The choice of medical facility and associated factors among Chinese advanced colorectal cancer patients: a cross-sectional multi-center study

Introduction

Colorectal cancer (CRC), one of the most common malignancies worldwide, was responsible for 1,931,590 new cases and 935,173 deaths in 2020 (1). In China, according to the latest Chinese Cancer Registration Report in 2019, new cases of CRC accounted for nearly 10% of all cancers, with CRC deaths accounting for 8% of all cancer deaths (2), posing a serious threat to public health. For decades, China has implemented a series of programs to improve CRC screening, early diagnosis, and treatment, leading to a significant increase in CRC survival (3). However, the 5-year survival rate is still only about 20% for advanced CRC patients (4), and the majority of CRC patients have advanced or metastatic cancer at their first diagnosis (5,6), suggesting that there are other barriers to care for CRC patients, especially for those at the advanced stage. Of note, it has been reported that disparities in access to diagnosis and treatment services may contribute to marked differences in CRC survival between countries or regions (7,8). The investigation of healthcare-seeking behaviors among CRC patients is critical for understanding access to health services, identifying service gaps, and improving quality of oncology services to reduce health risks caused by inadequate services (9). It can also help policymakers and hospital administrators to understand patterns and driving factors underlying patients’ healthcare-seeking behaviors, and thus to develop targeted strategies to alleviate the burden of CRC (10).

Healthcare-seeking behavior, referring to the actions taken by individuals perceiving their illness to obtain an adequate remedy (11), involves several decision-making processes such as whether and from whom to seek healthcare, as well as what kind of healthcare to seek (10). Previous studies on healthcare-seeking behaviors among CRC patients focused on delays in presentation, diagnosis, and treatment, and showed that factors such as gender, education, economic status, and awareness of the disease, among others, could affect delays in healthcare services (12-16). Beyond that, examining the choice of medical facility among CRC patients and its associated factors is also important for understanding patients’ needs for health services. This will help to identify potential gaps in service provision (17), and take measures to improve quality of oncology services for optimizing health outcomes (18). However, research on the choice of medical facility among CRC patients and its associated factors is very limited.

Here, we aimed to investigate the choice of medical facility and its associated factors among advanced CRC patients. Many factors may affect the utilization of health services. A behavioral model, initially developed by Andersen in the 1970s, has been commonly used to identify factors involved in healthcare utilization (19). This model suggests that the determinants of healthcare utilization can be classified into predisposing characteristics, enabling resources, and need factors. Predisposing factors, namely individual characteristics, include demographic, socio-structural, and attitudinal-belief variables (20). Enabling factors refer to the variation in availability of resources which may facilitate or impede individuals’ healthcare use (21). Need factors encompass the perceived and assessed needs for health services (22). Based on the conceptual framework, this study collected information on the medical experience of 4,589 individuals with advanced CRC to investigate their choice of medical facility and associated factors, thus providing clues for the improvement of quality of oncology services in China. We present the following article in accordance with the STROBE reporting checklist (available at https://atm.amegroups.com/article/view/10.21037/atm-22-1020/rc).

Methods

Study design and population

Data were obtained from the largest CRC survey conducted in China from March 2020 to March 2021. The cross-sectional multi-center study included 19 hospitals in 7 geographic regions of mainland China. Hospital selection was conducted as follows. According to the definition of traditional administrative districts, China is divided into seven geographic regions: Northwest, Northeast, North, Central, Southern, Southwest and Eastern. Each region shows different CRC burden levels (23). First, two cities in each geographic region were selected through simple random sampling. Second, one tertiary cancer hospital and/or one general hospital, which can provide health services including diagnosis, surgery, chemotherapy, radiotherapy, and regular follow-up care for CRC patients, and where patients come from multiple parts of the region, were selected in each city. As a result, nineteen tertiary hospitals inclusive of ten cancer hospitals and nine general hospitals, were involved in this study.

Patients from the selected hospitals were enrolled according to the following criteria: (I) diagnosis of stage III or IV CRC; (II) aged ≥18 years old. Patients who could not complete the questionnaire due to physical, cognitive, or verbal disorders were excluded. Sample size was calculated based on the number of advanced CRC patients in China, which was estimated to be 400,000 cases (24,25). To ensure geographical representativeness of the national survey, about 1% of the cases were taken into consideration, and taking into account the non-response rate of 10%, over 4,445 patients needed to be enrolled. In fact, a total of 4,589 individuals with advanced CRC were recruited in the current study. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This research was approved by the review board of Henan Cancer Hospital (No. 2019273), and the study was approved by all institutional review boards of the participating hospitals. All participants provided written informed consent.

