Esophageal cancer is the eighth most common cancer worldwide with estimates of 455,784 new cases and 400,156 deaths in 2012 [1]. The incidence varies in different physiographic regions, nations, and races. China had the highest number of new esophageal cancer cases (more than 223,000) worldwide in 2012 [2]. The number of new esophageal cancer cases in China will increase to 433,783 by the year 2035 [1]. In the meantime, in China, early detection of esophageal cancer has increased, while mortality has decreased [3, 4]. As the survival for esophageal cancer patients is prolonging and the average cost of esophageal cancer treatments is increasing, the economic burden of patients has continued to grow. Therefore, the need for prevention strategies is urgent, and the implementation of effective measures should be based on evidence from health economic analyses.

The Cancer Screening Program in Urban China (CanSPUC), which was supported by the National Health and Family Planning Commission (NHFPC) in China, has been a crucial cancer prevention initiative that began in August 2012 [5]. The primary objective of the CanSPUC was to explore an appropriate implementation approach for screening of population at high risk and early diagnosis of major cancers (including lung, female breast, colorectal, esophageal, gastric, and liver cancers) in urban populations in China, promoting the use of mature screening and early diagnosis technology for common cancers to reduce mortalities. Taking funding and local conditions into account, the program covered 13 provinces/municipalities in China [including eastern (Beijing, Hebei, Liaoning, Jiangsu, Zhejiang, Shandong, and Guangdong), central (Heilongjiang, Henan, and Hunan), and western regions (Chongqing, Gansu, and Xinjiang)] [6] by 2014 in the first stage and targets urban residents who were between 40 and 69 years of age. As a part of the CanSPUC, the main content of the health economic evaluation includes expenditure and economic burden assessment, quality of life and health utility surveys, multi-angle project sustainability assessment, clinical diagnosis and treatment, and cost-effective model-based analysis. The goal is to provide the scientific evidence for expanding cancer screening and health decision-making in early detection and treatment programs for the Chinese population.

Few studies have examined the medical expenditure of Chinese patients with esophageal cancer and their associated trends [7]. In this study, we used data collected from a retrospective survey conducted by the CanSPUC to estimate the medical expenditures for esophageal cancer during the past 10 years, from 2002 to 2011, and to analyze the time trends over the long term.

In the present study, we found that the overall average expenditure per patient for esophageal cancer increased by 6.27% per year from 2002 to 2011, with an estimate of 38,666 CNY (95% CI 38,178–39,155 CNY) during 2002–2011. Drugs accounted for the largest proportion of the overall expenditure per patient.

The first year after cancer diagnosis is a period with particularly high expenses and intensive treatment [8, 9]. We found that more than 90% of the expenses occurred within 12 months, which was higher than China’s GDP per capita (36,018 CNY in 2011) [10]. As a comparison, the newly diagnosed medical expenditure was less than 40% of the GDP per capita (2005) in the United Kingdom [11], less than 60% (2008–2010) in the United States [12], and less than 65% (1997–2007) in Canada [13]. According to data from China on the average family population (2.9) and the per capita disposable income (21,810 CNY) of urban households in 2011 [10], we found that the average medical expenditure per patient for esophageal cancer was 72.9% of the average family annual income for urban households, indicating that the economic burden imposed by esophageal cancer was relatively high for individual patients’ families in China, especially for low-income families.

Our finding that the medical expenditures in the eastern and central regions were higher than those in the western region mirrors the heterogeneity in economic development among the three regions in China. However, the overall average expenditure per esophageal cancer patient in the Xinjiang province is an exception, which was estimated at 52,167 CNY. The economic burden of esophageal cancer in Xinjiang province is unusually high [6]. Further analysis indicated that esophageal cancer patients in Xinjiang province had more clinical visits than those in other provinces/municipalities. The medical expenditures per esophageal cancer patient significantly differed between the provinces/municipalities; there was a threefold gap between the highest (Guangdong in the eastern region) and lowest (Gansu in the western region) expenditure regions. This result was in agreement with other expenditure studies on esophageal cancer in China [14, 15].

Our finding of an expenditure increase of 1.84 times observed over 10 years is consistent with the results of some other studies. For example, Hu et al. [16] analyzed hospitalization expenses of 1819 patients with esophageal cancer in a cancer hospital in the Shanxi province and found that the adjusted average expenditure per patient increased from 8022 CNY in 2002 to 33,723 CNY in 2011. The time trend of expenditures was almost the same as the time trend of the length of hospitalization stay, suggesting that the length of stay was an important factor affecting the overall average expenditure per patient. Therefore, controlling the length of hospitalization stay has been considered an effective measure for reducing medical expenditures. For this reason, some hospitals try to shorten the hospitalization stays of esophageal cancer patients through implementing clinical pathways, promoting daytime chemotherapy, and other approaches. The two peaks of medical expenditures observed in 2004 and 2008 in the present study might be attributed to the increase in clinical visits and hospitalization stay.

