Determinants of gastric cancer screening attendance in Southeastern China: a cross-sectional study

Abstract

Objectives

This study aimed to identify the determinants of gastric cancer screening attendance among individuals aged 40 years in a region with high gastric cancer in China.

Design

An anonymous, cross-sectional survey was conducted between October 2021 and March 2022.

Setting

A self-administered online survey was conducted in Fujian Province in Southeastern China.

Participants

People aged 40 years living in five selected cities in Fujian Province with no history of cancer.

Main outcome measures

Gastric cancer screening attendance was measured with the question ‘Have you ever been screened for gastric cancer in the past’.

Results

In total, 2547 complete responses were obtained. The mean age of respondents was 47.72±7.20 years, and 59.8% were men. A total of 42.6% of participants reported that they had undergone gastric cancer screening. The result of multivariable logistic regression analysis showed that participants with a first-degree relative affected with gastric cancer (OR=2.02, 95% CI: 1.58 to 2.59) and high perceived susceptibility of gastric cancer (OR=2.03, 95% CI: 1.58 to 2.59) were the strongest facilitators for screening attendance. Other factors positively associated with screening attendance were age 51–60 years (OR=1.69, 95% CI: 1.31 to 2.18), living in urban regions (OR=1.27, 95% CI: 1.05 to 1.55), friends/neighbours/colleagues with gastric cancer (OR=1.30, 95% CI: 1.07 to 1.58), history of chronic gastric disease (OR=1.90, 95% CI: 1.57 to 2.30), perceived high cost (OR=1.28, 95% CI: 1.01 to 1.61) and physician recommendation (OR=1.71, 95% CI: 1.36 to 2.16). On the other hand, factors negatively associated with screening attendance included perceived barriers, namely screening is only necessary when symptoms present (OR=0.71, 95% CI: 0.58 to 0.87) and perceived appointment for gastroscopy screening is difficult and time-consuming (OR=0.75, 95% CI: 0.60 to 0.94). No significant association was found between knowledge level and participation in screening.

Conclusion

This study highlights important individual-level factors and barriers to gastric cancer screening. Strategies targeting under-screened populations and eliminating patient-perceived barriers to gastric cancer screening are essential.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• We recruited a nearly equal number of residents from both rural and urban regions for each selected city to enhance the representativeness of residents in Fujian Province.

• The study was conducted in high-risk populations for gastric cancer; therefore, the findings would be valuable for establishing health promotion programmes specifically for the screening target population.

• The gastric cancer screening behaviour was self-report, thus, may be subjected to response biases.

• The current study evaluated a high-risk population from one province in China; therefore, the results cannot be generalised to other provinces or high-risk populations in China.

Introduction

Gastric cancer is the fifth most common cancer and has been a major global disease burden worldwide in recent years.¹ The incidence and mortality rates of gastric cancer in China account for approximately 50% of its global burden.² Despite global improvements in the clinical treatment of gastric cancer, the prognosis in China remains poor. An important reason for poor prognosis could be a delay in diagnosis, which may account for higher morbidity and mortality risks.³ More than 60% of patients with gastric cancer in China are diagnosed at an advanced stage, resulting in a poor 5-year overall survival rate of 35%–40%.⁴ In contrast, with an early diagnosis, the 5-year overall survival rate of patients with gastric cancer can exceed to 90%.^{5 6} Therefore, strategies to promote the early detection and timely treatment of gastric cancer are paramount.