Study procedures

In the first phase, workshops with researchers from all the centers were conducted for preparation, including questionnaire design, obtainment of prior agreement of the study launch, development of the implementation manual and staff training. Next, a pilot survey including 50 CRC patients was performed in Henan Cancer Hospital and The First Affiliated Hospital of Baotou Medical College to verify operating procedures and questionnaires. The formal survey was launched after ethical approval. It would take approximately 20 minutes to complete the questionnaire, and participants could get 30 yuan for their contribution after filling out the questionnaire. In addition, the survey was carried out by trained interviewers, and principles of good research practice was strictly adhered to during data collection.

Data collection

Socio-demographics, health-related quality of life (HRQOL) prior to the first treatment, and medical experience-related variables were collected as part of the protocol of the survey via a questionnaire during face-to-face interviews. Socio-demographics covered self-reported information on age, gender, occupation, education, marital status, annual household income of patients, and medical insurance. In addition, a semi-structured questionnaire was designed to collect data on patients’ medical experience in the diagnosis and treatment phases: the number of visited hospitals from initial medical care to the survey date, reasons for seeking initial medical care, the choice of medical facility in all the diagnosis and treatment phases, the choice of hospital department for initial medical care and the first definite diagnosis, and the choice and reasons for changing hospitals during the period from definite diagnosis to the first treatment, from the first treatment to the second treatment, and after recurrence and metastasis. Besides, clinical information such as metastasis status at the first definite diagnosis and the type of CRC were obtained though medical records.

Selection and definition of variables in logistic regression analyses

In the study, we defined the outcome variable as the level of medical facility visited by advanced CRC patients for their first treatment. The outcome variable was coded into 4 categories: hospitals at the national level, hospitals in provincial capital, hospitals at the prefecture level, and hospitals at the county level. In the analyses, we always used “hospitals at the county level” as a reference level.

We selected explanatory variables guided by the Andersen model (19), which has been commonly used to identify factors influencing the utilization of health services. In the study, variables considered as predisposing characteristics were age (≤40, 41–60, >60), gender (male, female), education (primary school or below, middle school, high school, college and above), occupation of patients or their families (non-healthcare related, healthcare related), and marital status (not married/divorced/widowed, married). Variables related to enabling resources included medical insurance (none, public, private) and annual household income of patients (<50,000 CNY, 50,000–100,000 CNY, >100,000 CNY). For need factors, metastasis status at the first definite diagnosis (no, yes), the type of cancer (colon, rectum, both), and HRQOL prior to the first treatment were selected. The traditional Chinese Functional Assessment of Cancer Therapy-Colorectal (FACT-C, version 4) was used to measure HRQOL. The scale includes 5 function subscales (physical well-being, social/family well-being, emotional well-being, functional well-being, and CRC subscale). Each item was valued on a 5-point Likert-type scale (0–4). The total scores were calculated (ranged from 0 to 136), and then were classified as “poor” (total score ≤100) and “good” (total score >100) (26).

Statistical analysis

Data were presented as mean ± standard deviation (SD) for continuous variables and percentages (%) for categorical variables. In the bivariate analyses, explanatory variables were selected based on the Andersen model, and multinomial logistic regression models were performed to analyze associations between the level of medical facility attended for the first treatment and potential factors. Odds ratios (ORs) as well as 95% confidence intervals (CIs) could be consequently calculated. Variables with an association of P value <0.1 were subsequently included in multivariate analyses among patients with complete information. All statistical analyses were performed using R v3.6.1 with a two-tailed P value of <0.05 being considered statistically significant.

Results

Participants’ characteristics

A total of 4,589 participants comprised of 2,730 males and 1,859 females were included in the survey. Demographic and health characteristics of the participants are summarized in Table 1. Only 270 (5.9%) individuals were aged 40 years or below, while 1,979 (43.1%) were in the age group of 41–60 years, and 2,340 (51.0%) were aged above 60 years. Most individuals (94.1%) were married, and only 569 (12.4%) individuals or their families had healthcare-related occupations. There were similar results in terms of educational attainment, with 32.2% completing middle school, followed by primary school or below (29.0%), high school (22.8%), and college and above (16.0%). More than half (57.4%) of the individuals had an annual household income of less than 50,000 CNY, while 28.3% were in the income group of 50,000-100,000 CNY, and 14.3% had a high household income (>100,000 CNY). For insurance type, 51 (1.1%) individuals had no medical insurance, while 220 (4.8%) had private insurance, and others (94.1%) only had public insurance. Of note, 1,709 (37.5%) individuals had developed metastatic cancer at the first definite diagnosis. In addition, 50.9% of individuals reported “poor” HRQOL prior to the first treatment, with 49.1% reporting “good”.

