The Relationship between IMPS-measured Stress Score and

The Relationship between IMPS-measured Stress Score and Biomedical Parameters Regarding Health Status among Public School Workers Kazuhiko Yamamoto1), Masahiro Irie1), Yoko Sakamoto1), Susumu Ohmori2) and Mototaka Yoshinari2) 1) Institute of Health Science, Kyushu University, Fukuoka 2) Kyushu Central Hospital, Fukuoka

Abstract The aim of this study was to examine the relationship between the stress score measured using the Inventory to Measure Psychosocial Stress (IMPS) and biomedical parameters regarding health status among apparently healthy subjects in order to evaluate the validity of the IMPS. Out of the 1,941 public school workers in Kyushu and Okinawa, Japan, who were admitted to a hospital for medical check-ups, 1,499 workers responded to questionnaires which assessed the degree of stress response (i.e., stress score) measured using the IMPS, and the degree of stress tolerance capacity (i.e., stress intolerance score) measured using the Inventory to Measure Stress Tolerance Capacity (IMST). One thousand two-hundred and one workers (684 men and 517 women) were analyzed, excluding 298 subjects who were taking medication for hypertension, hyperuricemia, hyperlipidemia and diabetes, or had a value for glycosylated hemoglobin (HbA1c) 6.0 percent. An increase in the stress score was positively associated with an increase in both body fat percentage and glycosylated hemoglobin values among men, while it was positively associated with an increase in plasma triglyceride concentrations among women. The stress score significantly correlated with the value for glycosylated hemoglobin even after controlling for age, body mass index, alcohol consumption, smoking, and exercise among men. An increase in the stress intolerance score was positively associated with an increase in body fat percentage among men, while it was positively associated with an increase in body weight, body mass index, and body fat percentage among women. Our result that the stress score measured using the IMPS was associated with obesity and unfavorable glycemic changes is in congruency with the model that psychosocial stress has a detrimental effect on humans by inducing obesity and insulin resistance, suggesting that the IMPS is a valid means to evaluate psychosocial stress levels among an otherwise healthy population. J Physiol Anthropol 26(2): 149–158, 2007 http://www.jstage.jst.go.jp/browse/jpa2 [DOI: 10.2114/jpa2.26.149]

Keywords: IMPS, psychosocial stress, body fat, obesity, glycosylated hemoglobin

Introduction Stress evokes physical and psychological responses in humans which enable them to restore inner stability through changes. However, it is assumed that if stress is strong or enduring, the ability of humans to adapt to stressful situations may decrease, resulting in a decline in health eventually leading to illness (Chrousos and Gold, 1992; McEwen, 1998). Numerous studies have shown that psychosocial distress is associated with coronary heart disease (Rozanski et al., 1999; Hemingway and Marmot, 1999; Ferketich et al., 2000; Rugulies, 2002; Hintsanen et al., 2005), obesity (Istvan et al., 1992; Ross, 1994; Siegel et al., 2000; Roberts et al., 2000), hypertension (Everson et al., 1997; Ming et al., 2004), insulin resistance (Winokur et al., 1988), and metabolic syndrome (Chrousos, 2000; Räikkönen et al., 2002; Kinder et al., 2004), implying that a certain amount of physical and psychological change accumulates as a result of stress response to psychosocial pressure, which is detrimental to the body. In order to stay healthy in the face of such stressors associated with modern life as social pressures, information overload, and rapid social changes, we must be aware of how much stress we are suffering from, as well as how to manage this stress. Awareness and management of stress may enable successful adaptation to stressful situations. Successful adaptation to stress is assumed to cause proper habituation, resulting in a decline in the stress response. Though it is vital to have a scale to measure psychosocial stress levels in order to promote stress management programs, to our knowledge there are few inventories from which to evaluate what degree of stress response a healthy person is experiencing. In a previous study, we developed the Inventory to Measure Psychosocial Stress (IMPS) and the Inventory to Measure Stress Tolerance

150

Psychosocial Stress and Biomedical Parameters

Capacity (IMST) among healthy subjects (Yamamoto, 2005). The IMPS was designed to assess psychosocial stress defined as the accumulation of stress response by measuring the amount of physical, psychological, and behavioral change caused by psychosocial stress. If the stress score measured using the IMPS is shown to be associated with biomedical parameters of health, universal application of the IMPS as a scale to measure psychosocial stress may be reinforced. With this background, this study was performed to examine the relationship between the IMPS-measured stress score and biomedical parameters, such as anthropometric measurements, blood pressure, liver function tests, serum lipids concentrations, plasma glucose concentrations and the value for glycosylated hemoglobin, among apparently healthy subjects. At the same time, the relationship between the stress intolerance score measured using the IMST and these parameters was investigated in an attempt to clarify the effects on health status of social support, lifestyle habits, and attitude toward life.

