We conducted a screening cohort study within the framework of the CanSPUC program. We used the household registration system in local communities to identify eligible permanent residents who were aged 40‐74 years and asymptomatic for UGC with no history of cancer diagnosis. People who were unable to give informed consent, had a prior cancer diagnosis, or were receiving therapy for other serious medical conditions were ineligible for participation. Television broadcasts, brochures, and websites were used to publicize cancer screening programs. Community-based telephone calls or home visits were carried out by trained staff to reach and inform the maximum number of eligible residents.

We used data from the UGC screening performed over the initial 6 years, spanning from October 2013 to August 2019, in 4 cities within Zhejiang Province: Hangzhou, Ningbo, Quzhou, and Jinhua city. In total, 153,079 eligible participants underwent risk assessment in our UGC screening program. Figure 1 displays a flowchart of the study population.

Participants deemed eligible were initially required to complete a risk evaluation before undergoing endoscopy procedure. To distinguish individuals with a high risk of developing UGC, we used a cancer-related risk assessment questionnaire developed by the CanSPUC group. This questionnaire covered a range of information, including basic information, dietary habits, living environment, lifestyle and habits, psychological and emotional aspects, medical history, and family history of cancer. To evaluate UGC risk, we used gender-specific risk scoring systems derived from the Harvard Cancer Risk Index [21]. A panel of experts assigned a score to each risk factor, considering the strength of its correlation with UGC. These individual risk scores were then totaled and divided by the average risk score in the general population, resulting in a relative risk assessment for each participant. The estimation process is illustrated in detail in Multimedia Appendix 1.

Individuals identified as high risk for UGC were advised to undertake a one-off endoscopy at a specialized tertiary-level hospital designated by the program. Further diagnostic or follow-up procedures were decided depending on the endoscopy results. Any anomalies discovered during the endoscopy were meticulously examined using biopsies. Additionally, visual documentation was made of any endoscopy findings. Individuals who did not have a complete endoscopy examination were encouraged to undergo a repeat endoscopy to fulfill the necessary clinical standards for diagnosis.

To address potential bias stemming from differences in the time during which participants in the screened cohort were required to stay alive and free from certain events (immortal time), we determined the entry date as the day of screening for participants in the screened group. In the nonscreened group, the cohort entry date was approximated based on the screening date of the participant in the screened group whose risk assessment date most closely matched that of the nonscreened group. The duration until the occurrence of UGC was computed as the time from the entry date to the earliest of 3 events: the development of UGC, death, or administrative censoring (up to June 30, 2021). Similarly, the duration until UGC-specific or all-cause death was calculated as the time from the cohort entry date to the earliest death or administrative censoring.

Our focus was on 3 key outcomes: new cases of UGC, UGC-related deaths, and deaths from all causes. We identified UGC using the International Classification of Diseases, 10th Revision, where EC was coded as C15 and GC as C16. Every 6 months, we collected outcome data from national sources, which included the cancer registry and death surveillance system.

Our data collection involved standardized web-based forms, including an epidemiological questionnaire and endoscopy reports, which were collected by our qualified staff and physicians face-to-face. A thorough verification of consistency was carried out, and any discrepancies were rectified by referring to the original documents. Each participant was assigned a unique identification number, which helped us keep track of all relevant documentation.

We compared the screened versus nonscreened groups and the high-risk versus low-risk groups using standardized differences. These differences were derived by dividing the mean or proportion variations by a combined SD estimate [22]. This analytical approach is not heavily influenced by sample size and is efficacious in discerning significant discrepancies, with standardized differences exceeding 0.1 being deemed significant. For UGC outcomes, the estimation of the cumulative incidence or mortality was carried out using the cumulative incidence or mortality function, with due consideration given to the competing risk of mortality or other causes. All-cause mortality was defined as any cause of mortality, and cancer-specific mortality was regarded as mortality from UGC. The Gray test was used to evaluate discrepancies between groups concerning cumulative UGC incidence and UGC mortality. All-cause mortality was assessed using the Kaplan-Meier method with a log-rank test to discern differences between groups. We used a Cox proportional hazards model to determine the respective association of endoscopy screening with each outcome. Hazard ratios (HRs) and 95% CIs were reported. All hypothesis tests were 2-sided. Statistical analyses were conducted using R software (version 3.5.1; R Foundation for Statistical Computing), considering P values less than .05 as indicative of statistical significance.

