Development and psychometric evaluation of the diabetic Men’s dietary behaviors inventory based on the theory of reasoned action

Diabetes mellitus (DM) is among the most prevalent metabolic disorders [1] which causes around four million deaths and many disabilities each year [2, 3]. The number of 20–79 year-old diabetic patients is estimated to reach 642 million by 2040 [4]. The prevalence of DM among Iranian adults was reported to be about 2–10% [5]. According to the recent reports, about 9.6% of Iranian men suffer from DM [6]. Although the prevalence of DM among Iranian men is lower than their female counterparts [7], the prevalence is progressively increasing [8]. Moreover, in Iran, pre-diabetes is more prevalent among men than women [9].

DM and its complications significantly affect the quality of life of the afflicted patients and their families [10]. Moreover, compared with healthy people, hypertension and renal disorders are respectively two and seventeen times more prevalent among diabetics [11]. Not only adults, but also the youth are afflicted by DM. DM-related costs are around 10% of the total healthcare costs [10]. Furthermore, diabetic and pre-diabetic people are more at risk for developing cancer [12].

According to the World Health Organization, one of the main components of DM management is engagement in self-care activities, particularly dietary modifications [13‐16]. Nonetheless, investigations show limited engagement in healthy dietary behaviors among patients with chronic conditions such as DM [15, 17‐19]. Successful DM management greatly depends on self-care activities such as having a healthy diet and performing adequate physical activity [17, 20].

Patients’ engagement in healthy behaviors greatly depends on their beliefs [20]. In other words, one of the factors behind patients’ poor adherence to their dietary regimens is their beliefs [20‐22]. Studies show that beliefs and perceptions of the outcomes of behaviors can predict dietary behaviors [23]. Moreover, diabetics’ perceptions of their family members’ ideas can motivate them for healthy eating and improve their dietary behaviors [16, 24]. Thus, a prerequisite to any intervention for improving their dietary behaviors is to assess and identify beliefs and perceptions contributing to such behaviors [17]. Some studies showed that beliefs and attitudes cannot always predict behaviors and thus, it is needed to use appropriate theories to assess all factors contributing to behaviors.

Health education theories and models can play a pivotal role in identifying factors contributing to behaviors. One of these theories is the Theory of Reasoned Action (TRA) [25, 26].TRA integrates both behavioral and normative beliefs [25] and holds that the most significant factor behind behaviors is intention. In other words, behavioral intention is the most significant predictor of behaviors. Behavioral intention, in turn, is determined by two factors, namely attitude toward behavior and subjective norms. These two factors are affected by personal beliefs. Attitude consists of behavioral beliefs and outcome evaluation while subjective norms comprises normative beliefs and motivation to comply [27].

It has been reported that knowledge and beliefs towards a certain behavior may be different among males and females. In this regard, the results of earlier reports shows that beliefs and behavior of men with diabetes are different than that of women with diabetes [28, 29]. Men do less attention to nutritional education than women and the taste of food play an important role in their food choices [30, 31], additionally, also it is less probable for men than women to control their weight and adhere with healthy foods [32]. As such, we cannot use the same measure for women as for men with diabetes. Although the necessity to assess diabetic men’s dietary beliefs and behaviors [23, 33], there is currently no gender-specific tool for assessing diabetic men’s dietary behaviors inventory. The present aimed at developing and evaluating the psychometric properties of the Diabetic Men’s Dietary Behaviors Inventory (DMDBI).

This cross-sectional study was carried out in 2018. The study was replicated on an earlier study sample, [34] with a time interval of 12 months, attending the Isfahan Diabetes Clinic.

The means of participants’ age and body mass index were 59.26 ± 9.74 and 26.70 ± 4.34, respectively. Table 1 shows participants’ demographic characteristics.

The preliminary DMDBI contained 78 items in six dimensions namely behavioral beliefs (19 items), outcome evaluation (8 items), normative beliefs (17 items), motivation to comply (18 items), behavioral intention (six items), and dietary behaviors (10 items).

The present study sought to develop and evaluate psychometric properties of the DMDBI based on TRA.In order to accomplish this aim, a preliminary 78-item inventory was developed based on the six constructs of TRA. Then, CVI and CVR were calculated and 60 items with acceptable CVI and CVR were kept. After that, item analysis was performed which yielded a 33-item inventory. Finally, exploratory factor analysis revealed a five-factor structure for the inventory. The extracted factors were remarkably similar to the theoretical underpinnings of the study, i.e. the constructs of TRA. TRA holds that “intention” results in “behavior” and is affected by “attitude”. “Attitude”, in turn, consists of “behavioral beliefs” and “evaluation of behavioral outcomes”. Another construct of the theory is “subjective norms” which affects “behavioral intention”. The “subjective norms” construct comprises “normative beliefs” and “motivation to comply”. However, the factor structure of the 33-item DMDBI was slightly different from the constructs of TRA. Internal consistency analysis and Cronbach’s alpha values revealed the acceptable reliability of the DMDBI.

