Breast cancer is the most common malignancy in women worldwide and the second leading cause of malignancy death in this population (1). In Iran, 16% of all cancers in women are breast cancer, which is the number one cancer among Iranian women (2). The incidence age of breast cancer in Iranian women is ten years earlier than that of women living in developed countries (3). There's a rising trend of death from breast cancer in Iran, although this rate is still relatively low compared to industrialized countries (4). The significant increase in treatment costs in the second and third stages underscores the importance of early diagnosis and policy-making for necessary interventions on this subject in Iran (5). According to the current knowledge, one-third to half of the cancers are potentially preventable. This finding owes to the greater understanding of the risk factors of some malignancies, including breast cancer, and the development of related preventive interventions (6). Studies show that prognosis decline is proportional to the cancer progression rate. The upper is the rate of progression, the worse will be the prognosis, and the lower the 5-year survival rate (7).
Moreover, most patients are not informed about how to perform early-stage breast cancer detection by touching, and self-examination training or screening tools should therefore be incorporated into women's health promotion programs (8). Early detection is crucial to improving outcomes and survival rates (9). Breast Self-Examination (BSE) could be a self-diagnostic method that is easy, effortless, private, safe, cost less, and equipment-free, and is helpful for screening (10). Nevertheless, BSE is a forgotten part of the health behaviors of many Iranian women; therefore, Iranian women should be further educated about the benefits of this behavior (11). In addition, targeted training based on health promotion models for Iranian women has been assessed as useful for increasing breast cancer awareness and controlling the burden of cancer in the future (12). Although having sufficient knowledge on various health issues is a prerequisite for health behaviors, there is not always a direct relation-ship between people's knowledge and behaviors. In other words, people with higher knowledge do not necessarily have better health behaviors. There is a relationship between the two factors: (i) People's attitudes, which can predict their behavior too; that is, when people's attitudes are improved, more appropriate actions can be expected from them; and (ii) Education, which can significantly improve a person's knowledge, attitude, and behavior about BSE (10). In addition, perceived self-efficacy can lead to behavior adoption, while perceived subjective or objective barriers can lead to the non-adoption of the BSE behavior. Therefore, educational interventions for promoting BSE health behavior should emphasize improving women's performance to promote their self-efficacy and modify the perceived barriers through proper education (13).
Different models are used in health education to study health behavior. Theories can answer planners' questions about why people do not have the desired behavior, how to change behaviors, and what factors should be considered in evaluating programs (14). Furthermore, the success rates of health education and promotion programs vary, and those based on proper theories and models are more likely to succeed. The-ories and models form the basis of educational inter-ventions, are considered tools for explaining and justifying health plans, and comprise a framework for health intervention planning; they also enable the repeating of health interventions and provide a structured process for analyzing successes and failures (15).
The Extended Parallel Process Model (EPPM) is one of these health education models that was introduced by Witte in 1992 based on the central variable of fear and aimed to answer the questions of how and when the instinct of fear works effectively and successfully to promote the right behavior and when and why it fails. According to this model, messages mixed with fear-arousal produce two assessments in a person. One is "threat assessment," and the other is "efficiency assessment", both of which lead to three reactions: Indifference, rejection of the message, and acceptance of the message. Initially, the target audience of these messages assesses the severity of the threat (How does breast cancer occur? What are its symptoms, risks, and complications?), and their sensitivity to the risks of the threat (Am I at risk for breast cancer?). By considering these two factors (perceived severity and sensitivity to the risk), the person feels fear and threat. Then, if the presented messages motivate the audience, the second assessment emerges, i.e., assessing the efficiency and effectiveness of the message, and the degree of importance of the threat is judged (16). The key distinguishing aspect of this model from other models based on the fear instinct is the ability to predict health-related behaviors in individuals by determining critical points. The audience is placed at one of two processes, either risk control (increased likelihood of a positive behavior change) or fears control (decreased likelihood of a positive behavior change), depending on whether the degree of perceived threat is dominant or the degree of perceived effectiveness. The numerical calculation of the critical point enables the researcher to predict the likelihood of positive behavior change in the future (17). Since breast cancer is a disease that leaves its first psychological traces in the patient in the form of fear and worries and given the two main EPPM constructs of risk control and fear control, a model has to be used that incorporates this concept and helps empower audiences to adopt self-care behaviors. To date, no research has been conducted on the effectiveness of the EPPM in promoting breast cancer screening behaviors in Iran. This study was conducted to increase women's sensitivity and perception of the risk of this disease to improve their self-care behaviors, and while increasing their awareness about the risk factors of malignant breast tumors and encouraging self-care behaviors, it helps them benefit from early cancer detection in the future and the reduction of the economic and health costs of this disease.

