Effect of a worksite-based intervention program on metabolic parameters in middle-aged male white-collar workers: A randomized controlled trial

Effect of a worksite-based intervention program on metabolic parameters in middle-aged male white-collar workers: A randomized controlled trial

Preventive Medicine 51 (2010) 11–17 Contents lists available at ScienceDirect Preventive Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i...

328KB Sizes 3 Downloads 18 Views

Preventive Medicine 51 (2010) 11–17

Contents lists available at ScienceDirect

Preventive Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y p m e d

Effect of a worksite-based intervention program on metabolic parameters in middle-aged male white-collar workers: A randomized controlled trial Chizuko Maruyama a,⁎, Mika Kimura b, Hisashi Okumura c, Kenji Hayashi c, Takashi Arao d a

Department of Food and Nutrition, Japan Women's University, 2-8-1, Mejirodai, Bunkyo-ku, Tokyo, 112-8681, Tokyo, Japan Center for Health Promotion, International Life Sciences Institute Japan, Tokyo, Japan Health Value Added Foods Business Division, Nichirei Foods Inc., Tokyo, Japan d School of Sports Science, Waseda University, Saitama, Japan b c

a r t i c l e

i n f o

Available online 18 April 2010 Keywords: Workplace Insulin resistance Risk factor Individual counseling Dietary habits Internet Lifestyle modification

a b s t r a c t Objective. An effective program for preventing metabolic diseases through lifestyle modification is urgently needed. We investigated the effects of the Life Style Modification Program for Physical Activity and Nutrition program (LiSM10!®) on metabolic parameters in middle-aged male Japanese white-collar workers. Methods. One hundred and one male office workers, 30 to 59 years of age, with metabolic syndrome risk factors, were randomly allocated into no-treatment control (n = 49) and LiSM intervention (n = 52) groups. The LiSM group attended individualized assessment and collaborative goal setting sessions based on food group intake and physical activity, followed by two individual counseling sessions with a registered dietitian and physical trainer, and received monthly website advice during the 4-month period from December 2006 to May 2007, in Tokyo, Japan. They were encouraged to enter current targeted food intakes and pedometer data on self-monitoring websites during the entire study period. Results. Habitual food group intakes changed significantly in the LiSM group, showing improvements in 14 anthropometric and biochemical parameters contributing to inter-group differences in body weight, body mass index, fasting plasma glucose, insulin and homeostasis model assessment of insulin resistance changes (p b 0.01). Conclusion. The LiSM10!® program effectively improved insulin resistance-related metabolic parameters in middle-aged male white-collar workers. © 2010 Elsevier Inc. All rights reserved.

Introduction Improvements in lifestyle habits focusing on physical activity and nutrient intake can reportedly normalize disorders of energy metabolism (Yamaoka and Tango, 2005; Lichtenstein et al., 2006; Teramoto et al., 2007). Metabolic syndrome (MetS) is widely regarded as targets for preventing the early stages of coronary artery disease (Lakka et al., 2002; Zimmet et al., 2005). Development of a simple yet effective program to prevent MetS through continuous lifestyle modification is urgently needed. People should be encouraged to maintain desirable lifestyle habits and become self-reliant over the long-term (Wing and Phelan, 2005). Intervention programs that include health behavior theories (Prochaska et al., 1992; Johnson et al., 2008), social cognitive theory (Bandura, 2001), and other theories based on behavioral sciences (Kaplan et al., 1977), have been suggested to be effective (Fujii and Muto, 2009). However, few studies have examined the results of

⁎ Corresponding author. Fax: +81 3 5981 3428. E-mail address: [email protected] (C. Maruyama). 0091-7435/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.ypmed.2010.04.008

health promotion programs in Japanese white-collar workers, prompting us to design this randomized controlled trial. We previously reported the effectiveness of a worksite health promotion program, the Life Style Modification Program for Physical Activity and Nutrition (LiSM10!®), on metabolic parameters in factory employees (Arao et al., 2007). The LiSM10!® is composed of individual structured counseling sessions, as well as social and environmental approaches. Looking back on comprehensive evaluation of the program, there are problems requiring resolution before this program can be effectively applied to white-collar workers. White-collar workers, whose lifestyles are more irregular because of their long commutes, long working hours, skipping meals, and dining with colleagues than those of subjects in other professions, reportedly have more physical symptoms (Kawada and Suzuki, 2008). Particularly with white-collar workers who eat out more frequently and are rather sedentary as compared to factory employees (The National Health and Nutrition Survey in Japan, 2005), opportunities to benefit from environmental support in the company cafeteria may be lacking. In terms of the counseling sessions, the subjects often find visits inconvenient due to their irregular job schedules. In the last decade, web/computer-based health education programs, tailored to individuals, have been developed and their effectiveness in modifying

