Factors associated with intention to undergo specific health guidance among Japanese workers using health belief model ()
1. INTRODUCTION
Among 40 - 74 year-old individuals in Japan, 50% of males and 20% of females are strongly suspected to have either metabolic syndrome or prodromal metabolic syndrome [1,2]. Therefore, healthcare insurance providers have been mandated by Japanese law to provide health checkups to their insured customers between the ages of 40 and 74 since April 2008 [3]. Moreover, insurance providers must give people at risk of metabolic syndrome opportunities to undergo a health consultation, which is called “specific health guidance” [3]. People who undergo specific health guidance achieve positive health outcomes, such as weight loss and decreased hemoglobin A1c and serum triglyceride levels [4-6]. However, the specific health guidance completion rate is low (7.7% in 2008, 12.8% in 2009) [7,8]. Thus, it is necessary to investigate why people choose to participate in or forego specific health guidance.
Most of those targeted for specific health guidance are workers; thus, they have various reasons not to participate in health promotion programs. Studies indicated that a lack of time or being busy was the most common reason not to participate in worksite health-promotion (WHP) programs [9-14]. For example, Gucciardi and colleagues reported that people preferred to participate in one 30 - 60 min education intervention session a year [15]. Cost was also an important problem for participation in WHP programs [12,16], and inconvenient location was also identified as a barrier [9]. Thus, overcoming these obstacles is desirable.
In addition to practical barriers, a perceived lack of benefit, feeling non-susceptible to illness, and a lack of motivation were cited as reasons not to participate in WHP programs [10,16]. This suggests that people would take part in WHP programs if they felt them to be beneficial and were thus motivated to participate. Indeed, perceived benefit was shown to facilitate participation in various health-promotion programs [12,17]. Previous studies investigated these benefits, such as being healthy, feeling good, weight loss, improved appearance, and participation in fun activities [12,17,18]. Additionally, although few studies have examined the benefits and barriers that are more strongly correlated with participation in WHP programs, both benefits and barriers are correlated with intention to participate [17,19].
In the Health Belief Model (HBM), which describes why individuals engage in health behaviors using factors such as perceived susceptibility, perceived severity, perceived benefits, and perceived barriers, Janz and Becker subtracted the perceived barriers from the perceived benefit. The resulting difference score (net benefit) has a direct effect on behavior [20]. The other main component of HBM is perceived threat. This derives from two subcomponents: perceived severity and perceived susceptibility. Perceived severity is an individual’s belief about the severity or seriousness of a disease. Perceived susceptibility is the perception of susceptibility to a disease. Threat perception is calculated as the product of perceived severity and perceived susceptibility, and it has a direct effect on behavior. In essence, HBM hypothesizes that if individuals regard themselves as susceptible to a condition, believe that condition has potentially serious consequences, believe that a course of action available to them would be beneficial in reducing either their susceptibility to or severity of the condition, and believe that the anticipated benefits of taking action outweigh the barriers to it, they are likely to take action that they believe will reduce their risk [21].
In HBM, the factors directly influencing behavior are perceived threat and net benefit; however, the HBM include four subcomponents: perceived severity, perceived susceptibility, perceived benefit, and perceived barriers. Therefore, most authors have used these four subcomponents in their analysis [22,23]. Perceived benefit and perceived barriers were strong predictors of behavior [22-26]. Perceived severity and perceived susceptibility were also predictors of behavior, but the correlations were weak or not significant, depending on the study [24,26,27].
Although the correlations between a behavior and the four subcomponents have been identified, the combined influence of the four subcomponents has not. Few studies have analyzed perceived threat and net benefit. One study of risky driving behaviors indicated that three of the four subcomponents, the exception being perceived barriers, failed to predict the likelihood of risky driving behavior, but perceived threat and net benefit did [28]. Thus, we investigated the predictors of intention to undergo specific health guidance using the main HBM components: perceived threat and net benefit. The purpose of this study was to investigate the predictors of intention to undergo specific health guidance using these HBM components. We hypothesized that perceived threat and net benefit would be associated with intention.
2. METHODS
2.1. Participants and Procedure
Subjects were 4861 Japanese health insurance union members. From August to September, 2010, anonymous self-administered questionnaires (original version in Japanese) and envelopes for their return were enclosed with a household medicine application form. Subjects returned questionnaires with the application form to the health insurance society. Next, the applicant ID on the questionnaire was used by the health insurance society to identify the health insurance member, and the authors received the questionnaires, ID, and information on sex and age. The study received approval from the Ochanomizu University Ethics Review Committee.
2.2. Measures
The questionnaires included questions about future intention to undergo specific health guidance, the HBM factors (perceived severity and perceived susceptibility for lifestyle diseases, and perceived benefits of and perceived barriers to undergoing specific health guidance), and other factors (recommendation by associates, family history of lifestyle disease, and self-rated health).
2.2.1. Intention
The intention to undergo specific health guidance was determined using the question “Will you undergo specific health guidance if you become a subject of guidance?” Participants answered on a 5-point scale [never undergo (1) to definitely undergo (5)].
2.2.2. HBM Subcomponents
Perceived severity of lifestyle diseases (perceived severity) was assessed by the question, “Do you think that lifestyle disease is serious illness?” Perceived susceptibility to lifestyle diseases (perceived susceptibility) was asked using the question, “Compared to other people your age, do you think that your chance of having lifestyle disease in the future is higher?” Perceived benefit of specific health guidance (perceived benefit) was determined as follows: “Do you think that you can prevent lifestyle diseases if you undergo specific health guidance?” Perceived barriers to specific health guidance (perceived barriers) were identified using one item, “Do you have any reason not to undergo specific health guidance?” Participants answered each item using a 5-point scale [strongly disagree (1) to strongly agree (5)].
