Introduction
Liver cancer ranks as the sixth most commonly diagnosed cancer and the third leading cause of cancer death, with 865,000 new cases and 757,948 deaths worldwide.1 The age-standardized mortality rate for liver cancer has declined over the past decades, but liver cancer remains the second leading cause of cancer-related deaths in China, representing a substantial disease burden.2 Despite guidelines from international and national organizations recommending liver cancer screening for high-risk populations,3–5 only a minority of liver cancer patients are diagnosed at an early stage.6,7 These poor outcomes are largely attributable to inadequate early screening and detection among high-risk individuals.8
Abdominal ultrasonography combined with alpha-fetoprotein (AFP) testing has been recommended for liver cancer screening by the American Association for the Study of Liver Disease (AASLD).9 This strategy is supported by a large randomized controlled trial among patients with hepatitis B virus infection or a history of chronic hepatitis, which demonstrated improved early-stage liver cancer detection and a 37% reduction in liver cancer mortality compared with the control group.10 A systematic review and meta-analysis of randomized controlled trials and cohort studies has shown that liver cancer screening increases the proportion of early-stage detection and reduces mortality in high-risk populations.11 Of note, population-based liver cancer screening, which aims to detect early-stage liver cancer among high-risk individuals in the general population, is distinct from hepatocellular carcinoma (HCC) surveillance. However, the potential benefits of screening can only be realized with high uptake among the target population. Insufficient participation limits the population-level impact of screening programs.
Notably, even within clinical HCC surveillance, where patients are already engaged in routine care, a systematic review and meta-analysis reported a pooled uptake of only 52%,12 population-based screening is likely to fare worse, as it must actively reach at-risk individuals outside routine care. Consistent with this, a global modelling study estimated that only 37% of HCC cases worldwide were detected through screening or surveillance.7 Identifying population-specific determinants of screening uptake is therefore essential, as these factors represent modifiable targets for improving programme reach.
Previous studies have primarily focused on individual-level factors associated with liver cancer screening participation, such as age, education, and health awareness, although provider recommendation has also been identified as a key facilitator.13 However, structural-level factor—including regional economic development, community mobilization strategies, accessibility of screening services, and financial incentives—have received comparatively less attention, despite their potential to influence population-level screening uptake.14 These structural determinants are potentially modifiable through program design and health policy and therefore represent important targets for intervention to improve screening participation at scale.
Following the standardized protocol of the national program, individuals aged 45–74 years were invited to undergo risk assessment based on a risk scoring system derived from the Harvard Cancer Risk Index, which integrates established liver cancer risk factors to identify high-risk individuals. In our study, we report findings from the liver cancer screening component of the Cancer Screening Program in Urban China (CanSPUC),15 conducted from October 2013 to October 2022 in Chongqing, a municipality in southwest China with a population of over 30 million. Our objectives were to assess liver cancer screening uptake rates among high-risk individuals, examine reasons for non-participation, and identify individual-level and structural factors associated with uptake.
In addition, we reported the diagnostic yield of liver cancer screening and provided evidence-based recommendations for optimizing screening strategies.
Methods
Study Design and Population
This study was conducted within the framework of the CanSPUC, an ongoing national cancer screening program funded by the Ministry of Finance and the National Health Commission of China.15 In Chongqing, 177 communities across 18 districts or counties were selected as participating areas based on complete cancer registration data, vital statistics, and relatively stable populations.16 The study comprised two components with distinct designs. First, a cross-sectional risk assessment was conducted, in which eligible participants aged 45–74 years were invited to undergo risk assessment for liver cancer. Detailed information on eligibility criteria is provided in Supplementary Appendix 1.
Second, a prospective follow-up was conducted, in which high-risk individuals completed an epidemiological questionnaire to collect individual-level and structural characteristics, and were then followed for their uptake of liver cancer screening within the program. The study was approved by the ethics committees of the Chinese Academy of Medical Sciences’ Cancer Hospital (approval number 15–070/997) and Chongqing University Cancer Hospital (approval number CZLS2022196-A). All enrolled participants provided written informed consent prior to the survey.