Methods Stress score and stress intolerance score The IMPS was developed for use in assessing distress by measuring the amount of stress response such as physical, psychological, and behavioral changes which a person under psychosocial stressors experiences (Yamamoto, 2005). It is known that persons suffering from psychosocial stress may become depressive or experience mood changes. They may also show behavioral changes or suffer from somatic symptoms such as diarrhea, dizziness, headache, and fatigue. We regard both the somatic symptoms and changes in mental status and behavior, which an otherwise healthy person experiencing psychosocial stress exhibits, as stress responses. The 40 items utilized in the IMPS were chosen from as broad a spectrum of known symptoms of psychosocial stress as possible, without exact repetition. The subjects were asked to choose their answer for each of the 40 items of the IMPS from one of: “yes” “rather yes” “rather no” and “no”. “No” and “rather no” were multiplied by zero, “rather yes” by 1 and “yes” by 2. The score was then calculated by adding all of the numbers together, ranging from 0 to 80. We refer to the score measured using the IMPS as the stress score in this study. We regard appropriate social support, a healthy lifestyle, and a positive attitude toward life as elements of stress tolerance capacity which are important for achieving good mental health, and believe that deviation from these elements indicates a decline in the stress tolerance capacity, that is, an increase in stress intolerance. A person with a low stress tolerance capacity is assumed to be more vulnerable to psychosocial stress than one with a high stress tolerance capacity. The 20item IMST was designed to evaluate a decline in the stress tolerance capacity by measuring the extent of deviation from having appropriate social support, a healthy lifestyle, and a positive attitude toward life (Yamamoto, 2005). The subjects

were asked to choose their answer for each of the 20 items of the IMST from one of: “yes” “rather yes” “rather no” and “no”. “Yes” and “rather yes” were multiplied by zero, “rather no” by 1 and “no” by 2. The score was calculated by adding all of the numbers together, ranging from 0 to 40. We refer to the score measured using the IMST as the stress intolerance score in this study.

Subjects and procedure Every year about 5,000 workers chosen randomly from among approximately 136,000 teaching and non-teaching workers employed at public elementary, junior high, and senior high schools in Kyushu and Okinawa, who want to undergo medical check-ups carried out for the purpose of detecting illness at an early stage, are admitted to Kyushu Central Hospital in Fukuoka, Japan, for three days. The subjects of this study came out of a smaller pool of 1,941 workers who were admitted to the hospital between November 2004 and March 2005. On the first day of admission, the subjects were asked to answer a demographic questionnaire which recorded age, gender, and occupation. We also recorded details of each subject’s medical and family history, cigarette smoking, alcohol consumption, physical activity, subjective somatic symptoms, significant morbidity if diagnosed by a physician, and current medication. The subjects were given an explanation of the study and asked for their voluntary participation. The subjects who gave their informed consent were asked the question “Have you been feeling stressed recently?”, then rated their individual subjective stress as 1 (no), 2 (rather no), 3 (rather yes) and 4 (yes), and then completed the IMPS and IMST. A total of 1,499 workers responded to the IMPS and IMST, yielding a response rate of 77.2%. Two hundred and ninety-eight workers who were taking medication for hypertension, hyperuricemia, hyperlipidemia and diabetes, or who had a value for glycosylated hemoglobin 6.0 percent were excluded, and the remaining 1,201 subjects were analyzed. All participants underwent blood sampling after an overnight fast. Anthropometric measurements and urinalysis were conducted under the same conditions. Physical examination, ECG and chest X-ray examination, and vision and hearing function tests were routinely carried out. If necessary, ultrasonographic examinations of the upper abdomen and thyroid gland, or a gastrographic examination were carried out. Height was measured to the nearest 0.001 m, and body weight was measured to the nearest 0.1 kg. Body mass index (BMI) was calculated as the weight in kilograms divided by the square of the height in meters (kg/m2). Body fat percentage was measured by the method of bioelectrical impedance analysis, using an Omron body fat analyzer (HBF302).

Statistical analysis Separate analysis was conducted for the men and women throughout the study. The unpaired t-test was used to test

Yamamoto, K et al.

J Physiol Anthropol, 26: 149–158, 2007

differences between men and women in the means of subjective stress, the stress score, and the stress intolerance score. For the parametric two-group comparisons, p values were based on Welch’s test for groups with unequal variances if the hypothesis of equal variance was rejected at the 0.05 level of significance. Pearson correlation coefficients were used to explore the correlates of the stress score and stress intolerance score to biomedical variables. The subjects were categorized into five groups according to their stress score: the 0–4 group, 5–9 group, 10–19 group, 20–39 group, and 40–80 group. They were also categorized into four groups according to their stress intolerance score: the 0–4 group, 5–9 group, 10–19 group, and 20–40 group. Analysis of covariance (ANCOVA) tested whether there existed any cross-sectional association of the five categories of the stress score and four categories of the stress intolerance score with biomedical variables. Partial correlations were used to assess the correlation of the stress score to body fat percentage, glycosylated hemoglobin, and triglyceride, with variables controlling for age, BMI, alcohol consumption, smoking, and exercise. P values of less than 0.05 were considered significant.