The study was approved by the ethics committee of the Cancer Hospital Chinese Academy of Medical Sciences (approval 15-070/997) and Zhejiang Cancer Hospital (approval IRB-2022‐271). All participants provided written informed consent.

We conducted a multicenter, large-scale, prospective cohort study to investigate the effectiveness of sequential UGC screening in a real-world situation. Our results highlight a substantial decrease in both all-cause mortality and UGC mortality among high-risk individuals subjected to one-off endoscopic screening compared to the nonscreened group. These findings carry significant implications for clinical practice and public health, emphasizing the real benefits associated with the implementation of effective UGC screening programs.

UGC continues to represent a substantial population health challenge on a global scale. While evaluations of the program’s effectiveness have been conducted over several decades, the routine impact of the screening program remains unclear. Countries such as Japan and Korea in East Asia have also implemented nationwide population-based screening strategies to address the high burden of UGC and reduce its incidence and mortality, showing that implementing gastroscopic screening in high-risk rural areas can lower the mortality related to EC and GC [5]. The Korean National Cancer Screening Program found an overall odds ratio of GC mortality among individuals ever screened as 0.79 (95% CI 0.77‐0.81) relative to those who had never undergone screening [13,23]. On the other hand, Hamashima et al [24] revealed a standardized mortality risk ratio of 0.43 (95% CI 0.30‐0.57) for the gastroscopic screening participants in Japan. Another study illustrated decreases in cumulative mortality from UGC, with a standardized mortality rate of 0.47 (95% CI 0.25‐0.88) [25]. Despite these studies, the effectiveness of UGC screening programs remains a topic of ongoing research and discussion. The effectiveness of UGC screening in lowering mortality may be ascribed to the identification of precancerous abnormalities. The majority of cancers detected through screening were in their early stages, including precancerous lesions. These early-stage malignancies have a lower likelihood of progressing compared to advanced-stage cancers if treated promptly. Hence, the sustained decrease in UGC mortality over time aligns with the natural course of UGC, affirming the continued impact of timely intervention.

Apart from the above studies, evidence derived from large-scale, high-quality cohort studies regarding the association with all-cause mortality is sparse. One of the most compelling aspects of our results is the impact on all-cause mortality. The fact that screening significantly lowered 36% of all-cause mortality indicates that the benefits of UGC screening extend beyond merely reducing UGC-related deaths. A positive impact on overall mortality may be more conspicuous in nations with constrained medical resources and fewer diagnostic capabilities compared to countries with ample medical provisions. Individuals who participate in such screening programs experience better overall health outcomes and a lower risk of death from various causes. This suggests that the positive effects of screening are not limited to cancer-specific outcomes but can have a broader impact on an individual’s overall well-being. However, research conducted among Western populations did not consistently demonstrate a meaningful correlation [26]. The disparity could be attributed to lower UGC incidences and mortality rates in Western populations compared to East Asian populations [1]. In Europe, Areia et al [26] pointed out that the cost-effectiveness of gastroscopic screening was only evident when accompanied by colonoscopy screening in countries where the risk of GC is ≥10 per 100,000.

As expected, a greater UGC incidence density was observed in both the high-risk and screened groups, which can be attributed to increased opportunities for early diagnosis and treatment. Early detection is a key determinant of improved cancer treatment outcomes [27,28]. When UGC is diagnosed at an advanced stage, the treatment options become limited, and the prognosis tends to be less favorable. However, screening programs, as demonstrated in our study, offer a powerful tool to address this issue. One of the notable outcomes we observed was a marked increase in incidence rates within the screened group. This is a positive outcome, and it underscores the fact that screening programs effectively identify cases of UGC at earlier, more treatable stages. By increasing the opportunities for early diagnosis, screening facilitates timely medical intervention, which is known to lead to better treatment success rates. The rationale behind this increase in incidence rates within the screened group is that the screening process identifies UGC cases that might have otherwise gone undetected until they reached a more advanced and less treatable stage. When UGC is diagnosed early, patients can receive appropriate medical care, including surgical intervention, targeted therapy, or radiation treatment, all of which are associated with higher success rates in terms of disease management and patient survival. Additionally, early detection allows for a more focused and less aggressive treatment approach. This means that patients with UGC detected through screening may undergo less invasive surgical procedures, experience fewer treatment-related side effects, and have a higher likelihood of achieving complete remission. As a result, the quality of life for these individuals can be significantly improved.