Procedures and Quality Control
Eligible participants aged 45–74 years were invited to complete a structured, computer-assisted standardized questionnaire. In addition, participants were tested for hepatitis B surface antigen (HBsAg). After completing the questionnaire and HBsAg testing, each participant underwent a risk assessment for liver cancer based on a risk scoring system derived from the Harvard Cancer Risk Index.
The risk assessment system is described in Supplementary Appendix 2. All high-risk individuals were advised to undergo free abdominal ultrasound combined with AFP testing at tertiary-level hospitals. Different strategies were used in each community to motivate the high-risk participants to attend liver cancer screening.
The normal range of AFP was determined by the reference interval of each sub-center’s instrument. Ultrasound examinations were performed by an attending physician or above with at least five years of ultrasound experience, who held a qualification certificate for large-scale ultrasound equipment. Positive cases were reviewed by a deputy chief physician or above.
Furthermore, individuals with positive findings within the past five years who had not been diagnosed with liver cancer underwent annual re-evaluation, including those with: (1) suspected liver cancer; (2) space-occupying lesions in the liver (excluding simple hepatic cysts and hepatic hemangiomas); and (3) positive HBsAg with abnormal AFP levels. Dedicated pathologists assessed all abnormal findings in accordance with the latest World Health Organization classification.17 The flowchart of participant enrollment is presented in Figure 1, and the flowchart of the entire screening process is shown in Supplementary Figure S1.Figure 1Flowchart of participants in liver cancer screening program.Flowchart sequence: Enrolled, Excluded, High risk, Excluded, Included, Accepted screening, Not accepted.Flowchart sequence: Participants aged 45 to 74 years enrolled for risk assessment, Excluded incomplete risk assessment questionnaire 69 existing liver cancer before risk assessment 164 death before risk assessment 1649 low risk participants 330130, Participants assessed to be high risk of liver cancer 37381, Excluded death before screening 2 screened before assessment 44 incomplete questionnaire on community structural characteristics 2129 individuals not contacted by the project outreach 172 incomplete screening 387, Participants included in analysis 34647, Accepted liver cancer screening 14062, Not accepted liver cancer screening 20585.
To identify potential factors associated with the liver cancer screening uptake, we considered multiple variables, including: economic status (developing or developed areas), distance from the community hospitals to the screening hospital, number of baseline comorbidities (respiratory disease, hepatobiliary disease, hypertension, hyperlipidemia, and diabetes), number of mobilization methods (face-to-face communication, text messages, WeChat, telephone, home visits, etc), rewards (financial or material incentives provided to participants), and allowances (covering transportation, meals, work delay, or auxiliary examinations).
These variables were measured at different levels: economic status was a district-level variable, based on the 2022 GDP of each district or county, which was used because the overall development pattern across the participating districts remained largely stable during the study period; mobilization methods, distance to the screening hospital, rewards and allowances were community-level variables, reflecting the strategies and services provided by each community to its residents; and baseline comorbidities were individual-level variables, based on self-reported information.
For quality control, participants’ screening-related information and images were transmitted to the Central Data Management Team at the National Cancer Center of China, where the databases were constructed and analyzed. Ultrasound images of solid liver lesions and undiagnosed cases were reviewed by the project team experts. A 3% sampling review was conducted for negative cases.
Outcome Definitions
The primary outcome was liver cancer screening uptake, defined as whether high-risk individuals underwent abdominal ultrasound combined with AFP testing within 6 months following the baseline risk assessment. Reasons for non-participation and individual-level and structural factors associated with uptake were also examined. The secondary outcome was the diagnostic yield of liver cancer screening, defined as a confirmed diagnosis of liver cancer, verified by hospitals at or above the secondary level with cancer diagnostic qualifications.
Diagnoses were coded in accordance with the International Classification of Diseases, 10th Revision (ICD-10), with liver cancer assigned to code C22.