Results Sample characteristics As Table 1 shows, the study subjects consisted of 684 men and 517 women, whose mean (SD) ages were 47.17.9 and 47.57.0 years, respectively. About 92% of the male subjects and 94% of the female subjects were teachers, and the rest were non-teaching workers. Women reported feeling more subjective stress than men (t(1138.6)5.38, p0.001). Women had a higher stress score than men (t(1033.8)6.02, p0.001), while men had a higher stress intolerance score than women (t(1161.1)4.17, p0.001). The lifestyle characteristics of the subjects are shown in Table 1. About 15 % of the male and 38% of the female subjects consumed no alcohol, while more than half of the male and 23% of the female subjects were frequent drinkers. Most of the female subjects had never smoked cigarettes, while 32% of the male subjects were former smokers and 28% current smokers. About half of the male and 68% of the female subjects engaged in no physical activity during the week, while 53% of the male and 32% of the female subjects engaged in physical activity once or more a week.

Stress score and biomedical parameters Table 2 shows the arithmetic means (SD) of the biomedical parameters and the results of ANCOVA testing on the association between parameters and the levels of stress score in the 0–4 group, 5–9 group, 10–19 group, 20–39 group and 40–80 group. The stress score was significantly associated with body fat percentage and the value for glycosylated hemoglobin among men (Table 2). There was a linear relationship between the stress score and body fat percentage or the value of glycosylated hemoglobin (p0.001 and p0.018,

151

Table 1 Sample Characteristics Characteristics Age 30–39 years, % 40–49 years, % 50–59 years, % 60–69 years, % Mean, y (SD) Occupation Teaching worker, % Non-teaching worker, % Subjective stress (SD) Stress score (SD) Stress intolerance score (SD) Alcohol consumption No drinking, % Occasional drinker, % Frequent drinker 360 ml of Sake¶/time, % 360 ml of Sake/time, % Smoking status Never smoked, % Former smoker, % Current smoker 1–19 cigarettes/day, % 20 cigarettes/ day, % Physical activity§ 0 day/week, % 1 day/week, % 2–3 days/ week, % 4–7 days/week, %

Men (N684)

Women (N517)

17.3 42.7 34.4 5.7 47.1 (7.9)

13.3 44.3 38.1 4.3 47.5 (7.0)

92.4 7.6 2.78 (0.93) 12.7 (11.1) 11.5 (5.6)

93.6 6.4 3.07 (0.88) 16.9 (12.6) 10.2 (5.1)

14.8 32.1

38.0 38.9

44.3 8.7

22.3 0.8

39.4 32.2

90.1 4.3

10.0 18.4

5.2 0.4

47.6 19.2 18.8 14.5

68.4 16.2 9.7 5.7

Note 1) ¶ Sake is a brewed alcoholic beverage made from fermented rice which is about 32 proof. § Physical activity is activity other than routine daily activities, which is engaged in for more than 30 minutes per day.

respectively). Figure 1 shows that the value for glycosylated hemoglobin tended to increase with an increase in the stress score among men. However, it showed no significant association with the values for body weight, BMI, WBC, RBC, hemoglobin, hematocrit, C-reactive protein, liver function tests, BUN, creatinine, uric acid, cholesterol, HDL cholesterol, LDL cholesterol, triglyceride, and systolic and diastolic blood pressures among men. A significant association was found between the stress score and triglyceride concentrations among women (Table 2). There was a linear relationship between the stress score and triglyceride concentrations (p0.001). There was no significant association between the stress score and other biomedical measurements among women. As shown in Table 3, an increase in stress score was significantly correlated with an increase in body fat percentage among men, controlling for age and BMI. However, when the correlation was controlled for alcohol consumption and smoking, the significant correlation attenuated, and was nullified when exercise entered into the analysis. The stress score among men remained significantly correlated with the value for glycosylated hemoglobin even after controlling for age, BMI, alcohol consumption, smoking, and exercise (Table 3). The stress score was shown to be significantly correlated

152 Table 2

Psychosocial Stress and Biomedical Parameters Arithmetic Means of Biomedical Parameters and the Relationship between the Stress Score and Parameters among Public School Workers Men (N684) Mean (SD) Height (cm) Body weight (kg) BMI (kg/m2) Body fat (%) WBC (103/m l) RBC (106/m l) Hemoglobin (g/dl) Hematocrit (%) C-reactive protein (mg/dl) Total protein (g/dl) GOT (lU/l) GPT (IU/l) r-GTP (IU/l) BUN (mg/dl) Creatinine (mg/dl) Uric acid (mg/dl) Serum cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Triglyceride (mg/dl) Glycosylated hemoglobin (%) Systolic BP (mm Hg) Diastolic BP (mm Hg)

Women (N517)

F(4, 678)

p

Mean (SD)

0.61 1.18 1.23 3.00 0.55 0.33 0.57 0.87 2.03 0.58 1.23 1.01 0.19 0.42 1.37 0.20 1.40 1.16 0.98 1.55 2.90 0.31 0.44

ns ns ns * ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns * ns ns

155.8 (5.4) 55.4 (7.8) 22.8 (3.2) 29.4 (4.9) 4.82 (1.32) 4.42 (0.32) 13.2 (1.2) 40.2 (3.1) 0.096 (0.248) 7.2 (0.4) 19.9 (7.2) 17.3 (11.5) 23.4 (24.3) 13.5 (3.2) 0.62 (0.09) 4.3 (0.9) 204.8 (34.3) 76.1 (16.3) 126.2 (32.4) 75.1 (36.8) 4.85 (0.34) 117.8 (16.4) 72.5 (10.4)

