key: cord-0076937-t0rw0t52
authors: Wei, Wenshuang; Zhang, Miao; Zuo, Dan; Li, Qinmei; Zhang, Min; Chen, Xinguang; Yu, Bin; Liu, Qing
title: Screening Intention Prediction of Colorectal Cancer among Urban Chinese Based on the Protection Motivation Theory
date: 2022-04-01
journal: Int J Environ Res Public Health
DOI: 10.3390/ijerph19074203
sha: 8fdb33fbb02e47861a85544ef6e147126c4ace6f
doc_id: 76937
cord_uid: t0rw0t52

Colorectal cancer poses a serious threat worldwide. Although early screening has been proved to be the most effective way to prevent and control colorectal cancer, the current situation of colorectal cancer screening remains not optimistic. The aim of this article is to apply the protection motivation theory (PMT) to examine the influencing factors on screening intention of colorectal cancer (CRC). This cross-sectional survey was launched in five communities in Wuhan, China. All the eligible urban Chinese were recruited and interviewed using paper-and-pencil questionnaires. The intention of colorectal cancer screening (CRCS) was measured using six PMT subconstructs, including perceived risk, perceived severity, fear arousal, response efficacy, response cost, and self-efficacy. Data on sociodemographic variables and knowledge of CRC were also collected. The structural equation modeling (SEM) method was used for data analysis. Among all the 569 respondents, 83.66% expressed willingness to participate in CRCS. Data of the research fit the proposed SEM model well (Chi-square/df = 2.04, GFI = 0.93, AGFI = 0.91, CFI = 0.91, IFI = 0.91, RMSEA = 0.04). Two subconstructs of PMT (response efficacy and self-efficacy) and CRC knowledge were directly and positively associated with screening intention. Age, social status, medical history, physical activity, and CRC knowledge were indirectly related to the screening intention through at least one of the two PMT subconstructs (response efficacy and self-efficacy). The findings of this study suggest the significance of enhancing response efficacy and self-efficacy in motivating urban Chinese adults to participate in CRC screening. Knowledge of CRC is significantly associated with screening intention. This study can provide useful information for the formulation and improvement of colorectal cancer screening strategies and plans.

In 2021, the World Health Organization released new data: Cancer is a leading cause of death worldwide and is expected to remain a major public health problem in the coming years. And colorectal cancer (CRC) is a major cause of cancer burden among both women and men, ranking third in morbidity and second in mortality around the world, causing 1.93 million new cases and almost 916 thousand deaths [1]. Studies from China have shown that the number of CRC has been increasing steadily over the past three decades, from 104.3 thousand cases in 1990 to 555 thousand cases in 2020 and from 81.1 thousand deaths in 1990 to 286 thousand deaths in 2020 [2, 3] . Obviously, colorectal cancer is one of the biggest threats to human health, leading to huge social and economic burdens accompanied with a decline in quality of life.

In this study, we aimed to: (1) assess knowledge of CRC and its screening among urban Chinese; (2) identify the possible influencing factors on CRC screening intention of urban Chinese; and (3) investigate the relationship between PMT and colorectal cancer screening intention. This study can give powerful PMT-based evidence to the local government and Health Committee for the development and improvement of colorectal cancer screening strategies among urban Chinese.

This cross-sectional study was conducted within the framework of a cancer screening program in urban China (CanSPUC). As a national cancer screening program, CanSPUC has been in progress in urban China since it was first launched in 2012. The objective of CanSPUC is to screen five main types of cancer: lung cancer free, upper gastrointestinal cancer, breast cancer, colorectal cancer, and liver cancer. Detailed information regarding the methodology of CanSPUC has been described in previous studies [24, 25] .

