This descriptive-analytical cross-sectional study was conducted on 410 average risk individuals of CRC who were referred to comprehensive health services centers in Urmia, Iran, 2021. The inclusion criteria included individuals aged 50 to 69 years with an average risk of CRC, physical and mental ability to answer questions, and consent to participate in the study. Exclusion criteria were incomplete completion of the questionnaire.
The minimum sample size required was determined 338 individuals according to a previous similar study and considering the standard deviation of 0.75 for the mean score of CRC screening [18], 95% confidence level (z = 1.96), maximum margin of error or precision (d = 0.08), and using the sample size determination formula for estimating a single mean. Then, to enhance the study power, the number of samples was finally considered 410 individuals.
$$n=\frac{Z_{1- {\propto }\!\left/ \ {2}\right.}^2{S}^2}{d^2}=\frac{1.96^2\ {0.75}^2}{0.08^2}=338$$
A multi-stage cluster sampling method was used for the sampling. First, the city of Urmia was divided into four geographical regions of north, south, east, and west. Then, an urban comprehensive health service center was selected from each region using a simple random sampling method and by lot. Next, by referring to the selected centers and coordinating with the head of the centers, the required samples were completed in proportion to the number of individuals referring to each selected center, from among the individuals who met the inclusion criteria and consented to contribute, via convenience sampling method.
In order to determine whether an individual is at average risk for CRC or not, when going to the health centers for sampling, the information of the health records of the samples available in the centers, as well as the information of the health staff of the centers were used. Also, before completing the questionnaires, the subjects themselves were also asked about the inclusion criteria, and finally, once that an individual was found to be at average risk for CRC and met the other inclusion criteria, he/she was enrolled into the study.
The data collection tool was a researcher-made questionnaire consisting of two parts. The first part captured demographic information and the medical history of participants. The second part involved questions designed based on constructs of motivational phase of HAPA (including risk perception, outcome expectancies and action self-efficacy), and TPB (including behavioral beliefs, outcome evaluation, normative beliefs, motivation to comply, control beliefs, and perceived power), as well as behavioral intention to undergo CRC screening.
The initial questions of the researcher-made questionnaire were designed based on a literature review and opinions of experts in fields related to research and scale development, after which its validity and reliability were measured and approved. In order to determine the validity, two methods of face validity (qualitative and quantitative type) and content validity (quantitative type) were used.
In the qualitative face validity, 20 individuals from the target group were interviewed face to face. They were asked about the suitability and proper relevance of the questions with each other and with the related construct, difficulty in understanding the words, phrases, and statements, as well as possibility of ambiguity and misinterpretations regarding the meanings of words, phrases, and statements. If there was a problem, their opinions would be taken and included in the questionnaire [19].
In the quantitative face validity, the impact score was calculated for each question. For this purpose, a panel of experts was employed, where the questionnaire was given to 10 experts in fields related to research and scale development (including 6 Health education specialists, 2 Epidemiologist, 1 Gastroenterologist, and 1 General surgeon); they were asked to assign each question a score of 1 to 5 in terms of their importance. A score of 1 indicates the lowest, while a score of 5 represents the highest importance. Questions with an impact score greater than 1.5 were deemed suitable for further analysis and remained in the questionnaire; otherwise, they were excluded [19].
In the quantitative content validity, the prepared pilot questionnaire was provided to the panel of experts mentioned above, where the content validity ratio (using the criterion of essentiality) and content validity index (using the relevance, clarity, and simplicity criteria) were calculated. Questions with a content validity ratio of greater than 0.62 and a content validity index of larger than 0.79 were accepted [19].
Cronbach’s alpha coefficient was used to assess the reliability of the researcher-made questionnaire. For this purpose, the prepared pilot questionnaire was given to 30 people in the target group, and after completing the questionnaires, Cronbach’s alpha coefficient was calculated. For all constructs, Cronbach’s alpha coefficient was above 0.7, so the reliability of the tools used in this study was optimal [19].
CVR, CVI, and Cronbach’s alpha were 0.916, 0.959, and 0.942, respectively, for risk perception constructs. For other constructs, the following were obtained: outcome expectancies (0.895, 0.934 and 0.832), outcome evaluation (0.895, 0.934 and 0.824), action self-efficacy (0.942, 0.970 and 0.946), normative beliefs (0.875, 0.913 and 0.925), motivation to comply (1, 1 and 0.820), control beliefs (0.847, 0.924 and 0.888), perceived power (0.847, 0.924 and 0.836), and behavioral intention (0.916, 0.927 and 0.912).
The initial questionnaire involved 111 construct questions, which decreased to 100 questions after dealing with validity and reliability. The final questionnaire included 12 questions associated with the construct of risk perception, 12 questions with outcome expectancies, 12 with outcome evaluation, 13 with action self-efficacy, 8 with normative beliefs, 4 with motivation to comply, 18 with control beliefs, 18 with perceived power, and three questions related to behavioral intention. Possible answers to constructs of motivational phase of HAPA and TPB were scored in 5-point Likert including strongly disagrees (1), somewhat disagrees (2), have no opinion (3), somewhat agree (4) and strongly agree (5). In general, obtaining a higher score in each construct would indicate a good condition of the subject in terms of the understudy construct. The questionnaires were completed by trained interviewers and through self-reporting technique.
Ethical considerations of the present study included receiving the ethics’ code from the research ethics committee of the Vice Chancellor for Research & Technology of Urmia University of Medical Sciences (IR.UMSU.REC.1398.201), receiving a written letter of introduction from relevant authorities to present to research environments, the presence of researchers in selected centers and stating the objectives of the study, obtaining informed consent from the volunteers to participate in the study, presenting sufficient explanation to them about the purpose of the study and the method of work, as well as assuring them that their participation in the study was entirely voluntary. If they did not wish to either participate or continue, they could withdraw from the study, and their information would be kept confidential by the researcher, and the study results would be reported only in general. The questionnaire had no first or last name.
Finally, the data obtained were analyzed in SPSS software version 23 using descriptive statistics (mean, standard deviation, min, max, percentage, and frequency) and analytical statistics including Kolmogorov-Smirnov (to check the normality of the data), Independent t-test (to compare the mean score of CRC screening intention among the two independent groups of the participants), One-way ANOVA (to compare the mean score of CRC screening intention among the three or more independent groups of the participants), Pearson correlation coefficient (to determine the degree of linear correlation between CRC screening intention and the independent variable), and Multiple linear regression with Enter method (to determine the predictive power of the constructs of motivational phase of HAPA and TPB on the CRC screening intention). The results were considered statistically significant at p < 0.05.