 

This experimental study was a randomized controlled trial on 80 women aged 20-60 years in Some Sara in Guilan Province of Iran. The sample size was estimated as 72 (36 per group) using the data from a related study (Nangle et al. 2003) and considering a high test power (90%), type-I error of 5%, and equal division of the subjects into the two groups. Considering the possibility of sample loss and to prevent a decline in the study's power, 10% was added to the sample size, and it finally reached 80 (40 patients per group). The inclusion criteria were: The lack of pregnancy and nursing, no history of malignant breast tumors, lack of malignant breast tumor detection during the study, incorrect or non-performance of breast self-examination by observation and the researcher's confirmation, willingness to regularly attend training courses, and the unlikelihood of immi-grating from Some Sara. The exclusion criteria were: Unwillingness to continue participation in the study, non-regular attendance of the training classes (for the intervention group), or immigration from the city. Sampling was carried out in several stages. First, a list of the health centers in the city of Some Sara was prepared; then, two centers with the largest population coverage and various demographic characteristics were selected. After obtaining written permission from the Ethics Committee (IR.SBMU.PHNM.1395.577), the researcher obtained written consent from the eligible women. Also, this study was confirmed at the Iranian Clinical Trial Registration Center with IRCT clinical trial registration number: IRCT20200525047565N1. Sampling continued (except on holidays) until the desired sample size from both centers was reached through convenience sampling on different days of the week (Sat., Mon., Wed. for the intervention group and Sun., Tues., Thu. for the control group). Due to differences in the physiological characteristics of women of reproductive age and the possible differences in self-care behaviors, participants in both groups were divided into two age groups: A group aged 20-39 years and a second group aged 40-60 years. The training package was prepared using health promotion and reproductive health experts' opinions based on the EPPM and was presented based on the steps recommended in this model (16, 17).
The required data were collected using a demo-graphic questionnaire, the Breast Self-Examination Behavior Predicting Scale (11 items), and the Risk Behavior Diagnosis Scale (RBDS) based on the EPPM to determine perceived threat (12 items) and perceived efficiency (12 items), which are generally used to determine people's subjective perception of risk and the probability of change in health behavior (16). Data were collected before, immediately after, and two months after the intervention. The RBDS was scored based on a 5-point Likert from 'strongly disagree' (=1) to 'strongly agree' (=5). The minimum and maximum scores in each construct were 12 and 60. The training sessions were four 60-min sessions held over one month in the intervention group. That is, an average of four one-hour sessions with an average attendance of seven to ten people were held every week, making for a total of 16 training sessions. Educational aids, such as anatomical moulages and video and slide display by video projectors, were used to convey the materials better and facilitate understanding the content. In the intervention group, the questionnaires were answered before, immediately after, and two months after the intervention. The questionnaires were completed before, immediately after, and two months after receiving routine care in the control group. The BSE Behavior Predicting Scale was also completed by the researcher before, immediately after, and two months after the last training session in both groups through observation. After the educational intervention was over, to adhere to ethical standards and not deprive the control group of the administered care, a summary of the educational materials was provided to the control group in one session through a speech over 20 minutes but without using the multiple EPPM stages.
The validity of the RBDS was determined by qualitative and quantitative content validity and qualitative face validity. The Content Validity Index (CVI) and Content Validity Ratio (CVR) of the questionnaire were calculated for each of the subscales of perceived sensitivity (0.96 and 0.95), perceived severity (0.96 and 0.93), perceived effectiveness (0.83 and 0.99), perceived self-efficacy (0.76 and 0.94) and performance (0.94 and 0.99), respectively. The total CVI and CVR of the tool were also calculated as 0.9 and 0.95, respectively. Also, the reliability of the tool was determined by measuring its internal consistency and Cronbach's alpha coefficient, which was 0.75, 0.8, 0.79, 0.85, and 0.75 for the subscales of perceived sensitivity, perceived severity, perceived effectiveness, perceived self-efficacy, and performance, respectively. In addition, the reliability of the tool over time was also assessed by the test-retest method with at least a two-week interval, and the correlation analysis of the two test scores yielded a reliability coefficient of 0.9. The crude data were analyzed in SPSS 21 (SPSS Inc., Chicago, Ill., USA) using the Chi-square test, Mann-Whitney test, Fisher's exact test, and therefore the inter-group and intra-group repeated measures ANOVA. The significance level was less than 0.05.