Briefly, urban residents aged 40 to 74 first received epidemiological questionnaires and high-risk assessments of the five types of cancer. High-risk assessment was based on the Harvard Risk Index and was appropriately adjusted to the characteristics of Chinese [24, 26] . Then individuals at high risk for those five types of cancer were screened free of charge. This study consisted of a survey on colorectal cancer screening intention as well as a screening of a high-risk population at baseline. Participants were recruited from five selected communities via convenience sampling method. WeChat, a widely used social application in China, was employed to call on local residents to participate in this survey. Specifically, the staff of the Community Health Center created separated WeChat groups, briefly introduced the purpose of CanSPUC and this study, and encouraged the groups to participate in the study. Thus, eligible participates from the five selected communities were identified and recruited following the Institutional Review Board (IRB) approved protocol.

The inclusion criteria for participation were: (a) 40-74 years old, (b) no clinical diagnosis of CRC, including precancerous lesions, (c) willing to participate in the study and having good understanding power and ability of expression, and (d) no severe organ dysfunction. Residents aged 40-74 years old were selected because of the high CRC risk during this age range [27, 28] . Those residents who were diagnosed with CRC were excluded. A total of 598 questionnaires were gathered. Among these questionnaires, 29 with missing data on key demographic variables and CRC knowledge were excluded, yielding a final sample of 569.

The data collection began in September 2020 and lasted for a month via in-person interviews, utilizing paper-and-pencil questionnaires. The questionnaire for data collection was pilot-tested to ensure feasibility and acceptability. Questionnaires were composed of demographic and other factors, knowledge of CRC, and PMT-based scale of CRC. Demographic and other factors were described in detail in the next part, and items about knowledge of CRC and the PMT-based scale of CRC were shown in Appendix A. It took 15-20 min for a typical respondent to complete the survey. Data collectors were second-year public health graduate students from Wuhan University. They received a three-day training session before participating in the data collection. The trained data collectors went to the Community Health Service Centers of the five selected communities to implement the data collection. Completed questionnaires were checked for completeness and accuracy, and double-entry and routine logic checks were performed to detect errors. Disagreements during the data entry were settled by reviewing the original questionnaire.

Intention for CRC screening was measured based on participants responses to the following question: "Are you willing to take up CRC screening in the future?" A 5-point Likert-scale was used, with 1 = "very much unwilling" and 5 = "very much willing." This variable was also dichotomized for analysis: Participants were coded as having no intention if they responded 3 (neutral) or less; otherwise, they were coded as having intention.

With reference to the previous study [24, 29] , the CRC Knowledge Scale consisting of 21 items with three subconstructs was developed. (1) the knowledge of risk factors contained 9 items, and a typical item was: "Do you believe the following items are risk factors of CRC or not?" (2) The knowledge of symptoms contained 8 items, and a typical item was: "To your knowledge, which of the following are CRC symptoms?" (3) The screening knowledge contained 4 items, and a typical item was: "Screening is needed for people with bloody stool or diarrhea alternating with constipation." Participants received one score for a correct answer, with higher scores indicating more knowledge.

Combined with the previous research [30] , which was a pilot study and expert consultation method, we designed a PMT-based Scale of CRC and revised it extensively to ensure good reliability and validity. Reliability and validity tests were conducted in the previous article [31] . In detail, the Cronbach's α for the overall scale was 0.763 and ranged from 0.584 to 0.771 for each dimension. The discrimination validity of the scale was examined via correlation coefficients: The correlation coefficients between items within the six dimensions ranged from 0.089 to 0.534, and the correlation coefficients between the dimensions and the total scale ranged from 0.165 to 0.586. The indicators of confirmatory factor analysis were better than the standard values, indicating a good construction validity of the scale.

The PMT-based scale of CRC is composed of two PMT pathways: the threat appraisal and the coping appraisal. The threat appraisal consisted of 3 subconstructs: risk perception (2 items), severity perception (2 items), and fear arousal (2 items). The risk perception referred to one's assessment of the likelihood of suffering from CRC (Cronbach alpha = 0.584); the severity perception referred to fear of the consequences from CRC (Cronbach alpha = 0.722); and the fear arousal referred to the assessment of worry of being affected by CRC (Cronbach alpha = 0.614). All items were assessed using a 5-point Likert-scale from 1 = "absolutely not" to 5 = absolutely yes.

