^aDivision of Epidemiology, Department of Public Health and Forensic Medicine, Tohoku University Graduate School of Medicine, Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan

^bDivision of Behavioral Medicine, Department of Functional Medical Science, Tohoku University Graduate School of Medicine, Seiryo-machi, Aoba-ku, Sendai, Miyagi, Japan

^cDepartment of Public Health, Yamagata University School of Medicine, Yamagata, Japan

Abstract

Objective

Obesity is an increasingly prevalent public health problem worldwide, and is associated with a higher risk of developing various noncommunicable diseases. To further examine the association between personality and overweight, obesity, or underweight, we conducted a cross-sectional analysis in Japan. We hypothesized that extraversion and psychoticism would have a positive association with overweight, and that neuroticism and lie would have an inverse association with overweight, whereas the association between personality and underweight would be the reverse image of overweight.

Methods

In 1990, 30,722 subjects (40–64 years of age) completed a self-administered questionnaire including body weight and height and the Japanese version of the Eysenck Personality Questionnaire-Revised Short Form. Multivariate logistic regression analysis was used to calculate odds ratios for overweight [body mass index (BMI)≥25.0 kg/m²] or underweight (BMI<18.5) relative to each category on the personality subscale.

Results

In men and women, extraversion and psychoticism had positive associations with overweight, whereas neuroticism had an inverse association. Lie had an inverse association with overweight in men. In men and women, only extraversion had an inverse association with underweight and neuroticism had a positive association with underweight.

Conclusion

Our findings indicate that personality is associated with both overweight and underweight. These results may provide clues to devising more effective measures for preventing overweight, obesity, or underweight or for weight control intervention.

Keywords: Adults; Japanese; Overweight; Personality; Underweight

Article Outline

Introduction

Research methods and procedures

Study population

Measurement

Statistical methods

Results

Discussion

Acknowledgements

References

Introduction

There have been a number of studies of interventions for preventing or controlling excess weight or obesity [4] and [5]. These interventions included physical activity, dietary pattern, pharmacological or behavioral therapy, or a variety of their combinations [5]. In practical terms, such obesity treatment programs may produce a wide range of responses in terms of both the magnitude and rate of weight changes due to differences in physiological and psychological factors among individuals [6]. A previous study has suggested that psychological aspects should be considered in weight loss strategies for obesity management [7]. Jonsson et al. [7] found that some psychological aspects were able to predict weight loss outcome 1 year after the end of their treatment in 28 obese patients.

Although consideration of personality in interventions for weight control is of potential importance, the psychological correlates and causes of overweight or obesity remain unclear [8]. To date, three reported cross-sectional studies have examined the association between the subscales of the Eysenck Personality Questionnaire (EPQ) and obesity. In the four subscales of EPQ, extraversion represents sociability, liveliness, and surgency; neuroticism represents emotional instability and anxiousness; psychoticism represents tough-mindedness, aggressiveness, coldness, and egocentricity; and lie represents unsophisticated dissimulation and social naivety or conformity [9]. Among previous studies, two found a positive association between extraversion and obesity [10] and [11], but Faith et al. [12] reported an inverse association in women and a positive association in men. With regard to the association between neuroticism and obesity, one study found an inverse association [10], one found no association [11], and one found a positive association in women and no association in men [12]. Only one study referred to the association between psychoticism and obesity, and reported a positive association in men but no association in women [12]. In contrast, two studies reported no association between the subscales of EPQ (extraversion and neuroticism) and underweight [13] and [14]. Furthermore, all of these studies were conducted in Western countries, and there is no report from Asian countries, where the prevalence of overweight, obesity, or underweight differs considerably from that in Western populations.

To further examine the association between personality and overweight, obesity, or underweight, we conducted a cross-sectional analysis in Japan. We hypothesized that extraversion and psychoticism would have a positive association with overweight and that neuroticism and lie would have an inverse association with overweight, whereas the association between personality and underweight would be the reverse image of overweight. Such a population would provide an opportunity to examine the association between personality and overweight as well as underweight.

Research methods and procedures

Study population

We analyzed cross-sectional data from a baseline survey conducted for the Miyagi Cohort Study. We have already reported the design of this prospective cohort study in detail elsewhere [15]. Briefly, we delivered two self-administered questionnaires to all 51,921 residents aged 40–64 years living in 14 municipalities of Miyagi Prefecture in rural northern Japan from June through August 1990. The first questionnaire asked about various health-related habits, and the second was the Japanese version of the EPQ-Revised (EPQ-R) Short Form [16]. The questionnaires were delivered to, and collected at, the subjects' residences by members of health promotion committees appointed by the municipal governments. The response rate for the first questionnaire was 91.7% (n=47,605), and that for the second questionnaire among the respondents to the first was 87.0% (n=41,424). We considered the return of self-administered questionnaires signed by the subjects to imply their consent to participate in the study. The study protocol was approved by the Institutional Review Board of Tohoku University Graduate School of Medicine.