Statistical Analysis
We calculated crude and standardized uptake rates of liver cancer screening among high-risk individuals. Standardized uptake rates were directly standardized by age, sex, education, and economic status, using the entire study population as the standard population. Due to the sensitivity of P-values to large sample sizes, standardized differences were used as the primary indicator for group comparisons.
To identify factors associated with screening uptake, we used multivariable unconditional logistic regression models and generalized linear mixed models (GLMMs). In the GLMMs, community-level random intercepts were included as random effects to account for clustering, and covariates were included as fixed effects. Missing data were handled by complete-case analysis.
All covariates were included in the models based on a priori hypotheses and clinical relevance, without automated variable selection. Additionally, sensitivity analyses were performed using both logistic regression and GLMMs to further evaluate the impact of these factors and ensure the robustness of our findings.
All statistical tests were two-sided, and P<0.05 was considered statistically significant. All analyses were performed using SPSS (version 26.0, IBM Corporation) and R (version 4.2.1) statistical software.
Results
Characteristics of the Study Population
Among the 369,393 participants who completed the risk assessment, 37,381 were evaluated as high risk for liver cancer in 177 communities from 2013 to 2022. After excluding those who were ineligible or had incomplete data, 34, 647 were included in the analysis. The characteristics of the study population are shown in Table 1.
The majority of the participants were female (56.81%). The mean age was 57.32 years (SD=7.19 years), and the largest age group was 50–54 years (23.48%). Approximately 48.78% of participants were overweight or obese, 93.18% were married, 74.54% had a medium or high education level, and 39.71% had a history of occupational exposure to hazards.
Approximately 44.43% of the participants were current smokers or former smokers, 47.94% of the participants drank alcohol, and 66.71% participated in physical activity fewer than three times per week. Hepatitis B and Hepatitis C infections were present in 32.56% and 9.31% of participants, respectively. Approximately 12.00% had a history of liver cirrhosis.
26.93% had three or more baseline comorbidities, 48.86% had a family history of liver cancer. Approximately 57.28% of the participants resided in a developed areas, and 39.30% were located within 5 km of the screening hospital. Approximately 42.96% were exposed to more than three mobilization methods, and 59.88% received more than two types of publicity about the screening program.
Approximately 41.36% received a reward for completing screening, and 29.66% received an allowance.Table 1Characteristics of the Study Population and Uptake Rates for Liver Cancer Screening Between Different GroupsCharacteristicsTotal (n=34647)Yes (n=14062)No (n=20585)Uptake Rate (%)Standardized differenceIndividual-levelAge, years0.111 45–495547(16.01)2460(17.49)3087(15.00)44.35 50–548134(23.48)3293(23.42)4841(23.52)40.48 55–597195(20.77)2944(20.94)4251(20.65)40.92 60–646953(20.07)2944(20.94)4009(19.48)42.34 65–695426(15.66)1977(14.06)3449(16.75)36.44 70–741392(4.01)444(3.15)948(4.60)31.90Gender0.065 Male14,965(43.19)5801(41.25)9164(44.52)38.76 Female19,682(56.81)8261(58.75)11,421(55.48)41.97BMI, kg/m20.055 <18.5750(2.16)281(2.00)469(2.28)37.47 