Although China has a lower incidence rate of gastric cancer than South Korea and Japan, the 5-year survival rate of patients with gastric cancer is significantly higher in these countries,⁷ mainly due to delays in diagnosis in China.⁸ Better gastric cancer prognoses in Korea and Japan may largely be due to their nationwide screening programmes,^{9 10} which provide free or subsidised biennial gastric cancer screening for the general public aged >40 years. Screening programmes have proven to be one of the most effective ways to promote early diagnosis of gastric cancer and reduce mortality. A study in Korea found that, by increasing participation in screening programmes from 45.1% to 65.4%, the rate of early diagnosis of gastric cancer increased from 60.5% in 2006 to 70.6% in 2011.¹¹ A case–control study conducted in Linzhou City, China, also reported that endoscopic screening led to a 28% reduction in mortality due to gastric cancer.¹² However, there is currently no nationwide cancer screening programme in China, and organised and opportunistic screening with endoscopy is the primary method of early gastric cancer detection.^{13 14} Since 2005, two organised cancer screening programmes for gastric cancer have been implemented in selected high-risk areas of China: the Cancer Screening Programme in Rural Areas (established in 2005) and the Cancer Screening Programme in Urban China (CanSPUC, established in 2012).^{15 16} High-risk individuals aged 40 years, as defined by the Chinese national guidelines, were invited for upper gastrointestinal cancer screening.¹⁷ By the end of 2018, more than 2.16 million rural people had undergone upper gastrointestinal endoscopy at 194 programme sites, reaching a cancer detection rate of 2.05%, with 70% detected at an early stage.¹⁸ However, such organised screening programmes only benefit selected high-risk areas, and regular screening largely relies on individuals. Consequently, the screening rate in the general Chinese population remains low. A recent cross-sectional survey reported that only 15.2% of China’s general population had undergone gastric cancer screening.¹⁹ More importantly, the screening rate of high-risk populations remains unclear. Therefore, it is important to identify the factors that promote gastric cancer screening to provide insights for improving health policies that promote gastric cancer screening among the general public in China.

Adequate knowledge and awareness of gastric cancer risk factors and alarming symptoms are paramount for effective screening and early diagnosis. Various factors that increase the risk of developing gastric cancer have been identified, including Helicobacter pylori infection, older age, smoking, alcohol consumption, lifestyle and family history.²⁰ H. pylori infection has been frequently reported as a risk factor for gastric cancer, accounting for almost 90% of non-cardiac gastric cancer cases.²¹ A previous study found that awareness of the association between H. pylori infection and gastric cancer was more likely to lead a person to undergo screening and receive treatment for H. pylori infection.²² In addition, knowledge of alarming symptoms is critical for the early diagnosis of cancer. The presence of an abdominal lump, fullness and pain are symptoms of gastric cancer.²³ Knowledge of alarm symptoms can lead to earlier presentation to medical care, resulting in earlier diagnosis and better outcomes.²⁴

In addition, individual decision-making regarding cancer screening is affected by personal attitudes and perspectives. The Health Belief Model (HBM) explains and predicts health behaviours by focusing on individual attitudes and beliefs. These include perceived susceptibility, severity, benefits, barriers, cues to action and self-efficacy. The HBM model has been used extensively to study health behaviours, including gastrointestinal cancer screening.^{25 26} HBM constructs have been recognised as important predictors of cancer screening uptake in many previous studies, including a study on Chinese colonoscopy compliance in a community-based colorectal cancer screening programme.²⁵

However, population-level gastric cancer screening studies in China are limited. We recently conducted a study exploring gastric cancer screening in a high-risk population (aged ≥40 years and living in a high-risk region) in Southeastern China, which revealed a high (80%) proportion of intention to be screened for gastric cancer.²⁷ However, an intention–action gap remains that warrants further investigation. Screening behaviour for gastric cancer has never been documented in this high-risk population. Therefore, this study reports the factors influencing gastric cancer screening uptake in this sample. Additionally, we explored the potential influence of individual-level, knowledge-based and health-belief factors on screening behaviours. Fujian Province in Southeastern China is a high-risk region for gastric cancer. Several cities in Fujian Province reported a twofold higher mortality rate than the national average (49.47/100 000 vs 21.9/100 000).²⁸ In view of the target screening population defined by the Chinese national guidelines,¹⁷ the sample comprised individuals aged >40 years in Fujian Province. Understanding the factors contributing to gastric cancer screening participation will provide information that will inform future gastric cancer screening programmes and reduce mortality due to gastric cancer through early detection and treatment.