169.0 (5.6) 70.9 (10.0) 24.8 (3.1) 23.4 (4.3) 5.60 (1.51) 5.04 (0.38) 15.5 (1.0) 46.6 (2.9) 0.103 (0.262) 7.3 (0.4) 25.1 (21.0) 30.0 (32.0) 50.7 (48.7) 15.0 (2.9) 0.88 (0.12) 6.2 (1.2) 205.6 (33.4) 59.5 (14.5) 136.5 (32.4) 128.2 (75.7) 4.92 (0.32) 129.6 (17.2) 79.9 (11.0)

F(4, 511) 1.61 0.61 0.96 0.08 1.34 0.22 0.63 0.19 1.05 0.33 0.66 1.04 1.03 1.69 1.16 0.69 0.63 1.15 0.51 2.84 0.28 0.37 0.81

p ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns * ns ns ns

Note 1) ANCOVA testing on the association between biomedical parameters and the levels of stress score in the 0–4 group, 5–9 group, 10–19 group, 20–39 group and 40–80 group, controlling for age. * p0.05; ** p0.01; ns, not significant.

with triglyceride concentrations among women, controlling for age and BMI, although when it was controlled for alcohol consumption, the significant correlation was nullified.

Stress intolerance score and biomedical parameters Table 4 shows the results of ANCOVA testing on the association between biomedical parameters and the levels of stress intolerance score in the 0–4 group, 5–9 group, 10–19 group and 20–40 group. The stress intolerance score was significantly associated with body fat percentage among men, while it was significantly associated with body weight, BMI, and body fat percentage among women (Table 4). Figure 2 shows that the body fat percentage among men and the BMI among women tended to increase with an increase in the stress intolerance score. There was a linear relationship between the stress intolerance score and body fat percentage among men (p0.015). There was also a linear relationship between the stress intolerance score and body weight or BMI or body fat percentage among women (p0.003, p0.002 and p0.017, respectively). Fig. 1 The relationship between stress score and glycosylated hemoglobin among men. Error bars indicate 95% confidence intervals. P value for glycosylated hemoglobin from an ANCOVA test indicates that the trend from the 0–4 stress score group to the 40–80 group is linear (p0.018).

Discussion The present study tested the associations between the IMPSmeasured stress score and biomedical parameters with regards to the health status of an apparently healthy adult population. The analysis was carried out separately for men and women, as

Yamamoto, K et al.


153

Table 3 Correlations between the Stress Score and Biomedical Parameters Controlling for the Effects of Age, BMI, Alcohol Consumption, Smoking, and Exercise among Public School Workers Men (N684)

Women (N517)

Control Variables Added Sequentially

Stress score Age Stress score Age BMI Stress score Age and BMI Alcohol consumption Stress score Age, BMI and alcohol consumption Smoking Stress score Age, BMI, alcohol consumption and smoking Exercise

Body fata

Glycosylated hemoglobina

Triglyceridea

0.084 (p0.028)

0.092 (p0.017)

0.123 (p0.005)

0.078 (p0.041)

0.085 (p0.027)

0.114 (p0.009)

0.075 (p0.051)

0.086 (p0.037)

0.067 (p0.132)

0.074 (p0.057)

0.082 (p0.048)

0.065 (p0.145)

0.057 (p0.144)

0.092 (p0.018)

0.048 (p0.284)

Note 1) a Partial correlation coefficient controlling for the variables indicated.

Table 4 Relationship between the Stress Intolerance Score and Biomedical Parameters among Public School Workers Men (N684)

Height (cm) Body weight (kg) BMI (kg/m2) Body fat (%) WBC (103/m l) RBC (106/m l) Hemoglobin (g/dl) Hematocrit (%) C-reactive protein (mg/dl) Total protein (g/dl) GOT (lU/l) GPT (IU/l) r-GTP (IU/l) BUN (mg/dl) Creatinine (mg/dl) Uric acid (mg/dl) Serum cholesterol (mg/dl) HDL cholesterol (mg/dl) LDL cholesterol (mg/dl) Triglyceride (mg/dl) Glycosylated hemoglobin (%) Systolic BP (mm Hg) Diastolic BP (mm Hg)

Women (N517)

F(3, 679)

p

F(3, 512)

p

1.42 1.23 2.45 2.90 2.08 0.24 0.32 0.85 1.29 1.17 0.35 1.11 1.83 0.43 1.51 1.02 0.64 1.47 0.36 1.43 1.65 0.97 1.31

ns ns ns * ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

0.46 3.66 4.56 3.52 2.25 0.41 1.54 1.21 0.57 0.08 0.75 1.08 0.29 0.53 0.63 0.99 0.61 1.45 0.41 0.73 1.03 1.52 1.62

ns * ** * ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns ns

Note 1) ANCOVA testing on the association between biomedical parameters and the levels of stress intolerance score in the 0–4 group, 5–9 group, 10–19 group and 20–40 group, controlling for age. * p0.05; ** p0.01; ns, not significant.