The first factor of the DMDBI was “Beliefs about behavioral outcomes”. The items loaded into this factor were extracted from the “behavioral beliefs” and “normative beliefs” constructs of TRA. Thus, this factor denotes an individual’s beliefs about the outcomes of a dietary behavior [41] and also about significant others’ reactions to the behavior [41, 42]. The second factor of the inventory was “Outcome evaluation”. This factor is the second component of the “attitude” construct of TRA and denotes an individual’s evaluation of healthy dietary behaviors outcomes. As a determining factor behind attitude, this factor is a significant predictor of behavioral intention. Thus, an individual’s attitude to a behavior can motivate him/her to show or modify that behavior. Aghamolaei et al. (2005) and Polly (1992) also confirmed that health-related beliefs are significantly correlated with diabetics’ self-care activities [43, 44]. Besides, Didarloo et al. (2011) [42]. Reported a significant positive correlation between attitude and behavioral intention these findings imply that people who hold stronger beliefs about self-care activities are more likely to engage in the activities.

The third factor of the DMDBI was “Dietary behaviors” which shows an individual’s abstinence from using unhealthy foods. Dietary behaviors have a pivotal role in preventing and managing DM and its complications [45‐47]. TRA-based DMDBI can be used to assess diabetic patients’ dietary behaviors and subsequently, develop strategies for modifying such behaviors and improving patient health [47]. The fourth factor extracted from the DMDBI was “Motivation to comply”. The items of this factor represent the motivation for considering the opinions of family members and significant othersin the time of decision making about dietary behaviors. This factor is completely congruent with the “motivation to comply” construct of TRA. Other people can affect the decisions about healthy or unhealthy behaviors. In other words, one may base his/her intention on the wills of others [42]. Our findings with regard to the “Motivation to comply” factor was congruent with the findings of previous studies. For example, Didarloo et al. (2011) made a cross-sectional study on 352 diabetic women and found that compared with people who did not experience social pressure, those who were under the pressure of their significant others were more likely to show an intended behavior. Moreover, they reported that through affecting intention, subjective norms contribute to the exhibition of self-care behaviors [42]. Nonetheless, Gholami et al. (2014), Kim (2003), and Stoel (2009) reported no significant correlation between subjective norms and the intention to consume fruits and eat a healthy diet [48, 49]. It seems that the relationship between intention and subjective norms widely varies according to the types of behaviors [50, 51]. The last factor of the DMDBI was “Intention to adjust and diversify dietary regimen”. This factor is relatively consistent with the “behavioral intention” construct of TRA which consists of thinking and deciding about showing a behavior [26]. Some previous studies also assessed the intention to adjust and diversify dietary regimen. Stoel (2009) found a significant correlation between attitude and behavioral intention in that people with more positive attitudes towards a diversified diet were more healthy and thus, more likely to show healthy behaviors. TRA holds that intention is the direct and essential prerequisite for behavior [25, 52]. Some of the previous studies also dealt with the intention to abstain from high-risk behaviors such as unhealthy eating [53] and smoking [54]. These studies indicated that the intention to abstain was significantly correlated with behavior, attitude, and subjective norms. Gholami et al. (2014) also noted that dietary habits and attitudes can affect the intention to consume fruits and vegetables [48].

The findings of the present study showed that the DMDBI has acceptable validity and reliability and can be used to study diabetic men’s dietary perceptions and beliefs. Exploratory factor analysis revealed that all DMDBI items had great construct validity. The factor loads of all five factors of the inventory were greater than 0.4, denoting acceptable factor load. Factor load values that are greater than 0.4 show that all items are important and the inventory has acceptable validity. The reliability of the DMDBI was also assessed through the internal consistency assessment method. Cronbach’s alpha values of three DMDBI factors were greater than 0.7, denoting high internal consistency while two factors had acceptable Cronbach’s alpha values (higher than 0.6).

Developed in the present study, the DMDBI was shown to be a valid and reliable instrument for measuring diabetic men’s dietary perceptions and behaviors. Thus, the inventory can be used in studies to assess the effects of nutrition educational interventions on men’s with diabetes. The usefulness of the inventory for female populations should be assessed in future studies.