Methods

Study design and participants

We conducted a cross-sectional web-based anonymous survey using an online questionnaire from October 2021 to March 2022. Convenience sampling was used to recruit participants for this study, using WeChat (the most popular social media platform in China) to advertise and circulate survey links to network members. Network members were asked to distribute survey invitations to all contacts who satisfied the inclusion criteria.

The target population of this study was people at high risk for gastric cancer, defined as those aged ≥40 and living in high-risk regions of the Fujian Province. The inclusion criteria were as follows: (1) age ≥40 years; (2) living in Fuzhou, Putian, Quanzhou, Xiamen and Zhangzhou cities of Fujian Province; and (3) no history of cancer. These five cities were selected because they have the highest incidence of gastric cancer in Fujian Province. The cumulative population of these five cities accounts for 73.43% of the total population of Fujian Province.²⁹

On completion of the survey, each respondent who provided a valid questionnaire was awarded an incentive of five Chinese Yuan (equivalent to US$0.75). To achieve more comprehensive recipient coverage, we also encouraged participants to disseminate the survey link to all their contacts with a thank you note. Participants were informed that their participation was voluntary, and consent was obtained through the completion of the questionnaire.

Instrument

The questionnaire was developed by a research team consisting of academics, epidemiologists and clinicians. The questionnaire was face-validated by a panel of experts and pilot-tested prior to administration. The survey questionnaire was divided into five sections: (1) demographic characteristics and general health status, (2) knowledge of gastric cancer-related risk factors and symptoms, (3) treatment-seeking history, (4) attitudes based on HBM constructs and (5) actual screening behaviour and intention towards gastric cancer screening. Online supplemental appendix 1 provides the full version of the questionnaire.

Demographic and general health

Demographic characteristics, namely age, sex, height, weight, highest educational achievement, marital status, current residential location (urban/rural), current residential city, type of occupation and average monthly household income, were obtained. Participants were asked if they had any first-degree relatives or anyone in their social network (eg, friends, neighbours and colleagues) diagnosed with gastric cancer. A history of mild chronic gastric disease (eg, chronic gastritis and gastric ulcers) was also recorded. Questions on general health assessed participants’ current self-rated health status.

Knowledge

This section contained a series of questions assessing the participants’ knowledge of risk factors (24-item scale) and symptoms of gastric cancer (14-item scale).²⁷ The response options were ‘true’, ‘false’ or ‘do not know’. Correct responses were scored as 1. An incorrect or ‘do not know’ response was scored as 0. The total possible knowledge score ranges from 0 to 38, with higher scores representing higher levels of knowledge. Participants were divided into high-knowledge and low-knowledge groups based on the median value of their total knowledge scores.

Attitudes

The HBM was used to explore five dimensions of health beliefs: perceived susceptibility to gastric cancer (three items),³⁰ perceived severity of gastric cancer (three items), perceived benefits of gastric cancer screening (two items), perceived barriers to the uptake of gastric cancer screening (five items) and cues for action (three items). The response options were ‘strongly agree, agree, disagree and strongly disagree’.

Sample size calculation

The minimal required sample size was calculated based on the formula ${\displaystyle N=[{\mu }_{\alpha /2}^2\times \pi \times (1-\pi )]/{\delta }^2}$. The prevalence rate was 15% (π) based on the gastric cancer screening rate reported in a previous study,¹⁹ with a significance level set at 0.05 (α) and an allowable error of 0.03 (δ). The minimum estimated sample size is 544.