12

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17

habitual dietary intake and physical activity has been reported (Brug et al., 2003; Portnoy et al., 2008). As personal computers are popular among white-collar workers in Japan, it was anticipated that using a personal computer might facilitate supporting ongoing lifestyle improvements. This study was designed to investigate the effectiveness of a worksite-based LiSM10!® program on metabolic parameters in middle-aged male Japanese white-collar workers requiring health guidance based on regular health check-up results. We report herein the results of the first period of a randomized, cross-over study. Methods Study design and subjects The study period was the first 4 months of study with a randomized crossover design. Sample size was calculated to detect the intervention effect of a 10% change within the group and between groups, using 0.05 for the alpha and 0.20 for the beta error. The necessary sample size was 45 subjects in each group. The protocol was approved by the Health Science Research Ethics Committee of Showa Women's University. Among office workers belonging to the health insurance association of the Nichirei Group Corporation in Tokyo and its surrounding area, aged from 30 to 59 years, those with MetS risk factors based on the results of regular health check-ups were enrolled in this study. Eight hundred male employees were informed about the study, 319 individuals agreed using their health check-up data and interested in participating in the study. After a detailed explanation of the program, 115 subjects agreed to participate. A subject was considered to be at risk of developing MetS if one or more abnormalities involving serum lipids, glucose levels and blood pressure were present, with visceral obesity (umbilical circumference: 85 cm or more) and/ or BMI ≧ 25. The following were considered to be abnormal: triglyceride (TG) ≧ 150 mg/dL and/or HDL-cholesterol (HDL-C) b 40 mg/dL, systolic blood pressure ≧ 130 mmHg and/or diastolic blood pressure ≧ 85 mmHg, fasting glucose ≧ 110 mg/dL and/or HbA1c ≧ 5.5%. Data collection The medical check-ups were conducted by the Tokyo Health Service Association (Shinjuku-ku, Tokyo), not involved in the study. The data from regular health check-ups conducted in December 2006 and May 2007 were used as baseline and post-completion data, respectively. A randomization code with equal numbers of alternative groups was generated from a list of all participants, using software SPSS (ver.15) at Waseda University. The Nichirei Inc staff members managing the study and contacting participants were not involved in this randomization process. However, as the participants received detailed explanations of the objectives and other aspects of this study, blinding to group assignments was not possible. Lifestyle data were collected at baseline in January and at completion in May 2007. All subjects were asked to answer a questionnaire on lifestyle, habitual food intake (Supplement 1), the stages and self-efficacies of changes in their habitual food intakes and efforts to increase physical activity. Subjects were given a pedometer (Walking style HJ-7101T Omron Health Care Co., Ltd. Japan) to count the number of steps in a week. Primary outcome measures were changes in food group intake and increased number of steps. Secondary outcome measures were anthropometric and biochemical parameters. Intervention program The LiSM10!® program was designed to promote healthy dietary habits and physical activity. Participants had monthly individual contact with a well trained dietitian and a physical trainer, both certified health counselors for this program. Just after the baseline data collection, participants attended an individual goal and action planning session, and at 1 and 2 months, they reviewed their plans with counselors. The fourth counseling session, at the end of the third month, was conducted through the website. The subjects were encouraged to use the website personal page of the LiSM10!® (Nichirei Foods Inc, Tokyo), and were required to enter their current weight, record their practices as regards targeted food intake and physical activity, upload