The choice of medical facility among advanced CRC patients

The median number of hospitals visited by study individuals was 2.00 (range, 1.00–7.00; mean 1.94). With regard to the reasons for initial care, 4,015 (88.0%) individuals found suspected symptoms (e.g., hematochezia, severe diarrhea, and abdominal pain) themselves, while 269 (5.9%) and 279 (6.1%) found suspected symptoms during health examination and treatment for other diseases, respectively.

The choice of medical facility of advanced CRC patients are presented in Table 2. Of the 4 levels of medical facilities, individuals preferred to visit hospitals at the prefecture level in phases including initial medical care (44.9%), the first definite diagnosis (46.3%), the first treatment (39.5%), and regular follow-up (38.9%). Intriguingly, individuals were more likely to choose hospitals at the national level for the second treatment (38.0%) and after recurrence and metastasis (45.9%). Furthermore, the most common hospital department sought by CRC patients was the department of gastrointestinal surgery for both initial medical care (55.8%) and the first definite diagnosis (60.3%).

The percentages of individuals changing hospitals were 31.8%, 28.9%, and 22.1% during the period from definite diagnosis to the first treatment, from the first treatment to the second treatment, and after recurrence and metastasis, respectively. Among those who visited multiple hospitals, the majority of individuals (57.3–62.3%) changed hospitals from their own willingness during the medical care seeking process, while some individuals followed their offspring’s advice (23.9–29.0%) or the doctor’s advice (10.2–12.7%), and only a very small proportion (0.7–2.4%) changed hospitals due to a lack of therapeutic drugs.

Factors associated with the choice of medical facility for the first treatment

We next performed multinomial logistic regression models to identify factors associated with the choice of medical facility for the first treatment. The results of the bivariate analysis are reported in Table 3. Several factors including age, gender, educational attainment, occupation, annual household income of patients, medical insurance, metastasis at the first definite diagnosis, and HRQOL prior to the first treatment were significantly associated with the level of facility (P<0.1).

Subsequently, we conducted a multivariate analysis by incorporating these predictive factors into multinomial logistic regression models. As shown in Table 4, the results demonstrated that gender was significantly associated with the level of medical facility, ranging from 1.31 to 1.41 times greater odds for females compared with males. Individuals who completed high school or college and above, as compared with primary education or below, were more likely to visit hospitals at a higher level [ORs ranged from 1.74 (95% CI: 1.20–2.53) to 7.26 (95% CI: 4.18–12.60)], while individuals who completed middle school only preferred to visit hospitals at the national level (OR =1.81, 95% CI: 1.33–2.47). Individuals with metastatic cancer at the first definite diagnosis preferred to visit hospitals in provincial capitals (OR =1.68, 95% CI: 1.27–2.22). Individuals with “good” HRQOL prior to the first treatment were less likely to visit hospitals at the prefecture level compared with hospitals at the county level (OR =0.63, 95% CI: 0.49–0.81).

Discussion

In this study, we investigated the choice of medical facility in multiple medical processes among Chinese advanced CRC patients using data from a nationwide multi-center survey. To further explore potential factors affecting the choice of medical facility for the first treatment, multinomial logistic regression analyses were performed based on the Andersen model. Our findings offer important evidence on the drivers of utilization of healthcare services by CRC patients, which may contribute to the improvement of referral patterns and the construction of a well-structured healthcare system.

The choice of medical facility among individuals with advanced CRC varied in different medical processes in this study. For initial medical care, the first definite diagnosis, the first treatment, and regular follow-up, the most common facilities sought by CRC patients were hospitals at the prefecture level. However, the first priority was changed to hospitals at the national level for the second treatment and after recurrence and metastasis. This may be because individuals with a relatively severe medical condition have an urgent need to seek a high-level healthcare facility, and hospitals at the national level are generally perceived to offer the best quality medical services. In regards to the choice of changing hospitals, most individuals did not change hospitals during the medical processes. Among those who changed hospitals, the majority took the action from their own willingness, followed by the offspring’s advice or the doctor’s advice. Only a very small proportion changed hospitals owing to a lack of therapeutic drugs. This is a reflection of the fairly good Essential Drug System, which facilitates the availability and quality of essential medicines.