Additionally, we asked subjects who endorsed a perceived barrier with a score between 2 and 5 (i.e., those who perceived a barrier) to identify their reasons for not undergoing specific health guidance. Participants responded by answering “yes” or “no” to each of eight items.
2.2.3. Other Factors Related to Undergoing Specific Health Guidance
Recommendation by associates (recommendation) was asked thus: “Have you ever received any recommendation about modifying your lifestyle from people around you such as family, friends, and physicians?” Participants answered on a 5-point scale [strongly disagree (1) to strongly agree (5)].
Family history of lifestyle diseases (family history) was determined with one item: “Have your parents, siblings, grandparents, uncles, or aunts ever been diagnosed with lifestyle diseases?” Participants were allowed to provide multiple answers.
Self-rated health was asked using one item: “How has the condition of your health been recently?” Participants answered on a 5-point scale [extremely bad (1) to extremely good (5)].
2.3. Statistical Analyses
A total of 3645 individuals responded to the questionnaire (response rate, 75.0%), and 3457 answered all variables (valid response rate, 71.1%). Sample size for an effect size of 0.1 using an alpha of 0.01 and beta of 0.8 was calculated to be 1169. Therefore, this study had sufficient power. Frequencies and median (interquartile range: IRQ) were determined for basic attributes, and the Mann-Whitney test and Spearman’s correlation coefficient were conducted to assess the associations between intention and basic attributes. New variables termed “perceived threat” and “net benefit” were generated. Perceived threat was calculated by multiplying the perceived severity score by the perceived susceptibility score, and net benefit was calculated by subtracting the perceived barriers score from the perceived benefit score. Then, each Likert-scale item was transformed to binary variables by dichotomizing the scores at the median score (See Tables 1 and 2).
Bivariate and multivariate logistic regression analyses were conducted to evaluate the associations between intention and the four HBM subcomponents, perceived threat, and net benefit. Bivariate and multivariate logistic regression analyses were conducted with intention as the dependent variable and the HBM factors as independent variables. In the multivariate analysis, the independent variables were the four HBM subcomponents from Model

Table 1. Frequency distribution and bivariate/multivariate logistic regression analyses of perceived severity, susceptibility, benefit, and barriers as predictors of intention (n = 3457).

Table 2. Frequency distribution and bivariate/multivariate logistic regression analysis of perceived threat and net benefit as predictors of intention (n = 3457).
1 and perceived threat and net benefit from Model 2. In the logistic regression analysis, we entered sex, age, recommendation, family history, and self-rated health as control variables.
Finally, the frequency of reasons not to undergo specific health guidance were summarized.
Statistical significance was considered at p < 0.01, and all tests were two-sided. Statistical analyses were conducted using SPSS, version 19 (IBM Japan, Ltd., Tokyo, Japan).
3. RESULTS
3.1. Descriptive Information
The majority of subjects were female (57.1%). The age of subjects ranged from 22 to 68 years, with a median (IQR) of 39.0 (31.0 - 47.0).
The median (IQR) intention score for males was 3.0 (3.0 - 4.0), and that for females was 3.0 (3.0 - 4.0). There was no significant correlation between intention and sex (U = 1,456.9, p = 0.80) or between intention and age (Spearman’s r = –0.026, p = 0.13).
3.2. Bivariate Correlation between Intention and HBM Components
A bivariate logistic regression analysis revealed a bivariate correlation between HBM components and intention (Tables 1 and 2). Perceived severity (OR: 1.83, 95% CI: 1.59 - 2.10), perceived benefit (OR: 6.91, 95% CI: 5.94 - 8.04), perceived barriers (OR: 0.11, 95% CI: 0.09 - 0.13), perceived threat (OR: 1.52, 95% CI: 1.31 - 1.76), and net benefit (OR: 11.28, 95% CI: 9.60 - 13.25) were correlated with intention, and all associations were statistically significant (all p < 0.01). With the exception of perceived barriers, the direction of all correlations was positive.
3.3. Multivariate Correlation between Intention and HBM Components
A multivariate logistic regression analysis was performed to examine the correlation of perceived severity, perceived susceptibility, perceived benefit, and perceived barriers with intention after adjusting for each factor. As shown in Table 1, perceived severity (OR: 1.52, 95% CI: 1.29 - 1.80) and perceived benefit (OR: 4.67, 95% CI: 3.95 - 5.51) showed positive correlations with intention, and perceived barriers showed a negative correlation (OR: 0.15, 95% CI: 0.13 - 0.18); this model had a discrimination rate of 75.6%.
According to the HBM, perceived threat and net benefit directly influence the likelihood of committing an action. A multivariate regression analysis was also performed to examine the correlation of these two HBM variables with intention. Both were statistically significant predictors of intention; this model had a discrimination rate of 77.2%. Although net benefit had a stronger correlation with intention (OR: 1.48, 95% CI: 1.25 - 1.76) than did perceived benefit and perceived barriers, perceived threat had a weaker correlation (OR: 11.23, 95% CI: 9.55 - 13.20) than did perceived severity (Table 2).
3.4. Barriers to Undergoing Specific Health Guidance
Table 3 summarizes the barriers to undergoing specific health guidance as reported by study participants. Of the total, 2473 subjects rated perceived barriers as 2 - 5 (2473/3457 = 71.5%), and 1981 provided reasons not to undergo specific health guidance (1981/2473 = 80.1%). “Can’t take the time” and “Don’t know contents” were the most frequently reported reasons not to undergo “specific health guidance”. Think that specific