Screening methods
EC Screening and early treatment was performed strictly following the National Guideline for Early Detection of Cancer[24]. The summarized screening procedures was that eligible residents aged 40 to 69 years can voluntarily participated the screening, after signed the informed consent, the participants were recommended to do the physical examination. Participants meeting the clinic screening criteria will then be examined by the standard endoscopy accompanied with Lugol’s iodine staining and indicative biopsy. All the histological diagnosis were conducted by the expert pathologists according to the AJCC cancer stage (Seventh Edition)[25]. Patients with clear histological diagnosis were recommended to undergo the following treatment. Individuals identified as LGIN were recommended to complete additional screening with endoscopy in 1-3 years. Patients diagnosed with intraductal carcinoma (IC) would undergo the standard treatment of the endoscopic submucosal dissection, while surgery was highly recommended for the cases with the submucosal cancer(SM) stage. For patients with moderate stage, surgery plus adjuvant chemoradiation was suggested, while cases with distant metastasis would be treated with chemoradiation or symptomatic treatment.
Model structure
A decision analysis Markov model of EC with 11 health states was built in Treeage Pro (2019). The health states included: normal, LGIN, IC, SM, moderate cancer stage (Mod), advanced cancer stage (Adv), disease-free survival of IC (DFS_IC), disease-free survival of SM (DFS_SM), disease-free survival of Moderate stage (DFS_Mod), disease progress free of Advanced stage (PFS_Adv), and death. Here, IC included high-grade intraepithelial neoplasia, while moderate stage included stage IB, stage II and stage III. Stage IV was classified as an advanced cancer stage. Overall, IC and SM both constituted the early EC stage, while moderate and advance stages were identified as the invasive EC stage. Fig.1 summarizes the state transition processes, with the arrows presenting the transition between states.
Individuals aged 40 to 69 years were assumed to be the participants. They were classified as six age groups, separated by five-year intervals (ages 40-44, 45-49, 50-54, 55-59, 60-64, and 65-69 years). Cohort simulation was performed until the cohort age reached 79 years or hypothetical death. A hypothetical cohort with 100,000 assigned for each age group. And three different strategies were assigned for each cohort: (1) non-screening, a strategy that assuming all the individuals were not screened, patients were diagnosed by clinical symptoms; (2) screening with follow-up for LGIN, a strategy that assuming all the individuals undergo the one-time standard endoscopic screening, patients were identified by the screening. Meanwhile, assuming the LGIN undergo the annual endoscopic surveillance; (3) screening without follow-up, a strategy that assuming all the individuals undergo the one-time standard endoscopic screening, patients were identified by the screening. However, endoscopic surveillance of LGIN individuals was not required. Assuming all the patients identified by the three strategies had the correct diagnosis and adhered to the standardized treatment scenarios.
Model simulation
QALY presented the effectiveness, and the incremental cost-effectiveness ratio (ICER) served as the economic evaluating indicator. ICER means the cost per unit of additional effectiveness. The calculation formula was: ICER= Cost-effectiveness analyses were used for the comparisons between the competing strategies, including “absolutely dominated strategy”, an option that had both more costs and less effectiveness, “extended dominated strategy”, an option that was less costs and less effectiveness than the alternative but had a higher ICER, and “undominated strategy”, an option that had the cost-effectiveness under the criterion recommended by the WHO[26]. Other outcomes assessed included costs, EC cumulative incidence, and mortality. Moreover, the willingness-to-pay (WTP) was set as three times the gross domestic product per capita (GDP) (USD 77,000) in 2017 in Zhejiang Province in China.
Parameters and Data Resources
Cohort initial probabilities
Initial cohorts’ probabilities in non-screening cohorts were calculated by the 2012 age-specific incidence of EC in Zhejiang province, multiplied by the stage distribution by the time of diagnosis that was obtained from the hospital-based retrospective study[27]. Whereas, the screening cohorts were computed by the age-specific detection rate of EC, which was identified by screening from 2010 to 2017 in Zhejiang province, then multiplied by the stage distribution that was distinguished by screening. Table 1 displays the age-specific incidence and detection rate of EC, while table 2 addressed the stage distributions.
Transition probability between Markov states
Whereas EC annual incidence was used to distinguish the EC cases and normal people in the non-screening cohorts, the risk ratio (RR) of the EC incidence in population after screening compared to people without screening was employed to adjust the EC incidence for the screening cohorts [19]. Also, stage distribution by the time at diagnosis was used to identify EC states in both scenarios. For screening scenario, the probabilities that transferred from LGIN to EC states were the EC annual incidence among the LGIN (table 1), multiplied by the stage distribution identified by screening. The calculation of EC incidence among the LGIN was the adjusted incidence according to the risk ratio of the EC incidence in LGIN compared to that of the true healthy population, combined with the detected proportion of LGIN during screening. The risk ratio was 3.66, which was summarized from published papers[8, 10, 28, 29]. No individuals would transfer between the LGIN and EC states under the non-screening scenario. Other transition probabilities between states were gained from various kinds of literature (table 3).
The age-specific annual death probabilities for the normal population were defined as the difference between the population death probabilities and EC death probabilities. Population mortality was drawn from the sixth population service survey, while the EC mortality was obtained from the 2012 age-specific EC mortality in Zhejiang Province[27, 30]. The LGIN was considered as the precancerous lesion. The EC-specific 5-year survival was 100% for IC and DFS_IC; therefore, people with LGIN or IC were less likely to die from EC. Consequently, the mortality for LGIN, IC, and DFS_IC was assumed to be the same as normal (table 1). Death probabilities of SM and invasive cancer were identified from published papers, while we adjusted the mortality risk according to the age (table 3).
Moreover, all the transition probabilities between states were presented as one-year probabilities. The transition between rate and probability was used to compute the transition probabilities from different follow-up periods. First, a one-year rate (r) was calculated by formula One-year probability (p) was then calculated by formula p=1- exp (-r), where t was the follow-up time[31].
Costs and utilities
Costs were estimated from a social perspective. And it consisted of screening costs, treatment costs, transportation and wage loss of patients and relatives due to the hospital visiting. Screening costs were calculated using the data of the screening program in Zhejiang Province, while treatment costs were extracted from the electronic medical record at Zhejiang Cancer Hospital and the price of medical services in provincial public hospitals in Zhejiang Province[32]. Every patient was assumed to have one accompanying relative. All costs were measured in the 2017 Chinese currency and were changed into US dollars using the purchasing power parities with 3.506 in 2017[33]. State-specific utilities were extracted from published papers and the results of the screening. A discount rate of 5% was used for both costs and effectiveness [34, 35]. All the items of costs in the study will inflate at the same inflation rate of 4.7%[36]. The total state-specific costs and utilities were displayed in table 4 and table 5.
Sensitivity analysis
Sensitivity analyses for cost-effectiveness screening strategies were conducted. Probabilistic sensitivity analyses were performed to examine the influence of the multiple parameters that varied simultaneously on the outcome. Initial cohort probabilities and death probabilities assumed as Beta distribution, while the discount rate and inflation rate identified as Triangular distribution. Gamma distribution was set for costs. Moreover, Beta and Dirichlet were assigned to transition probabilities. One-way sensitivity analyses were simulated to access the effect of the single parameter changed on the outcome. The following assumption was made. Initial probabilities, state transition probabilities, risk ratios, and health utility were varied by±20% of the base case value, while costs changed by±30% of the base case value. Besides, 0-8% was simulated for the discount rate, and 3.2- 6.2% was used for the inflation rate.