18.5–23.916,997(49.06)6987(49.69)10,010(48.63)41.11 24.0–27.913,447(38.81)5518(39.24)7929(38.52)41.04 ≥28.03453(9.97)1276(9.07)2177(10.57)36.95Education0.086 Low8822(25.46)3242(23.06)5580(27.11)36.75 Medium21,648(62.48)9022(64.16)12,626(61.34)41.68 High4177(12.06)1798(12.78)2379(11.55)43.05Marriage0.048 Unmarried/divorced/widowed2364(6.82)908(6.46)1456(7.07)38.41 Married32,283(93.18)13,154(93.54)19,129(92.93)40.75Occupational exposure to hazards0.226 No20,887(60.29)7557(53.74)13,330(64.76)36.18 Yes13,760(39.71)6505(46.26)7255(35.24)47.27Smoking0.097 Never19,252(55.57)7480(53.19)11,772(57.19)38.85 Current13,225(38.17)5787(41.15)7438(36.13)43.76 Former2170(6.26)795(5.66)1375(6.68)36.64Alcohol0.170 No18,038(52.06)6601(46.94)11,437(55.56)36.59 Yes16,609(47.94)7461(53.06)9148(44.44)44.92Physical activity0.175 <3 times/week23,114(66.71)10,036(71.37)13,078(63.53)43.42 ≥3 times/week11,533(33.29)4026(28.63)7507(36.47)34.91Hepatitis B0.198 No23,365(67.44)8742(62.17)14,623(71.04)37.41 Yes11,282(32.56)5320(37.83)5962(28.96)47.15Hepatitis C0.111 No31,422(90.69)12,592(89.55)18,830(91.47)40.07 Yes3225(9.31)1470(10.45)1755(8.53)45.58Liver cirrhosis0.218 No30,491(88.00)11,982(85.21)18,509(89.91)39.30 Yes4156(12.00)2080(14.79)2076(10.09)50.05Number of comorbidities0.213 07182(20.73)2026(14.41)5156(25.05)28.21 19254(26.71)3634(25.84)5620(27.30)39.27 28879(25.63)4078(29.00)4801(23.32)45.93 ≥39332(26.93)4324(30.75)5008(24.33)46.34Family history of liver cancer0.207 No17,719(51.14)6314(44.90)11,405(55.40)35.63 Yes16,928(48.86)7748(55.10)9180(44.60)45.77Structural-level Economic status0.211 Developing areas14,801(42.72)5138(36.54)9663(46.94)34.71 Developed areas19,846(57.28)8924(63.46)10,922(53.06)44.97Mobilization ways for screening0.112 1~319,762(57.04)7555(53.73)12,207(59.30)38.23 > 314,885(42.96)6507(46.27)8378(40.70)43.72Distance0.033 <5 Km13,616(39.30)5725(40.71)7891(38.33)42.05 5–10 Km11,456(33.06)4542(32.30)6914(33.59)39.65 >10 Km9575(27.64)3795(26.99)5780(28.08)39.63Publicity0.074 1~213,901(40.12)5342(37.99)8559(41.58)38.43 >220,746(59.88)8720(62.01)12,026(58.42)42.03Rewards0.080 No20,317(58.64)7915(56.29)12,402(60.25)38.96 Yes14,330(41.36)6147(43.71)8183(39.75)42.90Allowance0.122 No24,370(70.34)9456(67.25)14,914(72.45)38.80 Yes10,277(29.66)4606(32.75)5671(27.55)44.82Note: Bold indicates standardized difference > 0.1, suggesting statistically significant differences between groups.
Uptake Rates for Liver Cancer Screening
A total of 14,062 individuals underwent liver cancer screening, corresponding to a screening uptake rate of 40.59% (14,062/34,647). The uptake rate showed a significant downward trend with increasing age [44.42% (45–49 years) vs 31.97% (70–74 years)]. Significantly higher uptake rates were found among females (42.00%), highly educated individuals (43.10%), individuals who consumed alcohol (45.00%), and those residing in developed areas (45.03%).
The standardized uptake rates were consistent with the crude uptake rates (Figure 2).Figure 2Crude and standardized uptake rates of liver cancer screening by subgroups.A horizontal bar graph showing crude and standardized liver cancer screening compliance rates by subgroup.A horizontal bar graph comparing crude rate and standardized rate for liver cancer screening compliance by subgroup. X axis label, Compliance rate of L C screening left parenthesis percent right parenthesis, range 0 to 50. Y axis categories: Year, Age, Gender, Education, Alcohol, Economic status.
Legend: Crude rate and Standardized rate.