Data auditing

All the submitted survey data were audited by the research team. A pilot survey showed that respondents needed at least 200 seconds to complete the questionnaire. Therefore, questionnaires that took less than 200 seconds to complete were excluded. Pitfall questions were used to identify the invalid responses. Respondents were asked to indicate their current living location (‘urban’, ‘suburban’ or ‘rural’) twice: at the beginning and in the middle of the questionnaire. Questionnaires with inconsistent responses to these two questions were excluded.

Statistical analyses

Demographic characteristics and responses to each question are described using frequencies and percentages. Univariate and multivariate logistic regression analyses were performed to determine factors influencing gastric cancer screening. All factors found to be statistically significant (p value <0.05) in the univariate regression analysis were entered into a multivariate logistic regression model using the simultaneous forced entry method. ORs, 95% CIs and p values were calculated for each independent variable. The model fit of the multivariable logistic regression analysis was assessed using the Hosmer-Lemeshow goodness-of-fit test.³¹ All p values were based on a two-sided test with a statistical test level of α set at 0.05. All statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS V.26.0, IBM Corp, Armonk, New York).

Ethics and permission for data collection

This study was conducted in accordance with the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Fujian Medical University (FJMU No. 2020 [53]). A statement regarding informed consent was also provided. The respondents provided their consent to participate in the study by clicking on either the ‘Yes’ or ‘No’ option.

Patient and public involvement

None of the participants was involved in designing or developing the research questions or outcome measures. The participants were encouraged to disseminate the survey link to all their contacts.

Results

Demographics

Between October 2021 and March 2022, a total of 3525 complete surveys were received, of which 844 were deleted because the completion time was less than 200 seconds, and 224 were deleted because of inconsistent responses to the same two questions about their current living location. A total of 2547 responses were included in the final analysis. Online supplemental table 1 shows the demographics of the study participants compared with those of the general adult population in the five cities studied. The demographic characteristics of the participants are presented in table 1. The mean age of the study participants was 47.72 years (±7.20). Approximately half lived in urban areas (56.9%) and had an average monthly income below 5000 RMB (US$750) (48.0%). An almost equal proportion of the participants had the highest educational level of secondary school and below (33.3%), high school/technical school (35.1%) or university or above (31.6%). Only 18.8% of the respondents reported having a family history of gastric cancer, while 40.6% knew someone in their social network who had gastric cancer. In total, 40.0% of participants reported a history of mild chronic gastric disease. The 38 knowledge score items had a reliability (Cronbach’s α) of 0.954. The mean and SD for the total knowledge score was 23.9 (SD±9.8) out of a possible score of 38. The median score was 25 (IQR: 17–33). Knowledge scores were categorised as high or low based on the median split; therefore, 1209 (47.5%) respondents were categorised as having a high score, (25–38) and 1338 (52.5%) had a low score (0–24).

Table 1.

Demographic characteristics of respondents (N=2547)

Characteristics	No	%
Age, mean±SD	47.72±7.20
Age groups
40–50	1991	78.2
51–60	408	16.0
>60	148	5.8
Body mass index (kg/m2)
<18.5	175	6.9
18.5–24.9	1882	73.9
≥25.0	490	19.2
Sex
Male	1522	59.8
Female	1025	40.2
Educational Level
Secondary school and below	849	33.3
High school/technical school	894	35.1
University and above	804	31.6
Monthly average income (RMB)*
<2000	291	11.4
2000–5000	1034	40.6
>5000	1222	48.0
Current residing location
Urban	1448	56.9
Rural	1099	43.1
Current residing region
Fuzhou city	699	27.4
Putian city	425	16.7
Xiamen city	559	21.9
Zhangzhou city	430	16.9
Quanzhou city	434	17.1
Occupation
Professional and managerial	632	24.8
Office worker/service personnel	432	17.0
Industrial worker/farmers/others	770	30.2
Individual business/self-employed	443	17.4
Housewife/retired/unemployed	270	10.6
Knowledge of risk factors and warning symptoms
Total score, mean±SD	23.9±9.8
Median (IQR)	25 (17–33)
Low score (0–24)	1209	47.5
High score (25–38)	1338	52.5

*1 RMB=US$0.15.