there are significant differences in the structure of psychosocial stress and the stress tolerance capacity between males and females (Frankenhaeuser et al., 1978; Yamamoto, 2004). Our

results show that there was a significant association between the stress score measured using the IMPS and biomedical parameters such as body fat percentage and glycosylated hemoglobin among men. By contrast, a significant association was found between the stress score and triglyceride concentrations among women. This study is the first to demonstrate an association between psychosocial stress defined as the accumulation of stress response and biomedical parameters among apparently healthy subjects. The increase in obesity has been reported as a global phenomenon both in developing and developed countries (Popkin and Doak, 1998; Rössner, 2002). The mean body mass index was found to have increased from 25.3 to 26.3 for adult men and women aged 20 through 74 years between the 1976 to 1980 and 1988 to 1991 surveys in the USA (Kuczmarski et al., 1994). In Japan, an increasing trend in obesity among men was observed between the 1976 to 1980 and 1996 to 2000 surveys (Yoshiike et al., 2002). An increasingly sedentary lifestyle and dietary energy consumed in excess of metabolic expenditure are regarded as the most important factors associated with the worldwide trend of increases in overweight and obese persons (Popkin and Doak, 1998; Rössner, 2002). Conjunctively, it has been shown that depression and psychological distress are associated with obesity (Istvan et al., 1992; Ross, 1994; Rosmond et al., 1996; Roberts et al., 2000), and stress seems to be associated more significantly with abdominal obesity than with generalized obesity (Räikkönen et al., 1994a). Our results, which show that an increase in stress score was not associated with an increase in BMI and body weight, but was associated with an increase in body fat percentage among men, may parallel this finding that stress may be associated more significantly with abdominal obesity than with generalized obesity, inasmuch as an increase in body fat percentage may suggest, though not definitively, the presence of abdominal

154


Fig. 2 The relationship between stress intolerance score and body fat percentage among men (A) and BMI among women (B). Error bars indicate 95% confidence intervals. P values for body fat percentage among men and BMI among women from an ANCOVA test indicate that the trend from the 0–4 stress intolerance score group to the 20–40 group is linear (p0.015 and p0.002, respectively).

obesity. We found that the positive correlation between stress score and body fat percentage among men attenuated when these correlations were controlled for alcohol consumption and smoking, and lost significance when controlled for exercise, suggesting that lifestyle changes associated with an increase in psychosocial stress may be a potential pathway for the association between stress and body fat percentage. In contrast to body fat percentage, we found that the correlation between stress score and the value for glycosylated hemoglobin remained significant even after controlling for age, BMI, alcohol consumption, smoking, and exercise among men. These results indicate that the relationship between glycosylated hemoglobin and stress is not affected by changes in lifestyle. It is recognized that psychosocial stress sets in motion central and peripheral stress responses such as the perturbation of the regulation of the hypothalamic-pituitaryadrenal (HPA) axis and the activation of the sympathetic nervous system (Chrousos and Gold, 1992; McEwen, 1998, Chrousos, 2000). An increase in cortisol secretion induced by the perturbation of the regulation of the HPA axis is regarded to be primarily responsible for abdominal obesity (Rosmond et al., 1998), which is one of the main symptoms of metabolic syndrome (Ford et al., 2002). In addition, because insulin resistance is known to coexist both in metabolic syndrome and in related comorbidities such as abdominal obesity, hyperinsulinemia, dyslipidemia and hypertension, which are considered to be associated with stress (Räikkönen et al., 1996; Liese et al., 1997), it is reasonable to hypothesize that the level of glycosylated hemoglobin will increase in conjunction with an increase in stress response. In fact, psychosocial stress defined as feelings of excessive tiredness and psychological coronary risk factors such as hostile paranoia and vital exhaustion have been reported to cause perturbations of insulin and glucose metabolism (Räikkönen et al., 1994b, 1996;

McEwen, 1998). Our finding that stress score measured using the IMPS was positively correlated with an increase in glycosylated hemoglobin regardless of lifestyle changes among men seems to be consistent with the model that the perturbation of the regulation of the HPA axis along with the activation of the sympathetic nervous system caused by psychosocial stress is responsible for the unfavorable glycemic changes. However, we did find an association between the stress score and triglyceride concentrations among women, though no association was shown between the stress score and body fat percentage or glycosylated hemoglobin. The fact that there are differences in the structure of stress and the stress tolerance capacity between men and women (Frankenhaeuser et al., 1978; Yamamoto, 2004) may explain the different results for men and women in the present study. The positive correlation between the stress score and triglyceride concentrations was found to have lost its significance when it was controlled for alcohol consumption, suggesting that lifestyle changes associated with an increase in psychosocial stress may be a pathway for the association between stress and triglyceride concentrations among women. As hypertriglyceridemia is regarded to be a factor which features in a metabolic syndrome (Ford et al., 2002), our results suggest that psychosocial stress may exert a detrimental effect on the health status of women through dyslipidemia. It has been recognized that lifestyle changes associated with modernization are the primary causes of an increase in obesity and diabetes (Popkin and Doak, 1998; Rössner, 2002). In the present study, we assumed that having appropriate social support, a healthy lifestyle, and a positive attitude toward life were elements of the stress tolerance capacity, and measured the stress tolerance capacity of our subjects in order to elucidate whether deviation from these elements has an effect