Factors Associated with Uptake Rate for Liver Cancer Screening
We performed multivariable logistic regression to further explore the potential factors associated with screening uptake. Considering slight changes in uptake rates between years of recruitment, the GLMM with adjustment for year of recruitment was performed, the results are shown in Table 2. Compared with participants aged 45–49 years, those aged 65–69 years, and 70–74 years had 20% and 30% lower uptake rates, respectively (OR 0.80, 95% CI 0.74–0.87; OR 0.70, 95% CI 0.61–0.80).
Individuals with occupational exposure to hazards and those who drank alcohol were 30% and 9% more likely to undergo liver cancer screening, respectively (OR 1.30, 95% CI 1.23–1.36; OR 1.09, 95% CI 1.04–1.15). More physical activity (≥3 times a week) was associated with lower uptake rates (OR 0.81, 95% CI 0.77–0.85). Participants with hepatitis B infections and a history of liver cirrhosis were more likely to undergo screening (OR 1.40, 95% CI 1.36–1.50; OR 1.28, 95% CI 1.19–1.38), whereas no statistical association was found for hepatitis C infections.
Compared with participants without baseline comorbidities, those with 1, 2, and ≥3 comorbidities had 42%, 80% and 79% higher uptake rates, respectively (OR 1.42, 95%CI 1.33–1.53; OR 1.80, 95%CI 1.68–1.94; OR 1.79, 95%CI 1.66–1.93). Similarly, participants with family history of liver cancer were 39% more likely to undergo screening than those without (OR 1.39, 95% CI 1.32–1.47).Table 2Individual-Level and Structural-Level Factors Associated with Uptake Rates for Liver Cancer ScreeningCharacteristicsModel I*Model II#OR (95%CI)P-valueOR (95%CI)P-valueIndividual-levelAge,years 45–491.001.00 50–540.92(0.86–0.99)0.0290.96(0.89–1.03)0.221 55–590.95(0.88–1.02)0.1421.00(0.93–1.08)0.944 60–640.98(0.91–1.05)0.5850.98(0.91–1.05)0.516 65–690.83(0.77–0.90)<0.0010.80(0.74–0.87)<0.001 70–740.77(0.68–0.87)<0.0010.70(0.61–0.80)<0.001Occupational exposure to hazards No1.001.00 Yes1.25(1.19–1.31)<0.0011.30(1.23–1.36)<0.001Alcohol No1.001.00 Yes1.11(1.06–1.17)<0.0011.09(1.04–1.15)<0.001Physical activity <3times/week1.001.00 ≥3times/week0.81(0.77–0.85)<0.0010.81(0.77–0.85)<0.001Hepatitis B No1.001.00 Yes1.29(1.23–1.36)<0.0011.40(1.36–1.50)<0.001Hepatitis C No1.001.00 Yes0.92(0.85–1.00)0.0450.99(0.91–1.07)0.739Liver cirrhosis No1.001.00 Yes1.23(1.15–1.32)<0.0011.28(1.19–1.38)<0.001Number of comorbidities 01.001.00 11.37(1.28–1.47)<0.0011.42(1.33–1.53)<0.001 21.69(1.57–1.81)<0.0011.80(1.68–1.94)<0.001 ≥31.50(1.40–1.62)<0.0011.79(1.66–1.93)<0.001Family history of liver cancer No1.001.00 Yes1.35(1.29–1.41)<0.0011.39(1.32–1.47)<0.001Structural-levelEconomic status Developing areas1.001.00 Developed areas1.50(1.43–1.57)<0.0011.40(1.33–1.48)<0.001Mobilization ways for screening 1~31.001.00 >31.14(1.08–1.19)<0.0011.16(1.11–1.23)<0.001Allowance No1.001.00 Yes1.23(1.16–1.30)<0.0011.17(1.11–1.24)<0.001Note: Bold indicates P < 0.05. *Factors included in non-conditional binary logistic regression model. #Factors were additionally adjusted by year of recruitment in generalized linear mixed model (GLMM).