Participation in screening and its influencing factors

A total of 42.6% of participants reported that they had been screened for gastric cancer. Table 2 shows the significant findings of univariate and multivariate logistic regression analyses of the factors influencing screening behaviour. The multivariable logistic regression showed that gastric cancer history in first-degree relatives (OR=2.02, 95% CI: 1.58 to 2.59) and personally perceived susceptibility to gastric cancer (OR=2.03, 95% CI: 1.58 to 2.59) were the strongest factors influencing participation in gastric cancer screening. Other significant factors influencing participation in gastric cancer screening were age (51–60 compared with 40–50 years (OR=1.69, 95% CI: 1.31 to 2.18)), living in urban areas (OR=1.27, 95% CI: 1.05 to 1.55), knowing people in their social network diagnosed with gastric cancer (OR=1.30, 95% CI: 1.07 to 1.58), having a history of mild chronic gastric disease (OR=1.90, 95% CI: 1.57 to 2.30), perceived high screening cost (OR=1.28, 95% CI: 1.01 to 1.61) and physician recommendation (OR=1.71, 95% CI: 1.36 to 2.16). Factors associated with non-participation in gastric cancer screening were living in Zhangzhou city (OR=0.64, 95% CI: 0.47 to 0.87), the perception that screening is only necessary when symptoms are present (OR=0.71, 95% CI: 0.58 to 0.87) and appointment for gastroscopy screening perceived as difficult and time-consuming (OR=0.75, 95% CI: 0.60 to 0.94). No clear trends were observed between participation in gastric cancer screening and educational attainment. The results of the univariate and multivariate logistic regression analyses are shown in online supplemental table 2.

Table 2.

Factors associated with participation of gastric cancer screening (N=2547)