data from the computer-linkable pedometer and discuss awareness of their lifestyles for self-monitoring throughout the intervention period. The data obtained were automatically presented in figures on their individual website pages. To support their efforts, the subject's family members and the counselor could make comments and/or note their impressions of the data on the self-monitoring page. First goal setting session For the intervention group, individual counseling sessions were provided by a registered dietitian and physical trainer for 20 and 10 min, respectively. First, the dietitian encouraged assessment of the subjects' health check-up results and setting clinical goals to be met by the end of the 4 months. Then, participants assessed their own habitual food group intakes using the food frequency questionnaire, i.e. the “Check your dietary habits” sheet (Supplement 1). The sheet consisted of two major food groups and subgroups (group A: foods recommended to be increased: 5 subgroups: fish, soybeans/soybeanproducts, green/deep-yellow vegetables, white-vegetables, and mushrooms/ seaweed/konnyaku) (group B: foods recommended to be decreased: 11 subgroups: large servings of grains such as rice/bread/noodles, confectionaries, sweet-drinks, fatty-meats, meat-products, butter/margarine/dressing/ mayonnaise, eggs/liver, fried-dishes, pickles, soup, and alcoholic-drinks). The dietitian explained the significance and effects expected from consuming each A group food and avoidance or reduction of each B group food. Then, the subjects were encouraged to set their action plans to change dietary behaviors by increasing the consumption of 1 or 2 foods in the A group, and decreasing the consumption of 1 or 2 foods in the B group. The numbers of target food subgroups, which they wanted to change consumption of, were recommended by the dietitian according to their stages of change and self-efficacy. If the stage of change for food intake was precontemplation, they were advised to count the number of Japanese dishes consumed in a day, as low fat traditional Japanese dishes could be regarded as healthy food models. Each subject was required to record his personal goals and action plan on a commitment sheet. Next, with advice from the physical trainer, subjects assessed their recorded steps for seven days, and decided on an action plan based on how many steps had been taken, or on lifestyle changes aimed at increasing physical activity. Each subject was also required to record his personal action plan concerning physical activity on a commitment sheet. Follow-up counseling The participants were encouraged to enter their current targeted food intake and pedometer data on a website for self-monitoring during the entire study period. At the end of the first and then the second month, the counselors supported the participant in reviewing the month's achievements, to evaluate the level of achievement of his action plan based on the subject's own self-monitoring records, then encouraged him to consider the reason for the results and possible ways to more effectively implement or revise the plan. These face-to-face sessions each lasted 10 min. The participants received website advice after the personal counseling. At the end of the third month, the participants reported their conditions on their website pages, and the counselors both entered comments and advice, for example: answer his questions, acknowledge and praise what he was able to change, or advise him to lower the level and/or change the target if he was not able to change, remind him of behaviors he could to do with his family, let him think over the plan to change without effort if he felt stressed, and so on. Anthropometric measurements Body mass index (BMI) was calculated as weight(kg)/height(m)2. Umbilical circumference was measured during the late exhalation phase in the standing position. Blood pressure was measured using an automatic blood pressure manometer with the subject in the seated position. Blood sampling and analysis Fasting blood samples were obtained and blinded measurements were conducted in the laboratory of the Tokyo Health Service Association as follows (total cholesterol (TC) and TG, enzymatic method; HDL-C, direct method; LDL-cholesterol (LDL-C), Friedewald equation; Aspartate aminotransferase (AST), alanine aminotransferase (ALT) and gamma-glutamyl

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17

transferase (γ-GTP), UV and L-γ-glutamyl-3-carboxy-4-nitroanilide substrate methods; Uric acid, uricase method; PG, hexokinase-UV method; HbA1c, enzymatic method; insulin, chemiluminescence immunoassay; the homeostasis model assessment for insulin resistance (HOMA-IR), calculated as PG (mg/dL) × insulin (IRI) (μU/L) ÷ 405). Statistical analysis

13

in the control and four in the intervention group failed to attend the final outcome measurements. Thus, analyses were performed using the data of 47 and 52 subjects at baseline, and 39 and 48 at completion in the control and intervention groups, respectively (Fig. 1). The recruitment period was December 2006, and the blood and physical data were collected before (December 2006) and after the 4-month intervention (May, 2007).