Multivariate logistic regression analysis revealed that the level of medical facility sought by the CRC patients was significantly associated with gender, education, metastasis at the first definite diagnosis, and HRQOL prior to the first treatment. As a result, females were more likely to seek higher-level facilities versus hospitals at the county level. This was in line with previous studies where males exhibited poor healthcare-seeking behavior and healthcare utilization (27-30). The literature indicated that the healthcare-seeking behavior of males might be affected by the society and culture surrounding them, and in some cases, the concepts of masculinity could prevent males from seeking medical help (27). Another prominent factor strongly associated with the choice of medical facility was educational attainment. The results showed that well-educated individuals were more likely to visit higher-level health facilities compared with hospitals at the county level. This is probably because less educated individuals prefer to follow their doctor’s advice, and thus are less likely to bypass the primary care facility (31,32). Besides, we also observed that individuals suffering from metastatic CRC at diagnosis were more likely to seek hospitals in provincial capitals for the first treatment compared with hospitals at the county level. This may be due to the fact that hospitals in provincial capitals are generally perceived to have good health resources and the ability to offer quality services. Simultaneously, the complicated referral mechanism may reduce the willingness of CRC patients to visit hospitals at the national level. Additionally, individuals with “good” HRQOL prior to the first treatment were more likely to seek hospitals at the county level rather than hospitals at the prefecture level. The result may be attributed to the fact that compared with individuals with “poor” status, individuals with “good” HRQOL have a relatively less urgent need to seek higher level facilities. That is, they probably believe that hospitals at the county level meet their requirements for treatment.

Although data in this study was based on a nationwide multi-center survey and multivariate logistic regression analysis was performed to correct potential confounders, there are also several limitations. First, information on healthcare-seeking behaviors relied entirely on self-reporting, which might result in recall bias and the occurrence of under-reporting or over-reporting, and thus likely affected the reliability or validity of the results. Second, some potential factors which likely influence the choice of medical facility, such as distance from home to the medical facility and time to reach the facility, were not available. We will conduct further study to collect such information and adjust for more confounders to build strong support for the current results.

Overall, this study described the choice of medical facility among individuals with advanced CRC in different medical phases, and identified several factors associated with the choice of medical facility for the first treatment, such as gender, education, metastasis at the first definite diagnosis, and HRQOL prior to the first treatment. Our findings provide evidence to fill the policy-implementation gap and may contribute to the improvement of the efficiency of the healthcare system.

Acknowledgments

Funding: This research was funded by Beijing Love Book Cancer Foundation and Merck Serono Co. Ltd.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://atm.amegroups.com/article/view/10.21037/atm-22-1020/rc

Data Sharing Statement: Available at https://atm.amegroups.com/article/view/10.21037/atm-22-1020/dss