We also highlight variations in the effectiveness of screening among different age groups and genders. Notably, the benefits of screening appear more pronounced in older participants, particularly concerning UGC mortality. This may be linked to age-related differences in cancer progression and treatment response. It has been reported that advanced age is associated with an increased risk of developing UGC [29]. The UGC incidence, all-cause mortality, and UGC mortality in the age group of 55‐74 years are all remarkably higher than those in the age group of 40‐54 years, consistent with findings from other studies [30,31]. In addition, our study showed that older adult individuals showed markedly reduced risks of all-cause mortality (HR 0.65, 95% CI 0.49‐0.86) and UGC mortality (HR 0.19, 95% CI 0.05‐0.80) after the endoscopic screening in contrast to the younger group. Furthermore, we observed a more pronounced effect of screening in males, with significant reductions in both all-cause mortality and UGC mortality, whereas the effect was less significant in females. This could reflect gender-specific disparities in UGC incidence and survival. These differences may be associated with unfavorable lifestyle habits in males, such as smoking, alcohol consumption, and irregular dietary patterns [32]. Moreover, we observed similar reductions in all-cause mortality for smokers (0.61, 95% CI 0.44‐0.86) and nonsmokers (0.62, 95% CI 0.41‐0.95) after endoscopic screening, which means the screening initiative contributes to overall health outcomes, irrespective of smoking status. While the study reveals significant reductions in UGC mortality among smokers (HR 0.10, 95% CI 0.01‐0.76), the absence of a similar effect in nonsmokers raises intriguing questions. The explanation for this may lie in several factors, including distinct etiological pathways of UGC in smokers and nonsmokers, or differences in health awareness among these 2 populations. Further exploration of these nuances could enhance our understanding of the intricate interplay among risk exposures, screening effectiveness, and mortality outcomes.

Our study has several limitations. First, the follow-up duration of 6.05 years might be insufficient for capturing all death cases, potentially leading to an underestimation of the long-term impact of the screening program. Ideally, a more extended follow-up period would afford a comprehensive understanding of sustained benefits and potential late-emerging effects. Future research endeavors should consider prolonging follow-up periods to fully evaluate the program’s long-term effectiveness. Second, one must consider that the findings might not be universally representative across the entire general population of China. However, the study does offer a valuable scientific foundation for areas exhibiting equivalent socioeconomic profiles. Third, the screening compliance rate of 24.6% among high-risk individuals raises concerns about the generalizability of the findings to the entire high-risk individuals. The relatively low uptake rate may give rise to selection bias, influencing observed outcomes. However, it marginally surpassed the figure reported for UGC screening in Henan Province (18.4%) and Liaoning Province (22.3%), China [10,31,32]. Future research could explore strategies to enhance participation, such as community engagement programs, targeted awareness campaigns, or initiatives aimed at overcoming potential barriers to screening. These measures would ensure a more representative study population and enhance the generalizability of the findings. Future randomized controlled trials may be warranted to further validate the efficacy of risk-adapted sequential screening strategies.

In conclusion, our study suggests the significance of UGC screening in reducing both all-cause mortality and UGC mortality, particularly among high-risk individuals. Screening offers the opportunity for early UGC detection, enhancing treatment success rates and alleviating the disease burden for patients. However, screening effectiveness may vary by age and gender, necessitating consideration when implementing screening programs. Our research provides robust support for further exploration and implementation of UGC screening, with the potential to alleviate the disease burden of UGC.