Of the 41,424 subjects who responded to the two questionnaires, we excluded 54 subjects who responded “yes” or “no” to each of the 48 items and 8600 subjects for whom responses to any of the 48 items in the EPQ-R were missing. We further excluded 1133 subjects who had incomplete responses for body weight or height information. Because Nakaya et al. [17] reported that a neurotic tendency among subjects with a prevalence of cancer may have been a consequence, rather than a cause, of having been diagnosed with cancer, 915 subjects who had entered a history of cancer, stroke, or myocardial infarction in the self-reported questionnaire were further excluded. Finally, 30,722 subjects were used for the final analysis.

Measurement

Personality was measured using the Japanese version of the EPQ-R Short Form, one of a series of personality inventories developed by Eysenck and Eysenck [18]. The EPQ-R Short Form has 48 questions with dichotomized responses (yes or no); there are 12 questions for each of the four personality subscales (extraversion, neuroticism, psychoticism, and lie). Scores on each subscale range from 0 to 12, with higher scores indicating a greater tendency to possess the personality trait represented by each subscale. Extraversion represents sociability, liveliness, and surgency; neuroticism represents emotional instability and anxiousness; psychoticism represents tough-mindedness, aggressiveness, coldness, and egocentricity; and lie represents unsophisticated dissimulation and social naivety or conformity [9]. Several Eysenck's personality questionnaires have been translated into Japanese [16], [19] and [20]. In previous work, Hosokawa and Ohyama [16] developed the Japanese version of the EPQ-R and examined its reproducibility and validity among 329 college students and 253 adults. Cronbach's α coefficient, a measure of internal consistency, was greater than .70 for all subscales except psychoticism (.42 for college students and .48 for adults). Test–retest reliability coefficients of the four subscales over a 6-month period ranged from .70 to .85, indicating substantial stability. Confirmatory factor analysis supported the original theoretical structure of the four scales proposed by Eysenck and Eysenck [18].

The baseline survey included questions on self-reports of body weight and body height, and BMI was calculated as the weight divided by the square of the height (kg/m²). We grouped subjects into the following three categories: underweight (BMI<18.5), normal (BMI=18.5–24.9), and overweight (BMI≥25.0), based on the proposal by the World Health Organization in which BMI values of 18.5–24.9 and ≥25.0 are defined as normal and overweight or preobese, respectively [1] and [21].

We evaluated the validity of self-reported body weight and body height. Among the study subjects, 7153 individuals had their body height and weight measured during health examinations provided by local governments in 1990. The Pearson's correlation coefficient (r) between self-reported and measured values was .97 for body weight, .85 for body height, and .91 for BMI. Thus, self-reported height and weight at the time of the baseline questionnaire were sufficiently valid. The questions included the following data: age, education, marital status, history of diseases, cigarette smoking, alcohol consumption, walking time per day, physical activity time per week, and job type.

Statistical methods

Each personality subscale was divided into four categories by sex to obtain even-sized quartiles as closely as possible. We used multivariate logistic regression analysis to calculate odds ratios (ORs) and 95% confidence intervals (CIs) for overweight or underweight relative to each category of a personality subscale, with the lowest category treated as the reference group. Trend tests were performed by treating personality subscales as continuous variables.

In these analyses, we regarded the following data as covariates: age (continuous variable); education (up to 19 years of age or 19 years or older); marital status (whether or not living with spouse); history of hypertension, renal diseases, liver diseases, gallstone diseases, diabetes mellitus, peptic ulcers, or tuberculosis; cigarette smoking (never smoked, smoked in the past, or currently smoking); alcohol consumption (never drank alcohol, drank in the past, or currently drinking); walking time per day (less than 30 min, from 30 min to 1 h, or more than 1 h); physical activity time per week (less than 1 h, 1–2 h, 3–4 h, or more than 5 h); and job type (service-type job, farmer and manufacturing, or others).

We examined in detail potential confounding and effect modification by age and other covariates on the associations between personality scales and overweight or underweight. No statistically significant interaction was observed between personality and other confounding factors for overweight or underweight on a multiplicative scale (data not shown).

All statistical analyses were performed using SAS software, version 9.1 [22]. All the statistical tests reported were two sided. Differences at P<.05 were accepted as statistically significant.

Results

The subjects' characteristics according to the highest and the lowest categories (i.e., approximate quartiles) of each personality subscale are shown in Table 1. Subjects in the highest category of extraversion were less likely to have never smoked (P=.01 in men, P<.0001 in women), to have never drank (P<.0001 in men and women), or to have a history of diseases (P<.0001 in men and women) and were more likely to walk (P<.0001 in men and women) and to engage in exercise (P<.0001 in men and women) than those in the lowest category. Subjects in the highest category of neuroticism were less likely to walk (P=.0003 in men, P=.0001 in women) or to engage in exercise (P<.0001 in men and women) and were more likely to have a history of diseases (P<.0001 in men and women) than those in the lowest category. Subjects in the highest category of psychoticism were younger than those in the lowest category (P<.0001 in men and women) and were less likely to have never smoked (P<.0001 in men and women) or to have never drank (women only; P<.0001). Subjects in the highest category of the lie scales were older than those in the lowest category (P<.0001 in men and women) and were less likely to live with their spouse (P<.0001 in men and women) and were more likely to have never smoked (men only; P<.0001), to have never drank alcohol (P<.0001 in men and women), to walk (P<.0001 in men and women), to engage in exercise (P<.0001 in men and women), and to have attended school up to 19 years of age (women only; P<.0001).