Psychometric Properties of the Persian Version of Champion ’ s Health Belief Model Scale for Colorectal Cancer Screening

BACKGROUND
Colorectal cancer is a serious health problem. Early detection of colorectal cancer is crucial for treatment and reducing mortality. Beliefs related to colorectal cancer have been found to be a factor in a person's decision about colorectal cancer screening programs. To determine such beliefs, a valid and reliable instrument is necessary.


OBJECTIVE
The aim of this study was to adapt and determine the psychometric properties of the Persian version of Champion's Health Belief Model Scale of breast cancer screening in the measurement of beliefs toward colorectal cancer (CRC) screening.


MATERIALS AND METHODS
The 'forward-backward' procedure was applied to translate the instrument from English into Persian. This study was conducted in Iran from June 2012 to May 2013. A convenience sample of 200 individuals aged 50 years and older was recruited from the population at the outpatient clinics in the three teaching hospitals. Validity was assessed using content, face and construct validity. To test reliability, the internal consistency was assessed by using Cronbach's alpha coefficient and test-retest (intraclass correlation coefficient) analyses. Exploratory factor analysis was used to assess the construct validity and determine the factors of adapted Champion's Health Belief Model Scale.


RESULTS
The mean age of the participants were 62.5 years (SD=10.8 years) and the majority of them (75.5 percent) were female. The results of exploratory factor analysis indicated a six-factor solution for the questionnaire (benefits, motivation and confidence, seriousness, susceptibility, emotional barriers and background barriers) that jointly accounted for 55.52% of variance observed. Cronbach's alpha of the subscales ranged from 0.57 to 0.89 and test-retest reliability ranged from 0.81 to 0.93 indicating a good range of reliability.


CONCLUSIONS
The findings of this study suggest that the Persian version of Champion's Health Belief Model Scale of CRC screening has good psychometric properties and could be an appropriate measure for health beliefs related to CRC screening in national and international studies.


Introduction
Colorectal cancer (CRC) is a significant serious health problem and the fourth leading cause of cancer death over the world (Yusoff et al., 2012).It is the third most common cancer diagnosed and is the second leading cause of cancer death among both genders in the United States, accounting for 11% of all cancer deaths (James et al., 2002).Although previously a predominantly Western disease, the incidence of CRC is increasing in Asia (Sung et al., 2005).In Iran, the age-standardized incidence (ASR) from 2005-2009 is 38.0 per 100,000 population (Safaee et al., 2012) and is the most prevalent cancer among males and second among females (Alireza et al., 2005).Because of the nature of this disease and the increasing rate, it is highly preventable and suitable for screening (Katz et al., 2007).Early detection of colorectal cancer is crucial for treatment and detection in related mortality (Levin et al., 2008).Despite the importance of CRC

Psychometric Properties of the Persian Version of Champion's Health Belief Model Scale for Colorectal Cancer Screening
Zahra Taheri Kharameh 1 , Sahar Foroozanfar 2 , Hadi Zamanian 3 * screening, screening rates are low (Honda et al., 2005).Taylor believes that the lack of participation in screening programs can be justified based on health belief model (Taylor et al., 1999).The Health Belief Model (HBM) was created by social psychologists in the 1950s in the US Public Health Services to find why people did not participate in health-screening programs (Gipsh et al., 2004;Salz et al., 2009).There are six key concepts in the HBM including: i) perceived susceptibility (is one's belief regarding the chances of developing a condition); ii) perceived severity (a person's opinion of the seriousness of a condition and its sequelae); iii) perceived benefits (the opinion of the effectiveness of some advised action to reduce the risks of a condition); iv) perceived barriers (one's beliefs regarding the total costs of implementing the recommended action); v) confidence (one's intention to take the recommended action); and vi) motivation related to performing the health behaviors (Rosenstock, 1966;Rosenstock et al., 1988).
Few studies have examined beliefs and behaviors related to CRC screening in Iranians people.The results of a previous study conducted in Iran about knowledge toward CRC screening among people aged 50 years and older, showed that more than half of the participants had never heard about colorectal cancer screening tests (Salimzadeh et al., 2011). Javadzadeh et al. (2012) found that a significant relationship between knowledge and all domains of HBM excepting perceived benefits among individuals more than 50 years old (Javadzade et al., 2012).The most common reasons cited for not doing CRC screening were lack of knowledge and the belief that it is not necessary.Understanding of attitudinal components of health-related behavior has been important.Assessing patients' needs and designing educational strategies accordingly may lead to more effective cancer screening programs.
Jacobs adapted the CHBMS, only by substituting ''colon cancer'' for ''breast cancer'' in the wording of the questions with the same items (Jacobs, 2002).However, a valid and reliable instrument for determining the beliefs of Iranians aged 50 and older related to the recommendation of CRC screening has not been reported.Therefore, the aim of this study was to adapt for CRC screening, translate and assess the psychometric properties of the Persian version of Champion's Health Belief Model Scales (CHBMS) in the measurement of Iranians's beliefs about CRC screening.