	Frequency (%)	Univariate analysis		Unadjusted OR (95% CI)	P value	Multivariable logistic regression*	P value
		Ever screening or not				Yes vs no
		Yes n=1086	No n=1461			OR (95% CI)
Baseline demographic
Age group (years old)
40–50	1991 (78.2)	809 (40.6)	1182 (59.4)	Reference	<0.001	Reference
51–60	408 (16.0)	211 (51.7)	197 (48.3)	1.57 (1.26 to 1.94)		1.69 (1.31 to 2.18)	<0.001
>60	148 (5.8)	66 (44.6)	82 (55.4)	1.17 (0.84 to 1.65)		1.35 (0.88 to 2.06)	0.170
Highest education level
Primary school and below	283 (11.1)	121 (42.8)	162 (57.2)	Reference	0.007	Reference
Secondary school	566 (22.2)	209 (36.9)	357 (63.1)	0.78 (0.59 to 1.05)		0.61 (0.44 to 0.85)	0.004
High school/technical school	894 (35.1)	413 (46.2)	481 (53.8)	1.15 (0.88 to 1.51)		0.71 (0.50 to 0.99)	0.043
University and above	804 (31.6)	343 (42.7)	461 (57.3)	0.98 (0.76 to 1.31)		0.59 (0.41 to 0.87)	0.007
Current location
Urban	1448 (56.9)	675 (46.6)	773 (53.4)	1.46 (1.25 to 1.72)	<0.001	1.27 (1.05 to 1.55)	0.016
Rural	1099 (43.1)	411 (37.4)	688 (62.6)	Reference		Reference
Current residing region
Fuzhou city	699 (27.4)	323 (46.2)	376 (53.8)	1.12 (0.88 to 1.43)	<0.001	0.98 (0.75 to 1.29)	0.895
Putian city	425 (16.7)	208 (48.9)	217 (51.1)	1.25 (0.96 to 1.64)		1.34 (0.98 to 1.83)	0.065
Xiamen city	559 (21.9)	240 (42.9)	319 (57.1)	0.98 (0.76 to 1.27)		0.83 (0.62 to 1.10)	0.185
Zhangzhou city	430 (16.9)	127 (29.5)	303 (70.5)	0.55 (0.41 to 0.73)		0.64 (0.47 to 0.87)	0.005
Quanzhou city	434 (17.0)	188 (43.3)	246 (56.7)	Reference		Reference
Experience with gastric cancer
Ever known any first-degree relatives had gastric cancer
Yes	479 (18.8)	313 (65.3)	166 (34.7)	3.16 (2.57 to 3.89)	<0.001	2.02 (1.58 to 2.59)	<0.001
No	2068 (81.2)	773 (37.4)	1295 (62.6)	Reference		Reference
Ever known any friends, neighbours, colleagues had gastric cancer
Yes	1033 (40.6)	537 (35.5)	977 (64.5)	2.06 (1.76 to 2.43)	<0.001	1.30 (1.07 to 1.58)	0.008
No	1514 (59.4)	549 (53.1)	484 (46.9)	Reference		Reference
Health characteristics
Ever diagnosed with mild chronic gastric diseases (eg, chronic gastritis, gastric ulcers, etc)
Yes	1020 (40.0)	585 (57.4)	435 (42.6)	2.75 (2.34 to 3.24)	<0.001	1.90 (1.57 to 2.30)	<0.001
No	1527 (60.0)	501 (32.8)	1026 (67.2)	Reference		Reference
Health beliefs
Perceived susceptibility
I may have gastric cancer
Strongly agree/agree	967 (38.0)	559 (57.8)	408 (42.2)	2.74 (2.32 to 3.23)	<0.001	2.03 (1.66 to 2.48)	<0.001
Disagree/strongly disagree	1580 (62.0)	527 (33.4)	1053 (66.6)	Reference		Reference
Perceived barriers
Screening is only necessary when symptoms present
Strongly agree/agree	1442 (56.6)	550 (38.1)	892 (61.9)	0.66 (0.56 to 0.77)	<0.001	0.71 (0.58 to 0.87)	<0.001
Disagree/strongly disagree	1105 (43.4)	536 (48.5)	569 (51.5)	Reference		Reference
Cost of gastroscopy is very high
Strongly agree/agree	1588 (62.3)	648 (40.8)	940 (59.2)	0.82 (0.70 to 0.96)	0.016	1.28 (1.01 to 1.61)	0.039
Disagree/strongly disagree	959 (37.7)	438 (45.7)	521 (54.3)	Reference		Reference
It is difficult and time-consuming to have an appointment for gastroscopy screening
Strongly agree/agree	1692 (66.4)	683 (40.4)	1009 (59.6)	0.76 (0.64 to 0.90)	0.001	0.75 (0.60 to 0.94)	0.011
Disagree/strongly disagree	855 (33.6)	403 (47.1)	452 (52.9)	Reference		Reference
Cues-to-action
I only take screening when doctor recommends
Strongly agree/agree	1804 (70.8)	694 (38.5)	1110 (61.5)	Reference	<0.001	Reference
Disagree/strongly disagree	743 (29.2)	392 (52.8)	351 (47.2)	1.79 (1.50 to 2.12)		1.71 (1.36 to 2.16)	<0.001

*Hosmer and Lemeshow test, χ²: 488.216; p value: <0.001; Nagelkerke R²: 0.234.

Discussion

Many countries recommend regular gastric cancer screening for early diagnosis in individuals aged 40 years. As few studies have focused on this population, this study examined their participation in gastric cancer screening in China and contributed to the current literature on factors related to screening attendance.