In the present study, we compared the time trend of medical expenditures for esophageal cancer among various subgroups. The expenditures were higher in specialized hospitals than in general hospitals and higher in 3A hospitals than in 3A less hospitals, which might be attributed to the more standardized diagnosis and treatment in specialized and 3A hospitals. The average expenditure per patient increased with the number of clinical visits. There were minor differences in the expenditures among age groups at diagnosis. Medical expenditures were lower for those diagnosed at 65 years and older than for younger patients. This was consistent with results from some other studies, such as a study in the United States that reported the overall expenditures among early-stage esophageal cancer patients older than 65 years of age are, on average, 20% lower than those of younger patients [17]. This is probably because older patients have a smaller probability of cure, and physical or economic pressures forced the patients to abandon intensive treatment or to choose palliative care [18]. ESCC is the most common type of esophageal cancer in China [19]. The higher expenditure for ESCC than for other pathologic types was probably because of the availability of more therapy options and standardized procedures for ESCC. The average expenditure per esophageal cancer patient widely differed between therapy subgroups in the present study. The chemotherapy fee grew fastest, which was probably because of newly developed, expensive anticancer drugs. The expenditure of surgery were higher than those of chemotherapy or radiotherapy, which was probably because surgery is expensive and carries corresponding higher fees for diagnosis and inspection.

We found that the expenditure for stage II–III diseases was relatively higher than those for stage I and IV diseases, which seems to be reasonable if we take the differences of both survival and treatment among patients with I–IV diseases into consideration. As we all know, patients with stage I disease were mainly cured by surgery alone, which was relatively cheap, and those with stage IV disease was commonly treated with palliative care and with short survival period, whereas for patients with stage II and III diseases, multidisciplinary treatment were often adopted, which usually contain expensive antitumor drugs [20]. In addition, in the present study, the proportion of selected patients with stage I disease (7.56%) was lower than the target proportion (25%–30%), even after substantial effort, indicating that most esophageal cancer patients were diagnosed with a late stage disease in the absence of routine esophageal cancer screening or early detection and treatment. A cancer screening program may help with the early diagnosis and treatment of esophageal cancer, which could decrease medical expenditure [21].

Our data showed that drug fee was the largest component of medical expenditure and had never been less than 40% since 2003, although China’s health care reform (initiated in 2009) has planned to reduce the prices of drugs on essential drug lists [22]. The drug fee covers the cost of all the medicine used during the treatment, such as anticancer drugs and supportive drugs for symptoms, most of which are expensive joint venture or imported drugs. Additionally, esophageal cancer patients undergo multiple courses of chemotherapy, resulting in a high drug fee. Therefore, reducing drug fee is important for reducing the financial burden of esophageal cancer patients. In addition, the future policy on drug price addition and economic incentives in medical institutions may increase physician-induced demand, increasing the use of drugs and high-tech inspections [23]. In contrast, the proportions of diagnosis and nursing fees were very low, accounting for less than 3% of the overall medical expenditure, which also had a declining trend from 2002 to 2011. Overall, the expenditure proportion reflecting medical workers’ labor value was still relatively low in the diagnosis and treatment of esophageal cancer. Adjusting charges for medical services to appropriately reflect the value of medical staff services, reduce physician-induced demand, and decrease medical expenditures is very important.

The current analysis has several limitations. First, the gender-specific and stage-specific sample sizes were controlled at the investigation phase, inducing selection bias and potentially decreasing the generalizability of the data. Second, this hospital-based survey only abstracted the expenditure data within the study hospitals; however, some patients also received diagnoses and treatments from other hospitals, suggesting that our data probably under-estimated the economic burden of esophageal cancer. Further follow-up studies are needed to overcome this limitation. Third, in this study, we did not obtain insurance reimbursement information, making it impossible to analyze the actual economic burdens from out-of-pocket payments among esophageal cancer patients during 2002–2011. Finally, this analysis only reported medical expenditure data and did not include non-medical expenditure, which is another important component of expenditure estimation from a patient’s perspective. The non-medical expenditure will be separately reported by the health economic evaluation group of the CanSPUC program using data obtained from face-to-face interviews with esophageal cancer patients.

In conclusion, this study showed that medical expenditure for esophageal cancer in China was burdensome and substantially increased during 2002–2011. To dramatically decrease the economic burden on esophageal cancer patients, solutions are required to reduce personal payment ratio and to reduce medical expenditures incurred by services that are not covered by medical insurance. Our future analysis will use a model-based approach to perform a more aggressive estimate of the medical expenditure per esophageal cancer patient while considering variations, such as the numbers of clinical visits/admissions reported by individual study sites.