Study design and participants
This cross sectional study has been performed in June 2012 to May 2013 in Qom, a provincial city in the central region of Iran.A convenience sample of 200 individuals aged 50 years and older was recruited from population at the outpatient clinics in the three teaching hospitals (Nekooi, Shahid Beheshti, Kamkar hospitals).The inclusion criteria were being 50 years or older, not having had CRC and the ability to communicate in Persian language.

Instruments
For data collection, two questionnaires were used as follows: i) To collect clinical and socio-demographic information of the participants, a questionnaire was used that was consisted of questions about age, gender, marital status, employment, education, health insurance and family history of CRC; ii) Health belief was assessed using the CHBMS.This instrument was originally developed by Champion in 1993Champion in and revised in 1997Champion in and 1999 to assess beliefs towards breast cancer screening.It contains 61 items with 8 subscales.We substitute ''colon cancer'' for ''breast cancer'' in the wording of the questions in 6 subscales (with Champion's permission).The items related to breast self-examination in the confidence subscale and mamogram were divided from the instrument.Finally, 36 items remained to assess health beliefs for colorectal cancer screening according to 6 subscales; perceived susceptibility (5 items), perceived seriousness (7 items), perceived benefits (6 items), perceived barriers (6 items), health motivation (7 items) and confidence (5 items).Items were formatted with a 5-point likert scale, from 1 (strongly disagree) to 5 (strongly agree).

Translation
After receiving permission from the original author (Champion, 1999) for the translation and application of the CHBMS for colorectal cancer screening, the instrument was translated based on the international quality of life assessment project (IQoLAP) guideline (Gandek and Ware, 1998).The questionnaire was translated from English to Persian by two professional translators and the primary Persian version of the questionnaire was developed based on the comparison of the two translations.Next, the questions were adapted for CRC screening instead of breast cancer.The Persian version was backtranslated to English by two professional translators who has never seen the original version before.Back-translated versions and The original CHBMS were compared item by item and a final Persian version of the questionnaire was obtained.
To test the content validity of the scale, a multidisciplinary panel was developed including an Entrogastrologist, two general physicians, two health education professors and five professors in nursing.They were asked to comment on the reasonability, suitability, attractiveness and logical sequence of items as well as conciseness and comprehensiveness of the questionnaire.Moreover, in order to assess the questionnaire's face validity, it was given to 10 individuals aged 50 and older to test its comprehensibility and legibility.According to the presented comments and perspectives by the experts and participants, some items of the questionnaire were slightly simplified and modified.

Data collection
For data gathering, the study procedure was explained to the individuals who met the eligibility criteria.The instrument was applied by face-to-face interview technique.Questionnaire completion took between 10 and 15 minutes.

Data analysis
Participants characteristics and the score of each domain of the CHBMS were analyzed by using descriptive statistics.The construct validity of the questionnaire was performed using exploratory factor analysis.Principle components analysis with varimax rotation was applied.The Kaiser-Meyer-Olkin (KMO) and Bartlett's Test of DOI:http://dx.doi.org/10.7314/APJCP.2014.15.11.4595 Champion's Health Belief Model Scale for Colorectal Cancer Screening Sphericity were used to assess the appropriateness of the sample for the factor analysis.Eigen values above 1 and scree plot were used to determine the number of factors.Factor loadings equal or greater than 0.4 were considered appropriate (Nunnally, 2010).Internal consistency of each scale of the CHBMS was determined using Cronbach's alpha.Cronbach's α coefficient of 0.7 or above was considered to be satisfactory (Schneider, 2004).Test-retest reliability was assessed by computing the intraclass correlation coefficient.The time interval for this assessment was 2 weeks in this study.An ICC >0.80 indicated good test-retest reliability and stability (De Boer et al., 2004).

Ethical considerations
Permission to use the original CHBMS was obtained from Victoria Champion.The study research proposal was approved by the deputy of research, Qom University of Medical Sciences.Ethical approval was granted by the Medical Ethics Committee, the Qom University of Medical Sciences that corroborated the ethical considerations throughout the study.Participation in this study was free to withdraw from the study at any time without having any effect on their treatment process.The participants were provided with information about the study process.Those who voluntarily agreed to participate in this study signed written informed consent.