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://atm.amegroups.com/article/view/10.21037/atm-22-1020/coif). All authors report that this research was funded by Merck Serono Co. Ltd. The authors have no other conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). This research was approved by the review board of Henan Cancer Hospital (No. 2019273), and the study was approved by all institutional review boards of the participating hospitals. All participants provided written informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA Cancer J Clin 2021;71:209-49. [Crossref] [PubMed]
2. National Central Cancer Registry (China). China Cancer Registry Annual Report 2019. Beijing: People's Medical Publishing House; 2021.
3. Zeng H, Chen W, Zheng R, et al. Changing cancer survival in China during 2003-15: a pooled analysis of 17 population-based cancer registries. Lancet Glob Health 2018;6:e555-67. [Crossref] [PubMed]
4. Shen L, Li Q, Wang W, et al. Treatment patterns and direct medical costs of metastatic colorectal cancer patients: a retrospective study of electronic medical records from urban China. J Med Econ 2020;23:456-63. [Crossref] [PubMed]
5. Hammond WA, Swaika A, Mody K. Pharmacologic resistance in colorectal cancer: a review. Ther Adv Med Oncol 2016;8:57-84. [Crossref] [PubMed]
6. Siegel RL, Miller KD, Fedewa SA, et al. Colorectal cancer statistics, 2017. CA Cancer J Clin 2017;67:177-93. [Crossref] [PubMed]
7. Boyle P, Langman JS. ABC of colorectal cancer: Epidemiology. Bmj 2000;321:805-8. [Crossref] [PubMed]
8. Favoriti P, Carbone G, Greco M, et al. Worldwide burden of colorectal cancer: a review. Updates Surg 2016;68:7-11. [Crossref] [PubMed]
9. Treanor C, Donnelly M. An international review of the patterns and determinants of health service utilisation by adult cancer survivors. BMC Health Serv Res 2012;12:316. [Crossref] [PubMed]
10. Zeng Y, Wan Y, Yuan Z, et al. Healthcare-Seeking Behavior among Chinese Older Adults: Patterns and Predictive Factors. Int J Environ Res Public Health 2021;18:2969. [Crossref] [PubMed]
11. Kasl SV, Cobb S. Health behavior, illness behavior, and sick-role behavior. II. Sick-role behavior. Arch Environ Health 1966;12:531-41. [Crossref] [PubMed]
12. Chow Z, Osterhaus P, Huang B, et al. Factors Contributing to Delay in Specialist Care After Colorectal Cancer Diagnosis in Kentucky. J Surg Res 2021;259:420-30. [Crossref] [PubMed]
13. Schliemann D, Ismail R, Donnelly M, et al. Anticipated delay in help-seeking for cancer symptoms: Findings from a nationwide study of Malaysian adults. Cancer Epidemiol 2021;71:101882. [Crossref] [PubMed]
14. Smith KT, Monti D, Mir N, et al. Access Is Necessary but Not Sufficient: Factors Influencing Delay and Avoidance of Health Care Services. MDM Policy Pract 2018;3:2381468318760298. [Crossref] [PubMed]
15. Siminoff L, Thomson M, Dumenci L. Factors associated with delayed patient appraisal of colorectal cancer symptoms. Psychooncology 2014;23:981-8. [Crossref] [PubMed]
16. Zarcos-Pedrinaci I, Fernández-López A, Téllez T, et al. Factors that influence treatment delay in patients with colorectal cancer. Oncotarget 2017;8:36728-42. [Crossref] [PubMed]
17. Rasul FB, Kalmus O, Sarker M, et al. Determinants of health seeking behavior for chronic non-communicable diseases and related out-of-pocket expenditure: results from a cross-sectional survey in northern Bangladesh. J Health Popul Nutr 2019;38:48. [Crossref] [PubMed]
18. Kanté AM, Gutierrez HR, Larsen AM, et al. Childhood Illness Prevalence and Health Seeking Behavior Patterns in Rural Tanzania. BMC Public Health 2015;15:951. [Crossref] [PubMed]
19. Andersen RM. Revisiting the behavioral model and access to medical care: does it matter? J Health Soc Behav 1995;36:1-10. [Crossref] [PubMed]
20. Baek S, Choi EH, Lee J. Unmet Healthcare Needs of Children in Vulnerable Families in South Korea: Finding from the Community Child Center Child Panel Survey. Int J Environ Res Public Health 2020;17:8241. [Crossref] [PubMed]
21. Babitsch B, Gohl D, von Lengerke T. Re-revisiting Andersen's Behavioral Model of Health Services Use: a systematic review of studies from 1998-2011. Psychosoc Med 2012;9:Doc11. [PubMed]
22. Andersen RM. National health surveys and the behavioral model of health services use. Med Care 2008;46:647-53. [Crossref] [PubMed]
23. Yang Y, Han Z, Li X, et al. Epidemiology and risk factors of colorectal cancer in China. Chin J Cancer Res 2020;32:729-41. [Crossref] [PubMed]
24. Global Cancer Observatory. Available online: https://gco.iarc.fr/today/home
25. Yao H, Li X, Cui L, et al. Annual report of Chinese Colorectal Cancer Surgery Database in 2019: A nationwide registry study. Chinese Journal of Practical Surgery 2020;40:106-10,16.
26. Rausa E, Kelly ME, Bonavina L, et al. A systematic review examining quality of life following pelvic exenteration for locally advanced and recurrent rectal cancer. Colorectal Dis 2017;19:430-6. [Crossref] [PubMed]
27. Tong SF, Ho C, Tan HM. Managing the aging man in Asia: a review. Int J Urol 2011;18:32-42. [Crossref] [PubMed]
28. Courtenay WH. Constructions of masculinity and their influence on men's well-being: a theory of gender and health. Soc Sci Med 2000;50:1385-401. [Crossref] [PubMed]
29. Addis ME, Mahalik JR. Men, masculinity, and the contexts of help seeking. Am Psychol 2003;58:5-14. [Crossref] [PubMed]
30. Lee C, Owens RG. Issues for a Psychology of Men's Health. J Health Psychol 2002;7:209-17. [Crossref] [PubMed]
31. Victoor A, Delnoij DM, Friele RD, et al. Determinants of patient choice of healthcare providers: a scoping review. BMC Health Serv Res 2012;12:272. [Crossref] [PubMed]
32. Gabrani J, Schindler C, Wyss K. Health Seeking Behavior Among Adults and Elderly With Chronic Health Condition(s) in Albania. Front Public Health 2021;9:616014. [Crossref] [PubMed]

(English Language Editor: C. Betlazar-Maseh)