Furthermore, participants residing in developed areas (OR 1.40, 95% CI 1.33–1.48), those receiving more than three mobilization methods (OR 1.16, 95%CI 1.11–1.23), and those receiving allowances (OR 1.17, 95%CI 1.11–1.24) had higher uptake rates. Considering the impact of economic status on uptake, we conducted a subgroup analysis to explore the factors associated with liver cancer screening uptake across regions (Table 3). In developed areas, receiving more than three mobilization methods (OR 0.95, 95%CI 0.89–1.01) and receiving allowances (OR 1.07, 95%CI 1.00–1.15) were not significantly associated with uptake.Table 3Individual-Level and Structural-Level Factors Associated with Uptake Rates by Economic Status in the Screening ProgramCharacteristicsDeveloping areasDeveloped AreasOR (95%CI)P-valueOR (95%CI)P-valueIndividual-levelAge,years 45–491.001.00 50–540.97(0.86–1.10)0.6570.92(0.84–1.01)0.900 55–590.98(0.86–1.11)0.7290.98(0.89=1.08)0.663 60–641.00(0.88–1.14)0.9570.93(0.84–1.02)0.108 65–690.78(0.69–0.89)<0.0010.78(0.70–0.86)<0.001 70–740.69(0.57–0.82)<0.0010.64(0.52–0.79)<0.001Occupational exposure to hazards No1.001.00 Yes1.15(1.06–1.25)0.0011.40(1.31–1.49)<0.001Alcohol No1.001.00 Yes1.16(1.07–1.26)<0.0011.11(1.04–1.18)0.001Physical activity <3times/week1.001.00 ≥3times/week0.80(0.74–0.87)<0.0010.79(0.74–0.84)<0.001Hepatitis B No1.001.00 Yes1.40(1.29–1.52)<0.0011.44(1.34–1.53)<0.001Hepatitis C No1.001.00 Yes0.99(0.88–1.12)0.9221.11(0.99–1.25)0.067Liver cirrhosis No1.001.00 Yes1.38(1.25–1.53)<0.0011.22(1.09–1.36)<0.001Number of comorbidities 01.001.00 11.52(1.37–1.70)<0.0011.33(1.21–1.47)<0.001 21.88(1.68–2.10)<0.0011.73(1.58–1.91)<0.001 ≥31.60(141–1.81)<0.0011.92(1.74–2.12)<0.001Family history of liver cancer No1.001.00 Yes1.47(1.35–1.61)<0.0011.33(1.24–1.42)<0.001Structural-levelMobilization ways for screening 1~31.001.00 > 31.55(1.42–1.70)<0.0010.95(0.89–1.01)0.118Allowance No1.001.00 Yes1.30(1.17–1.45)<0.0011.07(1.00–1.15)0.065Note: Bold indicates P < 0.05.
A sensitivity analysis was conducted among 34,595 participants after excluding 52 individuals who had been diagnosed with liver cancer within 6 months of risk assessment. Among them, 14,062 (40.65%) underwent liver cancer screening. The results were consistent with the main analysis (Supplementary Tables 1 and 2).
Reasons for Unwillingness to Participate in Screening
Among the 34,647 high-risk participants, 20,585 did not undergo liver cancer screening. Participants could select more than one reason for non-participation. The five most frequently reported reasons were “no symptoms, do not think they would get liver cancer (93.61%)”, “afraid of being diagnosed with liver cancer (92.60%)”, “lack of knowledge about cancer prevention and treatment (89.30%)”,“unawareness of the benefits of cancer screening (86.99%)” and “unable to understand the screening process and content (76.12%)” (Figure 3).Figure 3Reasons for not participating in liver cancer screening; participants could select more than one reason.A horizontal bar graph showing reasons for not participating in liver cancer screening.