Sample characteristics
Mean and standard deviation of the participants' age was 62.47±10.78years and 75.5 percent of participants were female.Most of the patients (88%) were from urban districts.68.5 percent of participants were married.The majority of the samples (95%) have no previous CRC screening.More information about the personal characteristics of the study participants are presented in Table 1.

Validity
The Kaiser-Meyer-Olkin adequacy was 0.74 and Bartlett's test of sphericity was significant (p<0.001),showing sampling adequacy.
Factor analysis with principal component factor analysis and varimax rotation was used to determine construct validity by excluding items with factor loading less than 0.4.After varimax rotation, a total of 36 items loaded significantly on six factors which were slightly different from original CHBMS.All 6 factors had an eigenvalue greater than 1, with an explained variance    of 55.52%.A total of items except for two item loading ranged from 0.48 to 0.89.Two items were deleted (factor loading <0.4).

Reliability
Cronbach coefficients for the six subscales ranged between 0.57 and 0.89.For the test-retest reliability, the ICC coefficients ranged between 0.81 and 0.93 for the overall score significant (p<0.001).Table 3 shows the internal consistency and the test retest reliability of the CHBMS.

Discussion
Many studies have been conducted to identify beliefs regarding CRC.Most of these studies aimed at the designing intervention studies to promote CRC screening behaviors.To recognize such beliefs, a valid and reliable instrument is necessary.The aim of this study was to adapt and assess the psychometric properties of the CHBMS related to CRC.
In this study, the CHBMS was translated based on the instrument translation guidelines.The face and the content validity of the questionnaire were confirmed after minor revisions.
We employed the exploratory factor analysis method for evaluating the construct validity of the CHBMS.The KMO value was equal to 0.74 and the Bartlett's test was significant (p<0.001),indicating the appropriateness of the factor analysis model.The final instrument contained six subscales with 34 items.These subscales included benefits, motivation and confidence, seriousness, susceptibility, emotional barriers and background barriers.
The differences between the original and the Persian versions of CHBMS were: i) Two items of the instrument were deleted.''I want to discover health problems early'' was removed from the health motivation subscale.Similarly, the item, ''I can recognize normal and abnormal changes in my bowel habits'' was deleted from the confidence subscale because of loading lower than 0.40; ii)In the Persian version, the 'perceived barriers' sub-scale of the original version was divided into two sub-scales including 'emotional barriers' (consisting of three items) and 'background barriers' (consisting of three items).The literature concerning the transcultural reliability and validity of this scale related to CRC is limited.Ozsoy et al. (2007) evaluated the psychometric properties of the Turkish language version of the CHBMS and reported a structure consisting of five factors including: confidence, benefits, health motivation, susceptibility, barrier, health motivation and seriousness (Ozsoy et al., 2007).
The study findings also revealed that CHBMS had acceptable internal consistency.Coronbach's alpha for different subscales of CHBMS ranged from 0.57 to 0.89.Two sub-scales-including 'susceptibility' and 'background barriers'-had a Coronbach's alpha less than 0.70.Waltz et al. (1993) noted that the number of items of a measure directly contributed to the magnitude of its Coronbach's alpha.Champion reported Coronbach's alpha coefficients between 0.75 and 0.93 for the original version of the CHBMS.Ozsoy et al. (2007) reported Cronbach alpha coefficients for the 5 subscales ranged between 0.54 and 0.88.Coronbach's alpha coefficients in a study conducted by Jacobs et al. (2010) were between 0.60 and 0.78.
In this study, test-retest reliability coefficients ranged from 0.81 to 0.93 indicating a good range of reliability.Compared to the original version, the Iranian version had better results with regard to the domains.Ozsoy et al. (2007) reported a range of stability between 0.72 and 0.91 for the Turkish version of the CHBMS.
In conclusion, the findings of this study suggest that the CRC screening CHBMS is a reliable and valid instrument that can be used to measure's health beliefs related to CRC screening.This questionnaire can also be used to guide the development of more effective educational interventions.
Non-random sampling and a relatively small sample size restrict the generalizability of the study findings.Consequently, multi-cultural studies with larger sample sizes and with groups of first-degree relatives of individuals with CRC are recommended.DOI:http://dx.doi.org/10.7314/APJCP.2014.15.11.4595 Champion's Health Belief Model

Table 2 . Principal Component Analysis of the Persian Version of Champion's Health Belief Model Scales (CHBMS)
*CON indicates confidence; BEN, benefit; MOT, health motivation; SUS, susceptibility; BAR, barrier; SER, seriousness