Introduction
Lung cancer has been the second most commonly diagnosed cancer and the major cause of death from cancer worldwide, imposing a heavy disease burden on global health [
1]. In 2018, the global number of new lung cancer cases was approximately 2.09 million, ranking first among all cancer types. Other than causing 19.4% of total cancer deaths, lung cancer is also considered one of the main causes of cancer–caused disability–adjusted life years (DALYs) [
2,
3].
Lung cancer imposed a heavy burden on patients, their families, and the health system in China, with approximately 787,000 new cases in 2015. The age-standardized mortality rate in China reached 28.16 per 100,000 people, and approximately 30% of cancer deaths were due to lung cancer in 2015, both higher than most countries [
4]. For each lung cancer patient in 2015, the average expenses in the first year following diagnosis accounted for 171% of the household annual income, and the all-direct expenses within 5 years after diagnosis was $42,540 [
5]. For the whole country in the same year, about 0.6% of total health expenditure (RMB 24.31 billion) was on lung cancer treatment [
6]. The incidence of catastrophic expenditure on lung cancer was estimated at 42.78%, higher than that for gastric, liver, esophageal, and breast cancer in China [
7]. Under such circumstances, health insurance drew increasing public attention in that it could significantly get this burden down for families with lung cancer survivors.
China’s health insurance schemes for urban workers and urban residents bifurcate into the Urban Employee Basic Medical Insurance (UEBMI) and the Urban Resident Basic Medical Insurance (URBMI). UEBMI is compulsory and designed exclusively for urban employees. Contrarily, URBMI is a voluntary insurance program covering urban residents without formal employment, including young children, students, seniors, disabled, and other unemployed urban residents. The two health insurance schemes vary considerably in funding source, service coverage, and benefits packages [
8]. Based on the annual salary of employees, employers and employees contribute 6 and 2% to UEBMI, respectively. URBMI is co-financed by both individuals and the government, the government has higher subsidies compared to individual premium contributions. In 2016, the per capital fund for UEBMI and URBMI was RMB 3478 and RMB 626, respectively [
9]. UEBMI covers both outpatient and inpatient services whereas URBMI covers only inpatient services in most situations. Compared with URBMI, UEBMI provides a higher reimbursement rate, higher reimbursement ceiling, and more comprehensive service coverage, which means UEBMI has a better financial protection capacity for those enrolled. Importantly, a prior study stated that both UEBMI and URBMI schemes are pooled at the municipal level in China (approximately 333 UEBMI and 333 URBMI health insurance schemes under China’s fragmented health insurance system) [
8], leading to uneven benefit packages in different insurance types and cities.
The financial protection ability of health insurance schemes is either an incentive or a disincentive for patients to utilize health services [
10,
11]. Existing studies have compared patients’ medical costs for stroke [
12], schizophrenia [
13], and tuberculosis [
14], supported by different insurance schemes. However, research investigating disparities in medical costs for lung cancer patients supported by different insurance schemes is scarce. Further, considering the differences in health insurance schemes among cities could be significant. Therefore, we used 7-year claims data for lung cancer from UEBMI and URBMI schemes in China to elucidate how these two health insurance schemes and their municipal differences shaped healthcare access and medical costs for lung cancer inpatients.
Discussion
Based on claims data from CHIRA, this study revealed differences in medical costs for lung cancer patients with two health insurance schemes, UEBMI and URBMI, and in different cities. Overall, compared with the URBMI group, the UEBMI group had higher medical costs (including drugs, medical service, and medical consumable) and higher mean outside-insurance OOP expenses, but lower OOP expenses, which means that the UEBMI group has utilized more comparatively (or more expensive) health services and bore less individual financial burden. The differences varied by cities—differences in insurances among cities were greater than the differences in insurances within cities. In addition, our study specifically showed that the differences in OOP expenses among cities were greater than the differences between UEBMI and URBMI. Regarding the overall outside–insurance OOP expenses, the mean and median values had opposite performance when comparing the UEBMI and URBMI. Using the median value, overall, UEBMI had lower outside–insurance OOP expenses than URBMI, but URBMI had higher ones when using the mean value. However, these two types of results were not contradictory. This is because compared to those in the URBMI groups, the upper-most quartiles of median outside-insurance OOP expenses in the UEBMI groups were higher; thus, despite more comprehensive service coverage of UEBMI a few patients with the UEBMI insurance scheme might have utilized health services that were out of range. Further, health services not supported by UEBMI were usually more expensive and of higher quality compared to those not covered in UEBMI.
The results regarding differences between UEBMI and URBMI were consistent with prior studies [
17]. Using the China health and retirement longitudinal study (CHARLS) data, Wang et al. [
18] reported that UEBMI had a greater effect in improving healthcare utilization and causing higher medical costs compared with URBMI. Based on claims data from Guangzhou province in China, Zhang et al. [
19] found that the UEBMI dementia patients had higher hospitalization costs compared with the URBMI counterparts. Similarly, Chen et al. [
20] revealed that the UEBMI diabetic patients incurred higher expenditure compared to the URBMI patients. The present finding that UEBMI lung cancer patients have lower OOP, was also similar to that of Yang et al. [
12], who reported that the UEBMI stroke patients had fewer direct economic burdens than the URBMI counterparts. The differences in expenditure between the UEBMI and the URBMI patients were due to several possible reasons. From the patients’ socioeconomic status perspective, those in the UEBMI group were all urban workers or retired workers, compared with the disabled residents and unemployed patients who were covered by URBMI, usually having higher income (or pension), and better education. Income was an important contributor to healthcare utilization inequity, and people with high income had stronger incentive to utilize expensive health services and assume corresponding high medical costs [
21,
22]. This was also the reason that some UEBMI patients had higher outside-insurance OOP expenses (paying medical services, drugs, and medical consumables which were not supported and compensated by the insurance schemes) compared with the URBMI patients in this study. As UEBMI’s service packages were more comprehensive compared to URBMI’s service packages, theabove-mentioned medical services, drugs, and medical consumables that were not supported by health insurance scheme were generally non-basic and expensive. The URBMI patients with low income could be more conservative when utilizing health services and drugs. Thus, differences in income between the UEBMI and the URBMI groups may lead to differences in medical costs. Further, we speculate that people with better education had more knowledge of health and were more willing to pay for it [
23,
24], which attributed the UEBMI group’s higher medical costs to their better education.