Liver Cancer Screening Results
Table 4 presents the diagnostic yield of liver cancer screening. Within 6 months following screening, 14 cases of pathologically confirmed primary liver cancer were identified, corresponding to a detection rate of 0.10%. Other liver lesions detected included 26 cases of liver cirrhosis, 417 cases of hepatic hemangiomas, 2868 cases of fatty liver, and 1470 cases of hepatic cyst, with corresponding detection rates of 0.18%, 2.97%, 20.40%, and 10.45%, respectively.Table 4Liver Lesions Detected by Ultrasonography in the Screening ProgrammelesionsCases (n)Detection rate, % (95%CI)Liver cancer140.10 (0.06–0.17)Liver cirrhosis260.18 (0.12–0.26)Hepatic hemangioma4172.97 (2.70–3.26)Fatty liver286820.40 (19.74–21.07)Hepatic cyst147010.45 (9.96–10.97)Splenomegaly690.49 (0.39–0.62)Space-occupying lesion of the liver390.28 (0.21–0.38)Cholecystitis400.28 (0.21–0.38)Bile duct dilatation50.04 (0.02–0.09)Other lesions2191.56 (1.37–1.78)
Discussion
To our knowledge, this is one of the largest population-based prospective studies to evaluate the uptake and associated factors of liver cancer screening in Southwest China. Abdominal ultrasound combined with AFP testing is widely considered the optimal screening strategy. However, low participation rates remain a major challenge in population-based liver cancer screening.
In our study, the participation rate was 40.59%. This rate was slightly higher than those reported in Korea (32.59%)18 and the surveillance rate reported among patients with liver cirrhosis in the United States (18.4%),19 but slightly below the rates documented across four provinces in China (74.06%).20 Overall, poor uptake of liver cancer screening is a common problem in real-world population screening in countries with limited medical resources. If the uptake rate is insufficient, it negates the intended benefits of reducing morbidity and mortality and results in inefficient allocation of medical resources.
Therefore, addressing various barriers and improving screening uptake are key measures to enhance the implementation of liver cancer screening programs in high-risk populations.
Consistent with previous studies,21,22 we identified several individual characteristics as key determinants of screening uptake, including age, occupational exposure to hazards, alcohol consumption, physical activity, hepatitis B infection, liver cirrhosis, and family history of liver cancer. A novel finding of our study is the significant positive association between baseline comorbidity and liver cancer screening uptake, with a dose-response relationship, which may indicate that those with comorbidities had increased healthcare utilization and risk perception, as well as more frequent opportunities for provider recommendations,23 however, these mechanisms were not directly assessed and should be regarded as hypothesis-generating. Similarly, high-risk populations with a family history of liver cancer had a significantly higher participation rate, indicating that they have a more health-oriented consciousness and greater motivation for liver cancer screening.
These findings highlight the need for tailored interventions for these subgroups to address potential disparities in liver cancer screening.
The community health centers in China shoulder the responsibility for the first half of cancer screening implementation, including program publicity, risk assessment, screening counseling, and referral. Our results offer valuable insights for shaping community-based strategies aimed at increasing participation. For the first time, participants exposed to more than two types of publicity or receiving more than three mobilization methods for screening were associated with a higher uptake rate.
Similar findings have been reported in previous studies.24 Those exposed to multimedia campaigns or multichannel outreach may gradually develop an understanding of cancer, leading to a more positive attitude toward cancer screening. Furthermore, participation rates were higher in developed regions than in developing areas, where lower levels of income, education, and health awareness may reduce the willingness to engage. Our results also indicated that in developed regions, neither multiple mobilization strategies nor financial allowances were significantly associated with participation rates.
This finding underscores the need for liver cancer screening programs to tailor their implementation strategies to regional characteristics.
Our study identified key reasons for unwillingness to participate in screening: no symptoms and do not think they would get liver cancer, afraid of being diagnosed with liver cancer, and lack of knowledge about cancer prevention and treatment, which are consistent with those reported in previous studies. A systematic review also noted that the most commonly cited barriers included insufficient knowledge and awareness of liver cancer screening, unawareness of the need for early detection, and lack of physician recommendation.13 These findings reflect generally low public awareness of cancer prevention and a poor understanding of the benefits of screening. Previous study have reported significant associations between attitudes and screening practices, whereas knowledge levels were no correlated with screening practices.25,26 Our study further identified a major psychological barrier to screening: fear of being diagnosed with liver cancer.