Our study found a low gastric cancer screening rate in a sample of people aged 40 years and older in Southeastern China, which was much lower than that reported in Korea.³² The differences can largely be explained by the fact that Korea has established a nationwide screening programme. Screening programmes have proven to be one of the most effective ways to promote early diagnosis of gastric cancer. Owing to its large population and limited medical resources, such nationwide screening programmes may not be feasible in China in the near future. Although two organised cancer screening programmes have been implemented since 2005, including screening for gastric cancer, only selected high-risk areas have been covered.^{15 16} The most crucial step for China in the near future is to increase the screening uptake in asymptomatic individuals in high-risk populations.

It should be noted that most of our respondents were aged between 40 and 50 years (78.2%). Interpretation of results should be done with caution as it might not represent all of the potential target population for gastric cancer screening. The Cancer Screening Programme in Rural Areas includes high-risk individuals aged 40–69 years from selected rural areas of China. Their previous results showed that among 113 340 residents aged 40–69 years screened by endoscopy in two of the selected provinces, 46.27% were aged 40–49 years, 33.49% were aged 50–59 years and 20.24% were aged 60–69 years.³³ On the other hand, the CanSPUC targeted high-risk individuals aged 40–74 years in selected urban cities. Baseline results showed that individuals aged >50 years were the main population screened for gastric cancer (74.6%).³⁴ Meanwhile, under the CanSPUC schema, colorectal and breast cancer screening programmes also reported similar age distributions.^34–36 These findings imply that high-risk individuals aged >50 years are an important target population for cancer screening programmes. Efforts should be made to recruit more elderly people for future studies on cancer screening intentions.

In this study, more than half of the participants screened for gastric cancer had a history of mild chronic gastric disease. Most patients undergo endoscopy for gastrointestinal symptoms. The results of multivariate logistic regression also revealed that participants with a history of mild chronic gastric disease had a twofold higher participation rate. These findings imply that the general public aged 40 years and older should be encouraged to undergo gastric cancer screening despite the absence of gastric disease symptoms. Therefore, governments should introduce policies to enhance screening rates in asymptomatic target populations.

First-degree relatives are well-recognised as one of the factors influencing the early diagnosis of gastric cancer.³⁷ As found in this study, the results of multivariate logistic regression analysis revealed that in addition to a history of gastric disease, demographics and knowledge of family and social acquaintances with gastric cancer promote a person’s screening behaviour. Furthermore, regional differences in screening behaviours were observed in our study, whereby urban residents were more likely to participate in gastric cancer screening than those living in rural areas, which is in line with the findings of many other Asian studies.³⁸ Conceivably, urban regions have better access to diagnostic services than rural regions; therefore, the risk of delayed gastric cancer diagnosis is usually higher in rural patients. Similarly, a study in South Korea found that individuals living in remote rural areas had a significantly higher likelihood of delayed gastric cancer diagnosis.³⁸ A study conducted in China also reported that mountainous rural areas had more advanced cases of gastric cancer than urban areas.³⁹ As rurality is associated with poor gastric cancer screening uptake, health authorities should improve access to screening and increase health promotion to enhance screening behaviour in rural communities.

As hypothesised, the HBM constructs influenced health behaviours, including cancer screening. First, among all HBM components studied, perceived susceptibility to gastric cancer was found to be the most important facilitator of gastric cancer screening. A lack of recognition of cancer risk has been suggested as a barrier to participation in colorectal cancer screening.⁴⁰ Improvement of disease susceptibility information access is therefore important to consider when designing future gastric cancer awareness campaigns and screening promotions. Second, physician recommendations were the only significant factor that increased screening uptake among all the cues to action. Previous studies have reported that physician advice is a strong motivator for colorectal cancer screening participation.^{41 42} A survey conducted in the USA reported that clinician recommendations have the most important influence on up-to-date colorectal cancer screening in patients of all ages.⁴² Strengthening the role of physicians by enhancing interventions, such as training programmes, is necessary to encourage and equip physicians to counsel their patients regarding gastric cancer screening.⁴³ Importantly, our results revealed two prominent barriers to gastric cancer screening. There appears to be an erroneous belief that screening is only required if signs of the disease are present. Abundant evidence has shown that most patients with gastric cancer undergo endoscopic screening for gastric cancer after the onset of gastrointestinal symptoms, most of whom are diagnosed at an advanced stage.⁴⁴ Therefore, increasing screening awareness among high-risk, asymptomatic populations is crucial. The second obstacle to timely screening is related to the perceived difficulty in accessing screening services and long waiting times for endoscopy. In China, the average outpatient endoscopy waiting times under anaesthesia are approximately 5.9 days for ordinary gastroscopy and 8.1 days for gastroscopy, respectively.⁴⁵ It is well recognised that long waiting times may lead to reduced patient satisfaction and decreased willingness to seek preventive healthcare services.⁴⁶ It is time for the country to reevaluate wait times for patients referred for gastric cancer diagnosis and, eventually, treatment for gastric cancer.