The statistical analyses were performed using the SPSS12.0J software (SPSS Inc. Japan). Cronbach's α was calculated to check the internal consistency and reliability of the questionnaire, i.e. the “Check your dietary habits” sheet and habitual consumptions of food groups on the sheet were each given scores from zero to 5. Between-group comparisons at baseline were made using the Mann–Whitney U test for continuous data and Chisquared test for proportional data. The difference from baseline until the end of the study period for each group was examined using the paired t test. The difference in the program's effectiveness was examined as an inter-group difference in intra-group change, using repeated-measures analysis of variance. p values b 0.05 were considered significant.

The baseline clinical characteristics of the two groups were similar (Table 1). Habitual food intake scores for food groups A and B, number of steps in a day, and the mean values of self-efficacy for practicing healthy dietary and exercise habits at baseline did not differ between the two groups (Table 2). Cronbach's α for the “Check your dietary habits” questionnaire was 0.83 for 5 food group A items and 0.55 for 11 food group B items, for all subjects (n = 99). The distributions of the stages of change for each behavior did not differ between the two groups.

Results

Action plan

Trial profile and participant flow

For the dietary action plans aimed at food group A, 1, 2 and 3 items were selected to be increased by 20 (38.5 %), 24 (46.2 %) and 3 (5.8 %) subjects, respectively. The top items were white-vegetables, green/ deep-yellow vegetables and mushrooms/seaweed/konnyaku. As for food group B, 1, 2, 3 and 4 items were selected to be decreased by 20 (38.5%), 19 (36.5%), 3 (5.8%) and 2 (3.8%) subjects, respectively. The top items were confectionaries, alcoholic-drinks, sweet-drinks, large servings of grain and butter/margarine/dressing/mayonnaise. All

From the 115 enrolled subjects in the study, 14 receiving medical treatments were excluded. Thus, 101 participants provided written informed consent, and underwent baseline measurements. Two subjects suspected to have a lipoprotein lipase deficiency and/ or abnormal apoE in the control group, based on extremely high TG levels (N800 mg/dL), were excluded from the statistical analysis. Eight

Baseline equivalence between the two study groups

Fig. 1. Enrollment and retention. In total, 101 subjects 30 to 59 years of age, with MetS risk factors, were randomly allocated into no-treatment control (n = 49) or LiSM intervention (n = 52) groups. In the control group, 2 subjects with hyperlipidemia type III,V were excluded. Tokyo, Japan, December 2006–May 2007.

14

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17

Table 1 Clinical characteristics of control and LiSM groups at baseline.

Number Age Height Weight (kg) Body mass index Umbilical circumference (cm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Aspartate aminotransferase (IU/L) Alanine aminotransferase (IU/L) γ-glutamyl transferase (IU/L) Uric acid (mg/dL) Hemoglobin A1c (%) Fasting plasma glucose (mg/dL) Fasting insulin (μU/L) HOMA-IRc Total Cholesterol (mg/dL) LDL-cholesterol (mg/dL) HDL-cholesterol (mg/dL) Triglycerides (mg/dL)

Control

LiSMa

p

47 35.5 (8.1)a 171.3 (5.8) 75.8 (9.9) 25.8 (3.3) 90.4 (8.2) 127 (15) 81 (11) 27 (9) 35 (21) 80 (104) 6.4 (1.2) 5.1 (0.5) 96 (13) 9.1 (10.2) 2.28 (2.78) 212 (38) 123 (32) (n = 44)d 58 (13) 149 (111)

52 43.1 (7.7) 171.1 (5.5) 75.4 (11.5) 25.7 (3.7) 89.2 (9.3) 129 (14) 81 (10) 27 (11) 36 (28) 60 (50) 6.2 (1.2) 5.0 (0.4) 95 (13) 8.9 (9.5) 2.24 (2.9) 218 (34) 133 (33) 57 (13) 137 (68)

0.13 0.80 0.84 0.90 0.58 0.86 0.38 0.57 0.76 0.22 0.40 0.27 0.69 0.91 0.95 0.38 0.12 0.86 0.49

a: LiSM: Life Style Modification Program. b: Values are expressed as means (SD). c: HOMA-IR: homeostasis model assessment of insulin resistance. d: As 3 subjects in the control group had triglyceride levels higher than 400 mg/dL, the mean LDL-cholesterol concentration was determined using the data obtained from 44 control subjects. Tokyo, Japan, December 2006–May 2007.

