]. Our research indicated that income was significantly associated with healthcare-seeking behavior among elderly individuals. Compared to elderly patients with an annual family income exceeding ¥100,000, those with an annual family income ranging from ¥10,000 - ¥50,000 have a higher utilization rate of county medical services. In other words, patients with lower incomes are more inclined to choose county hospitals for their first treatment. Elderly patients with higher annual family incomes have a greater ability to bear the financial burden of illness, and the cost of medical services poses fewer restrictions on their healthcare choices [35, 36].

Need factors

The distribution of medical resources across different regions and the disparities between urban and rural medical resources may be associated with individuals’ choices of healthcare services [37–41]. Furthermore, we have also observed that patients with early-stage cancer preferred to choose county hospitals as their first treatment institution. Through interviews with hospital administrators, we have gained insights into the possible reasons that some of the county hospitals in the research area have earlier launched early cancer diagnosis and treatment programs, and the local hospitals have accumulated certain technology and experience in treating some commonly and high-incidence cancers. Therefore, they are largely capable of meeting the healthcare needs of patients with early-stage cancer.

This study has the following limitations. First, as a cross-sectional study, it can only reveal statistical associations between variables and cannot infer causal relationships. Second, since the study sample was recruited from inpatients in hospital oncology departments, there is a certain degree of selection bias due to hospital source, and it cannot fully represent the actual healthcare-seeking behaviors of the general population of elderly cancer patients in the community. Additionally, some key data, such as annual household income, were obtained through patient self-reports, which may be subject to recall bias. Third, the study sample is limited to Shandong Province. Healthcare-seeking behaviors among elderly cancer patients may vary across regions with different economic levels and distributions of medical resources, limiting the generalizability and national representativeness of the findings. Fourth, the study included only the choice of the first treatment institution as an outcome measure and did not comprehensively consider potential influencing factors such as cancer type and travel distance to medical facilities; the scope of variables requires further refinement.

Conclusion

In the background of an increasingly aging population, the healthcare seeking behavior of the elderly deserves attention. Our research indicates that tumor stage, age, marital status, region, and household income were associated with the choice of first treatment institutions among elderly cancer patients. Specifically, tumor stage reflects, to some extent, the urgency of the disease and the demand for medical technology, which may be related to patients’ choice of healthcare providers. Factors including age, marital status, household income, and geographic region may be associated with patients’ choice of healthcare resources at different levels through their impact on prognosis, level of care support available, and financial affordability.

This study can provide some policy recommendations. On one hand, the government should take effective measures in future work to enhance the cancer prevention and treatment capabilities of county medical institutions and improve the standardization of cancer prevention and treatment services. By strengthening vertical cooperation among hospitals of different levels, promoting the sinking of high-quality resources, and attracting high-level talent to county hospitals, the health service capacity of medical institutions in the county can be improved. This not only meets the demand of cancer patients for advanced medical standards and resources but also effectively encourages more rural cancer patients to stay within their counties for treatment, thereby lowering medical costs. On the other hand, based on a thorough assessment of medical quality and patient prognosis, more refined healthcare coverage and tiered diagnosis and treatment guidance strategies could be explored to encourage patients to make rational choices of healthcare facilities according to the severity of their illness.

It should be noted that while this study primarily identified factors influencing the choice of healthcare facilities, it did not further evaluate the impact of different choices on treatment outcomes or survival. Therefore, specific pathways for healthcare resource allocation and adjustments to health insurance policies still require validation through further causal inference studies and cost-effectiveness analyses.

Authors’ contributions

NZ designed and planned the study. KM did data analyses and wrote the first draft of the manuscript. YS and YF provided critical comments on the scientific interpretation of the results and on revised versions of the manuscript. YL coordinated the data acquisition. JW designed and supervised the overall conduct of the study. All authors provided critical comments on the manuscript.

Funding

This research was funded by the 2023 Shandong Provincial Health Policy Research Project, grant number WZY202375.

Data availability

The data supporting this study’s findings are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study adheres to the Declaration of Helsinki. Ethical approval was obtained from theEthics Committee of the Cancer Hospital Affiliated to Shandong First Medical University(protocol number SDTHEC202308026). The participants or the nextof-kins provided their written informed consent to participate in this study.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.