Yamamoto, K et al.


on their health status. We found that an increase in the stress intolerance score measured using the IMST was associated with an increase in body weight, BMI, and body fat percentage among women, and with an increase in body fat percentage among men, indicating that the less social support people have, the unhealthier their lifestyle, and all the more negative an attitude toward life they have, the more likely it is that they will be obese. A healthier lifestyle which includes physical exercise, smoking cessation, the restriction of alcohol consumption, and dieting has been shown to result in the prevention of type 2 diabetes among subjects with impaired glucose tolerance (Pan et al., 1997; Tuomilehto et al., 2001; Hu et al., 2001). Our findings suggest that an increase in the stress tolerance capacity has beneficial effects on health status through the prevention of an increase in body fat and obesity. We acknowledge several limitations of our study. First, the measurements of the stress score and stress intolerance score were self-reported, and are therefore subject to bias. The fact that the stress score of workers was lower than that of students (Yamamoto, 2005) suggests that overreporting has been avoided. The fact that the stress intolerance score of men was higher than that of students (Yamamoto, 2005) is interpreted as reasonable given that male workers who are busy with their jobs have less time to manage their lives than students. Second, as psychosocial stress was defined as the accumulation of stress response such as a deterioration in mental status, somatic symptoms, and behavioral changes in relation to stress, no evidence of an association between the stress score and physiological parameters of stress was shown. However, our results are in line with the reports that abdominal obesity and related comorbidities such as insulin resistance and dyslipidemia are associated with psychosocial stress (Räikkönen et al., 1996; Björntorp, 2001). Third, these data are cross-sectional, and therefore determine no causal relationship between the stress score measured using the IMPS and biomedical parameters regarding health status. In summary, we demonstrated that psychosocial stress defined as the accumulation of stress response was associated with an increase in body fat percentage and the value for glycosylated hemoglobin among men, while it was associated with an increase in triglyceride concentrations among women. We also found that a decline in the stress tolerance capacity defined as deviation from the assumed ideal lifestyle which a person should adopt in order to not be vulnerable to stress was associated with obesity, implying that an unhealthy lifestyle may lead to the accumulation of body fat through inappropriate health behaviors such as overeating, profuse alcohol consumption, and lack of physical activity. Our result that the stress score measured using the IMPS was associated with obesity and unfavorable glycemic changes is in parallel with the model that psychosocial stress has a detrimental effect on humans by inducing obesity and insulin resistance, suggesting that the IMPS is valid in evaluating psychosocial stress level among an otherwise healthy population.

155

Acknowledgements This study was supported by Grantin-Aid for Scientific Research No. 16207018 from the Japan Society for the Promotion of Science.

Appendix The questions used in this questionnaire are designed to help you to evaluate the degree of stress you are suffering from and how much stress tolerance capacity you have. Please select your answers from “yes”, “rather yes”, “rather no”, and “no”. Have you been feeling stressed recently? (yes rather no no)

rather yes

The Inventory to Measure Psychosocial Stress (IMPS) 1a. I have difficulty falling asleep. (yes rather yes rather no no) 1b. I awake several times while sleeping. (yes rather yes rather no no) 1c. I awake early and cannot fall asleep again. (yes rather yes rather no no) 2. I feel my sexual drive is declining. (yes rather yes rather no no) 3. I cannot sit still without fidgeting. (yes rather yes rather no no) 4. I get headaches without reason. (yes rather yes rather no no) 5. I get muscle cramps in my arms or legs. (yes rather yes rather no no) 6. I feel tired without reason. (yes rather yes rather no no) 7. I suffer from stomach trouble without reason. (yes rather yes rather no no) 8. I am prone to overeating without reason. (yes rather yes rather no no) 9. I lose my appetite without reason. (yes rather yes rather no no) 10. I get diarrhea or constipation without reason. (yes rather yes rather no no) 11. I get cold sweats without reason. (yes rather yes rather no no) 12. I have facial twitching. (yes rather yes rather no no) 13. I feel nauseous without reason. (yes rather yes rather no no) 14. I feel dizzy without reason. (yes rather yes rather no no) 15. I feel breathlessness or chest discomfort without reason. (yes rather yes rather no no) 16. I feel like crying without reason. (yes rather yes rather no no) 17. I do not enjoy my life. (yes rather yes rather no no) 18. I am impotent to solve the problems in my life. (yes rather yes rather no no) 19. I feel irritated without reason. (yes rather yes rather no no)