The coping appraisal consisted of another 3 subconstructs: response efficacy (3 items), response cost (3 items), and self-efficacy (4 items). Response efficacy referred to the judgement about the role of planned colorectal cancer screening in its early detection and the preferable following treatment (Cronbach alpha = 0.737). Response cost referred to the likely economic or social costs of curing or improving CRC (Cronbach alpha = 0.712). Self-efficacy indicated whether participants have enough confidence to adopt regular CRC screening (Cronbach alpha = 0.771). All items were assessed with a 5-point Likert-scale from "1 = strongly disagree" to "5 = strongly agree."

Demographic variables were: age (in years), occupation (administrative/technician, trader/service staff, peasant, housework/no work), education (primary or below, secondary, college or above), social status (created based on occupation and education), regular physical activity (yes/no), lower digestive tract lesions (yes/no), family history of cancer (yes/no). Medical history was formed by the combination of lower digestive tract lesions and family history of cancer. Regular physical activity referred to more than 3 times a week for more than 30 min each time.

Descriptive statistics (i.e., mean, standard deviation (SD), proportion, and rate) were used to describe the study sample and the distribution of key variables. The chi-square test and Student's t-test were performed to compare differences in factors between respondents with and without screening intention. Structural equation modeling (SEM) analysis was used to examine PMT-based models linking influential factors with screening intention. Data-model fit was assessed using multiple criteria, including GFI (>0.90), AGFI (>0.90), CFI (>0.90), IFI (>0.90), RMSEA (<0.08), and chi-square/df ratio (<3.0). p < 0.05 (two-sided) was used for statistical inference. Statistical analyses were conducted using IBM Statistical Package for the Social Sciences (SPSS) version 25.0 and Analysis of Moment Structures (Amos) version 21.0 for Windows (Chicago, IL, USA). Table 1 summarizes the basic characteristics for the study sample (n = 569). The average age of the total study population was 59.05 (SD = 8.54) years old. The male to female ratio was 1:1.03. Of the total sample, 10.02% were farmers and 86.11% had secondary or more education. Approximately 70% (69.07%) of the total population reported having regular physical activity. Among all the subjects, those with lower digestive tract lesions and a family history of cancer accounted for 12.65% and 38.31%, respectively. 83.66% of all participates expressed willingness to participate in CRCS. There were no significant differences on screening intention among different characteristics groups (p > 0.05). Table 2 exhibits the knowledge scores of CRC, containing risk factors, symptoms, and screening. The results from the Student's t-test revealed that people with screening intention had higher scores about CRC knowledge compared with those participates without intention (mean = 14.16, SD = 4.56 vs. Mean = 12.43, SD = 5.67, p = 0.001). Moreover, the comparisons of each subscale between the two groups (with and without screening intention) showed that those respondents who had screening intention got higher scores than those without screening intention. The typical items regarding knowledge of CRC risk factors and symptoms were "smoking frequently" (62.18% vs. 44.09%, p = 0.001) and "change in bowel habits" (77.10% vs. 59.14%, p < 0.001). Table 3 presents the results from Pearson correlation analysis. Screening intention X1 was significantly related to X8, the CRC knowledge score (r = 0.24, p < 0.01), and five PMT subconstructs, particularly X7, the subconstruct of self-efficacy (r = 0.52, p < 0.01). In addition, CRC knowledge was significantly associated with four of the six PMT subconstructs except risk perception and severity perception. Three subconstructs of PMT (response efficacy X5, response cost X6, and self-efficacy X7) were all significantly associated with subdomains of the CRC knowledge scale (knowledge of CRC risk factors X9, knowledge of CRC symptoms X10, knowledge of CRC screening X11). 