and LiSM groups, respectively, analysis by the intention to treat method was not possible. Increased consumption of food group A and decreased consumption of food group B were observed in the LiSM (p = 0.00) but not in the control group. Mean differences in habitual food intake changes differed between the two groups. The magnitudes of the intervention effects were 0.31 and 0.35 for food groups A and B, respectively. Changes in numbers of steps did not differ between the two groups (Table 3). Effectiveness of intervention on secondary outcome measures In the LiSM group, 14 of 17 outcome measures showed improvement after the intervention, while only four items in the control group. Mean inter-group differences in changes were significant for body weight (p b 0.05), BMI (p b 0.05), AST (p b 0.05), PG (p b 0.05), insulin (p b 0.05) and HOMA-IR (p = 0.00). Large effect sizes were obtained for insulin and HOMA-IR, 0.73 and 0.77, respectively, and a moderate value of 0.53 for AST. The HbA1c increased in both groups, though the mean inter-group difference was less significant in the LiSM (p = 0.05) than in the control group (Table 4). The LiSM group included a significantly higher percentage of subjects who showed improvements in clinical parameters, e.g. achieving BMI and HOMA-IR (p b 0.05) within normal ranges, as compared to the control group. For TG, the normalized percentage tended to be higher in the LiSM group (p = 0.05) (Table 5). Discussion

subjects decided to count steps as their physical action plans. Thirtytwo subjects (61.6%) decided to walk more than 10,000 steps daily. (Supplement 2). Effectiveness of intervention on primary outcome measures As the numbers of subjects answering the questionnaire at the end of the program were 24/47 (51.1%) and 39/52 (75.0%) in the control

Table 2 Habitual subject behaviors at baseline in control and LiSM groups. LiSMa

Control

Self-check for food group A Self-check for food group B Walking steps

p

n

Means (SD)

n

Means (SD)

42

12.7 (3.6)b

51

13.2 (3.7)

0.56

42

52.9 (5.7)

51

38.3 (4.1)

0.11

31

8974 (1967)

51

17.8 (4.2) 15.1 (4.7)

49 48

Self-efficacy: plan To eat a healthy diet 42 To exercise 43 Stages of dietary change Pre-contemplation 2 (4.3%) Contemplation 8 (17.0%) Preparation 8 (17.0%) Action 18 (38.3%) Maintenance 6 (12.8%) No answer 5 (10.6%)

7 (13.5%) 7 (13.5%) 11 (21.2%) 15 (28.9%) 10 (19.2%) 2 (3.8%)

Stages of change for exercise Pre-contemplation 6 (12.8%) Contemplation 6 (12.8%) Preparation 20 (42.6%) Action 4 (8.5%) Maintenance 6 (12.8%) No answer 5 (10.6%) a: LiSM: Life Style Modification Program. b:Values are expressed as means (SD). Tokyo, Japan, December 2006–May 2007.

9834 (3174) 0.22

16.8 (4.6) 14.9 (5.3)

0.39 0.89

0.34

11 (21.2%) 12 (23.1%) 17 (32.7%) 0 (0%) 9 (17.3%) 3 (5.8%)

0.13

The worksite-based LiSM10!® program produced significant improvements in parameters relating to insulin resistance such as body weight, umbilical circumference, insulin and lipid levels. In particular, insulin and HOMA-IR were markedly decreased and a rise in HbA1c was prevented. There have been many reports on the results of programs designed to give detailed advice about achieving the nutrient quantity intake specified by the intervention program (Ornish et al., 1998; Tuomilehto et al., 2001; Dansinger et al., 2005; Wadden et al., 2005; Villareal et al., 2006). While our subjects made efforts to achieve more approximate goals, as expressed by food group choices, our program was equally efficacious in improving the metabolic parameters. As HbA1c levels were within normal range in all but one of the LiSM group subjects, and the levels roughly estimates glycemic control 100 days prior to its measurement (Tahara and Shima, 1995), a longer study period would be needed to obtain significant decreases in HbA1c. Our present results show individual dietary assessments, made by the subject's themselves, to be useful. The FFQ named “check your dietary habits” sheet was designed for use as a tool for goal setting and the validity is now being studied. In our previous study, habitual food group intakes changed significantly for vegetable, butter and fried food intakes after the intervention (Arao et al., 2007). The internal consistency and reliability of the questionnaire were verified by a high Cronbach's α for food group A. As for food group A items recommended for greater daily consumption, increases in whitevegetables, green/deep-yellow vegetables and mushrooms/seaweed/konnyaku were often selected as components of the dietary action plan. This suggested that these had been consumed less than fish or soybeans/soybean-products before the intervention and that increased consumption was considered to be easily practiced. The top items selected for avoidance in food group B were confectionaries, alcoholic-drinks, sweet-drinks, large servings of grains and butter/ margarine/dressing/mayonnaise. It appears that the subjects might have expected that these foods should be decreased so as to reduce energy intake from carbohydrates, alcohol and fats/oils, and that reducing consumptions would thereby result in body fat reductions. A wider number of items should be selected from food group B than