156


20. I am concerned about my health. (yes rather yes rather no no) 21. I think of my life as a failure. (yes rather yes rather no no) 22. I dislike myself. (yes rather yes rather no no) 23. I have little interest in other people. (yes rather yes rather no no) 24. I conceal my feelings. (yes rather yes rather no no) 25. I feel resentment or hatred toward other people. (yes rather yes rather no no) 26. I feel I am losing my youth or beauty. (yes rather yes rather no no) 27. I cannot relax. (yes rather yes rather no no) 28. Other people are hostile toward me. (yes rather yes rather no no) 29. Other people ignore me. (yes rather yes rather no no) 30. Other people disrespect me. (yes rather yes rather no no) 31. I do not play a functional role in my family. (yes rather yes rather no no) 32. I am concerned about my future. (yes rather yes rather no no) 33. Other people do not accept or understand me. (yes rather yes rather no no) 34. I feel isolated. (yes rather yes rather no no) 35. I have difficulty concentrating on what I am doing. (yes rather yes rather no no) 36. I rush on to the next task before finishing the first one. (yes rather yes rather no no) 37. I dislike seeing my acquaintances or other people. (yes rather yes rather no no) 38. I feel discomfort when I accidentally touch other people. (yes rather yes rather no no) Stress score How many “yes” answers did you choose? ( How many “rather yes” answers did you choose? ( How many “rather no” answers did you choose? ( How many “no” answers did you choose? ( Stress score

(

)2( ) )1( ) )00 )00

)

The Inventory to Measure Stress Tolerance Capacity (IMST) 1. I have a hot, balanced meal at least once a day. (yes rather yes rather no no) 2. I sleep 7–8 hours at least four days a week. (yes rather yes rather no no) 3. I give and receive affection regularly. (yes rather yes rather no no) 4. I have a reliable relative within one hour from me. (yes rather yes rather no no) 5. I perspire in physical exercise at least twice a week. (yes rather yes rather no no) 6. I do not smoke, or smoke 10 cigarettes or less a day. (yes rather yes rather no no)

7. I do not drink alcohol, or drink three times or less a week. (yes rather yes rather no no) 8. My weight is proportionate to my height. (yes rather yes rather no no) 9. I have enough income to support myself. (yes rather yes rather no no) 10. Religious belief gives me psychological strength. (yes rather yes rather no no) 11. I regularly participate in social activities. (yes rather yes rather no no) 12. I have a network of friends and acquaintances. (yes rather yes rather no no) 13. I have at least one friend to talk with about personal matters. (yes rather yes rather no no) 14. I am in good health, including eyesight, hearing, and teeth. (yes rather yes rather no no) 15. I express my feelings openly when angry or worried. (yes rather yes rather no no) 16. I talk about family matters with my cohabitants. (yes rather yes rather no no) 17. I do my favorite pastime once a week. (yes rather yes rather no no) 18. I am able to organize my time effectively. (yes rather yes rather no no) 19. I do not drink caffeinated beverages, or drink three cups or less per day. (yes rather yes rather no no) 20. I have quiet time reserved for myself once a day. (yes rather yes rather no no) Stress intolerance score How many “yes” answers did you choose? ( How many “rather yes” answers did you choose? ( How many “rather no” answers did you choose? ( How many “no” answers did you choose? ( Stress intolerance score

(

)00 )00 )1( ) )2( )

)

References Björntorp P (2001) Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2: 73–86 Chrousos GP (2000) The role of stress and the hypothalamicpituitary-adrenal axis in the pathogenesis of the metabolic syndrome: neuro-endocrine and target tissue-related causes. Int J Obes 24 (suppl 2): S50–S55 Chrousos GP, Gold PW (1992) The concepts of stress and stress system disorders. JAMA 267: 1244–1252 Everson SA, Lynch JW, Chesney MA, Kaplan GA, Goldberg DE, Shade SB, Cohen RD, Salonen R, Salonen JT (1997) Interaction of workplace demands and cardiovascular reactivity in progression of carotid atherosclerosis: population based study. BMJ 314: 553–558 Ferketich AK, Schwartzbaum JA, Frid DJ, Moeschberger ML (2000) Depression as an antecedent to heart disease among women and men in the NHANES I study. Arch Intern Med

Yamamoto, K et al.