These findings underscore the need for clearer professional and patient education, and possibly the involvement of psychologists, to improve screening participation among high-risk populations.
The detection rate for liver cancer in our study was estimated at 0.10%, which was slightly higher than the rate observed in Zhejiang Province, China (0.06%),27 but lower than that reported from 12 rural sites in China (0.41%).20 Of noted, direct comparisons across screening programmes are limited by differences in risk selection, underlying populations, and participation rates. Nevertheless, given the lower screening participation observed in less developed regions, policy support for liver cancer screening in these areas may warrant consideration. Finally, because this study was conducted within a single municipality in Southwest China, these findings should be interpreted cautiously and confirmed in other settings.
Given the relatively low participation and detection rates in liver cancer screening, the following issues should to be addressed in future research. First, multifaceted interventions should be implemented to enhance screening uptake. Interventions targeting individual and structural characteristics play an important role in this regard.
Individual interventions, such as education, counseling, and reminders, can raise health awareness and knowledge. Structural interventions, including multimedia campaigns, multichannel outreach, improved geographical accessibility, dedicated “green channels”, and streamlined screening procedures, can facilitate access to services. Second, a multi-dimensional risk stratification model based on core risk factors should be developed, and differentiated screening protocols tailored to regional economic development levels should be established.
This approach aims to precisely align screening resources with population risk stratification and enhance the overall detection efficacy of screening programs.28
Our study has several strengths and limitations that should be considered when interpreting the results. Among its strengths, this study was the first to explore uptake across multifaceted characteristics and the diagnostic yield of liver cancer screening within a continuing large-scale population-based national cancer screening program in China. In addition, standardized procedures and rigorous quality control were adopted to ensure the quality of the research data.
Several limitations should also be acknowledged. First, the study population was recruited from 18 districts or counties in Chongqing without representative sampling, due to the challenges of rapid, large-scale recruitment in real-world settings; however, we considered the economic and geographical diversity in the selected districts or counties to partially mitigate this limitation. Second, the questionnaire-based evaluation relied on self-reported data, which may have introduced reporting bias; to address this, we provided rigorous quality control training for investigators to ensure that participants answered all questions truthfully.
Third, follow-up of screened patients is ongoing, and critical clinical outcomes, such as incidence and mortality, were not yet fully available. Thus, our findings reported preliminary short-term results, and long-term benefits in terms of incidence and mortality may be evaluated with extended follow-up. Fourth, the small number of liver cancer cases detected (n=14) limits the strength of conclusions regarding diagnostic yield and downstream clinical benefit.
Fifth, the exclusion of individuals with incomplete structural questionnaires and those with partial screening may have introduced selection bias, which could affect the generalizability of our findings. Finally, the classification of economic status was based on the 2022 GDP. Although the overall development pattern of the 18 participating districts or counties remained largely stable during the study period, individual districts or counties may have experienced differential economic growth.
Conclusion
In conclusion, this study showed that the uptake of liver cancer screening among high-risk populations in Chongqing was suboptimal (40.59%), and that participation was associated with both individual-level factors (including age, baseline comorbidities, hepatitis B infection, liver cirrhosis, family history of liver cancer, and lifestyle factors) and structural factors (including regional economic status, publicity and mobilization strategies, and financial allowances). The most frequently reported barriers to participation were the absence of symptoms, fear of diagnosis, lacked knowledge about cancer prevention. The short-term diagnostic yield of screening was 0.10%.
These findings suggest that multifaceted interventions addressing both individual and structural barriers—such as tailored education and counseling, multimedia campaigns, and improved service accessibility—may help improve screening uptake in regions with limited medical resources. However, these interventions were not directly evaluated in this observational study, and further research is needed to assess their effectiveness. Further studies with extended follow-up are warranted to evaluate the impact of screening on incidence and mortality.