Various studies have consistently reported low knowledge about gastric cancer as a barrier to timely screening and treatment.^{47 48} Our previous study found that a high level of knowledge about gastric cancer and screening was associated with a stronger intention to undergo screening.²⁷ However, the current study did not find a significant association between knowledge levels and screening participation. This study found that gastric cancer screening behaviours were significantly influenced by cancer history in first-degree relatives and social acquaintances, personally perceived susceptibility to gastric cancer, and physician recommendations. As gastric cancer-related knowledge is not associated with favourable screening behaviours, imparting appropriate gastric cancer knowledge to the population may not be sufficient to improve actual screening behaviours. Hence, comprehensive health promotion programmes should address the factors identified in this study.

Importantly, this study found that the high perceived cost of gastric cancer screening was not a barrier to screening in this sample. Further studies are needed to confirm this finding through a larger sample or qualitative enquiry to probe the perceived financial barriers to gastric cancer screening. It is possible that with the improvement in economic status in China, people consider health to be more important and are willing to dedicate more money to it. Finally, this study provides evidence of an intention–action gap in gastric cancer screening in China. Despite the high intentions reported in our earlier publication,²⁷ the sample population showed low screening behaviour. The facilitators of gastric cancer screening identified in this study may be useful in closing current intention–action gaps.

Limitations

Because this was a cross-sectional study, the main limitation was that the temporal link between gastric cancer screening behaviour and the investigated factors could not be determined. Second, because an online survey was used, the exact number of people who received the questionnaire was unavailable; therefore, the response rate could not be calculated. Third, gastric cancer screening behaviour was self-reported; thus, it may be subject to response bias. Further studies with larger sample sizes are required to validate our findings. Fourth, the current study evaluated a high-risk population from one province in China; therefore, the results cannot be generalised to other provinces or high-risk populations in China. Fifth, owing to the use of an online survey, most respondents were 40–60 years old and younger than the potential screening target population in Fujian Province. Future studies should recruit more elderly individuals, especially those older than 60 years. Finally, the potential harm of gastric cancer screening was not included in the survey, which might have led survey respondents to overestimate the benefits of gastric cancer screening.

Conclusions

In conclusion, to enhance gastric cancer screening behaviours in high-risk populations in China, the present study demonstrates the importance of overcoming various individual-level factors, personal subjective perceptions of susceptibility and attitudinal and structural barriers. This study also highlights the fact that imparting appropriate gastric cancer knowledge to the population is appropriate but may not be sufficient to improve screening behaviours. An important highlight of this finding is that regional disparities in the uptake of gastric cancer screening found in this study imply that screening recommendations should target underscreened regions, particularly those enabling equal access to gastric screening regardless of rurality. More importantly, this study also provides evidence of an intention–action gap and shows that the facilitators of gastric cancer screening identified in this study are extremely useful for an action plan to close this discrepancy.

Data availability statement

Data are available upon reasonable request. To request access to the underlying research data, please contact Dr. Yulan Lin: yulanlin@fjmu.edu.cn

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Fujian Medical University (FJMU No. 2020 [53]).