All data fit the proposed structural equation models quite well, with GFI = 0.93, AGFI = 0.91, CFI = 0.91, IFI = 0.91, RMSEA = 0.04, and chi−square/df = 2.04. Results from the modeling are presented in Figure 1 . First, of the six PMT constructs, two (selfefficacy and response efficacy) were significantly associated with CRC screening intention. The self-efficacy measures the perceived ability to take part in CRC screening, while the response efficacy measures the perceived effectiveness of screening in CRC prevention and treatment. Table 3 presents the results from Pearson correlation analysis. Screening intention was significantly related to X8, the CRC knowledge score (r = 0.24, p < 0.01), and five PM subconstructs, particularly X7, the subconstruct of self-efficacy (r = 0.52, p < 0.01). In ad tion, CRC knowledge was significantly associated with four of the six PMT subconstru except risk perception and severity perception. Three subconstructs of PMT (response ficacy X5, response cost X6, and self-efficacy X7) were all significantly associated w subdomains of the CRC knowledge scale (knowledge of CRC risk factors X9, knowled of CRC symptoms X10, knowledge of CRC screening X11). 

All data fit the proposed structural equation models quite well, with GFI = 0.93, AG = 0.91, CFI = 0.91, IFI = 0.91, RMSEA = 0.04, and chi−square/df = 2.04. Results from modeling are presented in Figure 1 . First, of the six PMT constructs, two (self-efficacy a response efficacy) were significantly associated with CRC screening intention. The se efficacy measures the perceived ability to take part in CRC screening, while the respon efficacy measures the perceived effectiveness of screening in CRC prevention and tre ment. All the four external factors (social status, medical history, physical activity, and CRC knowledge) were significantly associated with the PMT subconstruct of self-efficacy, suggesting the role of these factors in promoting CRC screening through self-efficacy. In addition, age was positively associated with self-efficacy and CRC knowledge was directly and positively associated with the intention for CRC screening.

Of the four external factors, three (social status, medical history, and CRC knowledge) were associated with response efficacy, another PMT subconstruct that was significantly associated with CRC screening intention. In other words, these factors can promote screening intention by increasing the perceived effectiveness of screening for CRC prevention and treatment.

In this current study, a multivariate approach-SEM was utilized to explore relationships between colorectal cancer screening intention and the two pathways of PMT (threat appraisal process and coping appraisal process). To our knowledge, this is the first study to assess colorectal cancer screening intention by employing theoretical framework PMT among urban Chinese. The findings of this study enable us to add new evidence to advance colorectal cancer research around the world, there being currently millions of colorectal cancer patients. The findings can strengthen the theoretical basis of the underlying mechanisms regarding urban Chinese screening intention and behaviors. All these data can promote the development of behavioral theory-based programs and increase CRC screening rates.

Results from SEM indicated the good fit of the whole model, which meant that PMT was a good tool to explore the relationships between sociodemographic factors and screening intention. Consistent with the findings of other researchers, not all constructs of PMT were equally strong in predicting health-related behaviors [21, 32, 33] . Data from this study indicated that the two PMT variables (response efficacy and self-efficacy) had significant associations with CRC screening intention, demonstrating a stronger impact on the perception of coping appraisal than on the perception of threat appraisal [34] .

Notably, findings of this research suggested that self-efficacy was the most powerful predictor of the six subconstructs of PMT, similar to research findings of other investigators [32, 35, 36] . This positive connection showed that those individuals with greater perceptions of self-efficacy tended to overcome more obstacles, and had much more confidence to participate in CRC screening [37, 38] . Hence, it is necessary to emphasize self-efficacy when designing and carrying out various health education interventions to further improve screening uptake and mitigate the threat of CRC [39, 40] .

As indicated by the results of previous surveys and the present study, response efficacy should also be considered in an effective screening program of colorectal cancer, acting as a promoting factor for screening willingness and behaviors [41, 42] . Participants perceiving higher response efficacy were more likely to believe in the benefits of CRCS and have stronger intention of CRCS, while low response efficacy was reported to be associated with consistent refusal of screening participation [43, 44] . This finding was important for urban Chinese because they lacked adequate education about CRC and enough perception of effectiveness regarding screening [45] .