160: 1261–1268 Ford ES, Giles WH, Dietz WH (2002) Prevalence of the metabolic syndrome among US adults. JAMA 287: 356–359 Frankenhaeuser M, von Wright MR, Collins A, von Wright J, Sedvall G, Swahn C-G (1978) Sex differences in psychoneuroendocrine reactions to examination stress. Psychosom Med 40: 334–343 Hemingway H, Marmot M (1999) Psychosocial factors in the aetiology and prognosis of coronary heart disease: systematic review of prospective cohort studies. BMJ 318: 1460–1467 Hintsanen M, Kivimäki M, Elovainio M, Pulkki-Råback L, Keskivaara P, Juonala M, Raitakari OT, KeltikangasJärvinen L (2005) Job strain and early atherosclerosis: the cardiovascular risk in young Finns study. Psychosom Med 67: 740–747 Hu FB, Manson JE, Stampfer MJ, Colditz G, Liu S, Solomon CG, Willett WC (2001) Diet, lifestyle, and the risk of type 2 diabetes mellitus in women. N Engl J Med 345: 790–797 Istvan J, Zavela K, Weidner G (1992) Body weight and psychological distress in NHANES I. Int J Obes 16: 999–1003 Kinder LS, Carnethon MR, Palaniappan LP, King AC, Fortmann SP (2004) Depression and the metabolic syndrome in young adults: findings from the Third National Health and Nutrition Examination Survey. Psychosom Med 66: 316–322 Kuczmarski RJ, Flegal KM, Campbell SM, Johnson CL (1994) Increasing prevalence of overweight among US adults. JAMA 272: 205–211 Liese AD, Mayer-Davis EJ, Tyroler HA, Davis CE, Keil U, Duncan BB, Heiss G (1997) Development of the multiple metabolic syndrome in the ARIC cohort: joint contribution of insulin, BMI, and WHR. Ann Epidemiol 7: 407–416 McEwen BS (1998) Protective and damaging effects of stress mediators. N Engl J Med 338: 171–179 Ming EE, Adler GK, Kessler RC Fogg LF, Matthews KA, Herd JA, Rose RM (2004) Cardiovascular reactivity to work stress predicts subsequent onset of hypertension: the Air Traffic Controller Health Change Study. Psychosom Med 66: 459–465 Pan XR, Li GW, Hu YH, Wang JX, Yang WY, An ZX, Hu ZX, Lin J, Xiao JZ, Cao HB, Liu PA, Jiang XG, Jiang YY, Wang JP, Zheng H, Zhang H, Bennett PH, Howard BV (1997) Effects of diet and exercise in preventing NIDDM in people with impaired glucose tolerance. Diabetes Care 20: 537–544 Popkin BM, Doak CM (1998) The obesity epidemic is a worldwide phenomenon. Nutr Rev 56: 106–114 Räikkönen K, Hautanen A, Keltikangas-Järvinen L (1994a) Association of stress and depression with regional fat distribution in healthy middle-aged men. J Behav Med 17: 605–616 Räikkönen K, Keltikangas-Järvinen L, Adlercreutz H, Hautanen A (1996) Psychosocial stress and the insulin resistance syndrome. Metabolism 45: 1533–1538

157

Räikkönen K, Keltikangas-Järvinen L, Hautanen A (1994b) The role of psychological coronary risk factors in insulin and glucose metabolism. J Psychosom Res 38: 705–713 Räikkönen K, Matthews KA, Kuller LH (2002) The relationship between psychological risk attributes and the metabolic syndrome in healthy women: antecedent or consequence? Metabolism 51: 1573–1577 Roberts RE, Kaplan GA, Shema SJ, Strawbridge WJ (2000) Are the obese at greater risk for depression? Am J Epidemiol 152: 163–170 Rosmond R, Dallman MF, Björntorp P (1998) Stress-related cortisol secretion in men: relationships with abdominal obesity and endocrine, metabolic and hemodynamic abnormalities. J Clin Endocrinol Metab 83: 1853–1859 Rosmond R, Lapidus L, Mårin P, Björntorp P (1996) Mental distress, obesity and body fat distribution in middle-aged men. Obes Res 4: 245–252 Ross CE (1994) Overweight and depression. J Health Soc Behav 35: 63–79 Rössner S (2002) Obesity: the disease of the twenty-first century. Int J Obes 26 (suppl 4): S2–S4 Rozanski A, Blumenthal JA, Kaplan J (1999) Impact of psychological factors on the pathogenesis of cardiovascular disease and implications for therapy. Circulation 99: 2192–2217 Rugulies R (2002) Depression as a predictor for coronary heart disease: a review and meta-analysis. Am J Prev Med 23: 51–61 Siegel JM, Yancey AK, McCarthy WJ (2000) Overweight and depressive symptoms among African-American women. Prev Med 31: 232–240 Tuomilehto J, Lindström J, Eriksson JG, Valle TT, Hämäläinen H, Ilanne-Parikka P, Keinänen-Kiukaanniemi S, Laakso M, Louheranta A, Rastas M, Salminen V, Uusitupa M (2001) Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N Engl J Med 344: 1343–1350 Winokur A, Maislin G, Phillips JL, Amsterdam JD (1988) Insulin resistance after oral glucose tolerance testing in patients with major depression. Am J Psychiatry 145: 325–330 Yamamoto K (2004) Gender differences of psychosocial stress and stress tolerance among Japanese college students. Jpn J Physiol Anthropol 9 (suppl 2): 84–85 [In Japanese] Yamamoto K (2005) Development of the inventories to measure psychosocial stress and stress tolerance. Jpn J Physiol Anthropol 10: 67–77 [In Japanese] Yoshiike N, Kaneda F, Takimoto H (2002) Epidemiology of obesity and public health strategies for its control in Japan. Asia Pac J Clin Nutr 11 (suppl): S727–S731 This article was presented at the 8th International Congress of Physiological Anthropology, 2006 (ICPA 2006), in Kamakura, Japan. Received: September 22, 2006

158


Accepted: November 7, 2006 Correspondence to: Kazuhiko Yamamoto MD, PhD, Institute of Health Science, Kyushu University, Shiobaru, Minami-ku, Fukuoka, 815–8540 Japan

Phone: 81 80 3963 7881 Fax: 81 92 553 4424 e-mail: [email protected]