From the perspective of the financial protection ability of insurance schemes, UEBMI provided a higher reimbursement rate and higher reimbursement ceiling compared to URBMI. Prior studies have proved that patients covered by health insurances with better financial protection tend to seek better quality health services in higher-level of hospitals [
25]. The present results indicated that more patients in the UEBMI groups accepted treatment in tertiary hospitals, subsequently causing higher medical costs. Health insurance schemes with better financial protection had a greater effect on motivating patients to utilize more health services [
19]. By disregarding catastrophic health expenditure, the UEBMI beneficiaries were more willing to use expensive drugs and medical consumables compared to the URBMI patients. In addition, patients with different insurances may choose different therapeutic schedules, naturally resulting in differences in the medical costs, drug costs, medical service costs, and medical consumable costs [
26]. Conversely, doctors may also provide more reasonable treatments to reduce the economic burden for patients with URBMI [
27]. The present findings regarding the differences in the composition of medical costs supported these two speculations indirectly. We believe that the higher reimbursement rate and higher reimbursed ceiling also caused the UEBMI patients to incur fewer OOP expenses than compared with URBMI patients. In brief, the UEBMI funding pool has incurred the highest medical costs for its beneficiaries, while URBMI funding pool was not as generous.
From the population characteristics perspective, the UEBMI group had more male patients than the URBMI group. Compared with female patients, male patients were more likely to smoke, which is the most threatening risk factor for lung cancer [
28]. Therefore, male patients had a larger population attributable fraction (PAF) of lung cancer deaths caused by smoking, and higher medical costs compared to female patients [
29,
30]. The higher medical costs for male lung cancer patients may have contributed to higher medical costs for the UEBMI group. Second, compared to the URBMI group, more patients in the UEBMI group had comorbidity, which has been proved to be significantly associated with high medical costs [
31].
The presentstudy also foundthat the differences in medical costs among cities were greater than the differences in medical costs by insurances within cities; the same case applies to the OOP expenses. First, the different cities had separate UEBMI and URBMI funding pools, leading to the different service coverage and benefits packages. Hence the differences in financial protection between these two insurance schemes primarily played a role in the differences in medical costs and OOP expenses among cities [
8]. Second, with different economic development levels, the UEBMI workers in different cities also had different levels of salaries, which caused the variance in medical costs and OOP expenses. Third, the prevalence of comorbidities such as hypertension and diabetes mellitus differed between cities. The comorbidity was associated with an increased risk of disease severity and medical costs [
31‐
33]. Fourth, hospitals’health resources and medical technologies varied by cities in China, including the three cities above [
34]. We believe that in some cities poor medical technology could have prolonged LOS for patients Table
4 showes that patients in city A had shorter median LOS than patients in city C, which proved this speculation directly. The prolonged LOS was significantly associated with medical costs [
35].
In 2016, the Chinese government officially integrated URBMI and the new rural cooperative medical insurance (NCMS, initially designed for rural patients) to establish a unified health insurance scheme, Urban-Rural Residents Basic Medical Insurance (URRBMI), covering rural residents and those earlier covered by URBMI. Although URRBMI has significantly promoted equity in access to health care utilization especially for rural residents [
36,
37], it did not significantly improve benefit packages for the original URBMI residents. Gaps between URBMI and UEBMI remain. Differences among lung cancer patients regarding medical costs and OOP expenses called for further integration of the fragmented insurance schemes in China. Notably, the current insurance integration in China was implemented within each municipal city, which improved the NCMS funding pools from county level to upper municipal level. A broader funding pool coudld resist economic risk more strongly [
8]. The integration was conducive to changing the status quo of fragmented management involving health insurance schemes in China; however, it failed to counteract the role of income, or the presence of the UEBMI, in increasing inequality on healthcare utilization [
38]. Our results indicate that the level of the UEBMI and URBMI funding pools could be further merged and improved to province level (even national level), providing residents with equal benefit packages and financial protection to reduce the gap between UEBMI and URBMI, and between different cities for lung cancer patients.
This study had several limitations. First, since URBMI and NCMS have been merged, a comparison between UEBMI and URRBMI could be a better choice. While the new insurance scheme did not drastically improve the benefit packages for the URBMI patients, the present results still reflect the differences between UEBMI and URBMI. Second, the claims data lacked clinical outcomes for lung cancer patients, thus, it was unclear whether UEBMI patients had a higher survival rate after paying higher medical costs. Third, this study did not include the indirect medical costs between the UEBMI and URBMI groups. Finally, owing to lacking detailed information regarding the cancer stage, histology type of lung cancer, and detailed lung cancer treatment, it was unclear how the cancer characteristics shaped the medical costs.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.