Another noteworthy discovery of this study facilitates effective implementation of PMT-based screening policies and prevention activities. Results from pathway analysis suggested that knowledge of CRC had a positive relationship with the scores of screening intention both in direct and indirect pathways. Likewise, published literature studies revealed that the respondents who got higher scores in knowledge were more likely to participate in CRC screening [46, 47] . Those who had deficient knowledge of CRC risk factors, clinical symptoms, and detection would be discouraged from attending screening. So, implementation of educational interventions on the part of participants and staff of the primary healthcare system for removing barriers to CRC knowledge was quite essential, as it has already been proved to be a successful facilitator of CRC screening [48] .

Findings from this survey reinforce the importance of focusing on the key social demographic and other factors. Age, physical activity, social status (education and occupation), and medical history (lower digestive tract lesions and family history of cancer) were found to be correlated to the screening intention mediated through at least one of the two PMT subconstructs (response efficacy and self-efficacy). Previous studies have shown that people with a personal history of colorectal disease were more likely to undergo CRC screening, whilst the uptake of screening was lower among older residents and those of lower socioeconomic status [49, 50] . Individuals of higher social status and regular physical activity tend to pay more attention to their health and hold a positive attitude to CRCS. Therefore, targeted educational campaigns and screening programs based on PMT should be designed and established, taking the above factors into account.

Although this study found that only two subconstructs of PMT were significantly related to screening intention, it was not meant to imply that those subconstructs with no statistical significance made no contributions. The previous reports found that risk perception and severity perception were influencing factors of CRC screening [51, 52] . Moreover, in this model, the total variance of the outcome variable was only partially explained, suggesting that other possible influencing factors were not included and further studies were needed.

Nevertheless, some limitations of this survey need to be noted. Firstly, the generalization of the results from this study should be considered with caution given the possible sampling selection bias, since the study sample was obtained from five communities in one city of central China via convenience sampling and the sample size was small. Secondly, due to the inherent deficiency of cross-sectional design, the causal relationship between screening intention and other factors can't be warranted. Therefore, multi-center cohort studies with large samples will be needed in the future to produce stronger evidence. Third, in view of possible reporting bias due to time constraints, the question about screening intention could be further refined in future studies, for example, "Do you plan to have a CRC screening in the next three (or five) years?"

Despite these limitations, this is the first study to assess colorectal cancer screening intention among urban Chinese based on PMT. Moreover, the results of the study offer new data supporting the implementation of effective theory-based cognitive behavioral prevention interventions to enhance willingness to receive colorectal cancer screening. Most importantly, the success of this survey indicates that the PMT developed in the United States could also be used as a good tool to investigate willingness of colorectal cancer screening among urban residents in other developing countries. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study. Written informed consent has been obtained from the participants to publish this paper.

The data presented in this study are openly available contacting corresponding Author.

The authors would like to thank all those who supported this study, including graduate students from the School of Public Health of Wuhan University and community workers and staff in charge of the urban cancer early diagnosis and treatment project in Wuhan, Hubei Province.

The authors declare no conflict of interest. Q19. Screening is needed for people with bloody stool or diarrhea alternating with constipation Q20. People with family history of colorectal polyps or CRC must screen. Q21. CRC screening tests include fecal occult blood tests and colonoscopy. 

Q8. Through colonoscopy, I can find out if it is colorectal cancer.

Q9. If one would like to cure colorectal cancer, it can't be done without colonoscopy.

Response cost Q10. Going to a hospital for colonoscopy wastes a lot of time.

Q11. It takes a long way to a hospital for colonoscopy.

Q12. I felt very embarrassed to go to the hospital for colonoscopy.

Self-efficacy Q13. Colonoscopy is easy to accept for me.

Q14. I have sufficient time to go to a hospital for colonoscopy.

Q15. Even though other people say colonoscopy is not necessary, I will go for it myself.

Q16. Even though colonoscopy costs me money, I will do it.

Trends in and predictions of colorectal cancer incidence and mortality in China from 1990 to 2025. Front

Changing profiles of cancer burden worldwide and in China: A secondary analysis of the global cancer statistics 2020

Colorectal cancer screening

Planning cancer control in Latin America and the Caribbean

The health belief model and colorectal cancer screening in the general population: A systematic review

Colorectal cancer population screening programs worldwide in 2016: An update

Factors associated with colonoscopy compliance based on health belief model in a community-based colorectal cancer screening program Shanghai, China. Int. Q Community Health Educ

Analysis of the screening results of the urban cancer early diagnosis and treatment project in Hubei province from 2018 to 2019

Determinants of participation in colorectal cancer screening among community-dwelling Chinese older people: Testing a comprehensive model using a descriptive correlational study

Protection motivation theory and preventive health: Beyond the health belief model

Protection motivation and self-efficacy: A revised theory of fear appeals and attitude change

The propensity for consumers to offset health risks through the use of functional foods and nutraceuticals: The case of lycopene

Comparison of risk perceptions and beliefs across common chronic diseases

Avoiding cancer risk information

Cultural worldviews and perceived risk of colon cancer and diabetes

Protection motivation theory and physical activity in the general population: A systematic literature review

Investigating the effects of cancer risk and efficacy perceptions on cancer prevention adherence and intentions

Development of an educational video to improve patient knowledge and communication with their healthcare providers about colorectal cancer screening

Development of a formal system for representing behaviour-change theories

Protection motivation theory in predicting intention to receive cervical cancer screening in rural Chinese women

Path analysis of skin cancer preventive behavior among the rural women based on protection motivation theory

Factors associated with initial participation in a population-based screening for colorectal cancer in Catalonia, Spain: A mixed-methods study

Participation and yield of a population-based colorectal cancer screening programme in China

Construction and validation of a lung cancer risk prediction model for non-smokers in China

Determinants of participation and detection rate of upper gastrointestinal cancer from population-based screening program in China

Variation of adenoma prevalence by age, sex, race, and colon location in a large population: Implications for screening and quality programs

Cancer statistics for adults aged 85 years and older

Colonoscopy screening behaviour and associated factors amongst first-degree relatives of people with colorectal cancer in China: Testing the health belief model using a cross-sectional design

Development and validation of a questionnaire to assess knowledge, threat and coping appraisal, and intention to practice healthy behaviors related to non-communicable diseases in the Thai population

Construction of colorectal cancer screening questionnaire based on protection motivation theory and its validity and reliability study

Protection motivation theory in predicting intention to engage in protective behaviors against schistosomiasis among middle school students in rural China

The protection motivation theory for predict intention of COVID-19 vaccination in Iran: A structural equation modeling approach

Effects through 24 months of an HIV/AIDS prevention intervention program based on protection motivation theory among preadolescents in the Bahamas

Protection motivation theory and rabies protective behaviors among school students in Chonburi Province

Predictive factors for preventing hookah smoking and health promotion among young people based on the protection motivation theory

Colorectal cancer screening knowledge, attitudes and behavioural intention among indigenous western Australians

Mediators of screening uptake in a colorectal cancer screening intervention among Hispanics

Predictors of colorectal cancer screening participation in Southern Khorasan (Iran)

Predictors of colorectal cancer screening intention among Iranian adults: An application of the preventive health model

Factorial validity and invariance of a survey measuring psychosocial correlates of colorectal cancer screening among African Americans and Caucasians

A randomized controlled trial of the impact of targeted and tailored interventions on colorectal cancer screening

Psycho-social determinants of colorectal cancer screening in Iran

Behavioural and demographic predictors of adherence to three consecutive faecal occult blood test screening opportunities: A population study

Colorectal cancer screening: An updated review of the available options

Community-based survey of knowledge of, attitudes to and practice of colorectal cancer screening in Hong Kong

A qualitative study of barriers and enablers associated with colorectal cancer screening among somali men in Minnesota

Barriers and facilitators for colorectal cancer screening in a low-income urban community in Mexico City

Awareness, attitude and barriers of colorectal cancer screening among high-risk populations in China: A cross-sectional study

Colorectal cancer screening in rural and remote areas: Analysis of the national bowel cancer screening program data for South Australia

Factors associated with participation in colorectal cancer screening: A population-based study of 7200 individuals

Multifactorial causal beliefs and colorectal cancer screening: A structural equation modeling investigation