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17

15

Table 3 Mean intra-group differences, mean inter-group differences in change, and the effect size of each primary outcome measure. LiSMa

Control n

Mean differencec

CI

p

n

−1264 −1.1 −0.8

855 0.9 1.3

0.69 0.87 0.62

44 39 39

b Walking steps Self-check for A group Self-check for B group

22 24 24

Inter-group difference

Mean differencec

95%

95%

CI

p

Mean differenced

95%

CI

p

−379 1.4 1.9

1269 3.4 4.9

0.28 0.00 0.00

942 2.3 2.7

−379 1.0 0.9

2262 3.7 4.5

0.16 0.00 0.00

Effect sizee

b

−204 (2389) −0.1 (2.4) 0.3 (2.5)

445 (2710) 2.4 (3) 3.5 (4.5)

0.91 0.31 0.35

a: LiSM: Life Style Modification Program. b: Values are expressed as means (SD). c: Mean intra-group differences were calculated by subtracting pre- from post-intervention values. d: Mean inter-group differences in change between the control and LiSM group. e: The effect size of the LiSM group versus the control group was calculated by dividing the difference between the two adjusted group means by the underlying standard deviation of all subjects. Tokyo, Japan, December 2006–May 2007.

from food group A as action plan targets. Thus, subjects could select such food groups with the expectation of an equal effect on the target problem, as all items in food A group are effective in ameliorating impaired glucose tolerance, hypertension, and hyperlipidemia (Japan Association of Diabetes Care and Education, 2003; de Castro et al., 2006; Cope et al., 2008). However, items in food group B had different effects on various disorders, possibly requiring the selection of greater numbers of food items by subjects with multiple risk factors. For food group B, the questionnaire requires further refinement. If a subject has low self-efficacy and/or is at an early stage of change, small achievable goals are recognized as being the best strategy for coping effectively with individual challenges (Glanz et al., 1994; Cullen et al., 2001; Sternfeld et al., 2009). Therefore, in this program, recommendations on the numbers of target food subgroups, which subjects wanted to change consumption of, were based on their stages of change and self-efficacy. At most, changes in two items each from food groups A and B were recommended. However, it is suggested that the targets be decided according to medical evidence rather than the feasibility or achievability of a given target. This means that some subjects should probably be guided to a medically appropriate conclusion.

Japanese white-collar workers are often thought of as going out to drinking parties at which highly caloric dishes lacking dietary fiber are consumed. These are social-business events with supervisors, often involving entertainment of business clients and partners. This custom makes dietary improvement extremely challenging. After the 4month intervention, habitual food intakes had changed significantly, with increases in food group A items and decreases in food group B items. There have been numerous reports on increasing fruit and vegetable intakes to protect against or treat cardiovascular risk factors (Appel et al., 1997; Serra-Majem et al., 2006; Shimazu et al., 2007; Sofi et al., 2008; Park et al., 2009). In Japan, patients with impaired energy metabolism are usually encouraged to consume mushrooms/seaweed/konnyaku in order to satisfy dietary fiber requirements without increasing energy intake (Japan Association of Diabetes Care and Education, 2003). In this study, the subjects and their family members described their efforts and challenges involving food selections at restaurants and in cafeterias, cooking these foods, and considerations of the effects of consuming these foods, by posting experiences on their web pages. This indicates that consumption of these foods together with vegetables is a potential strategy for achieving dietary changes in Japanese white-collar workers.

Table 4 Mean intra-group differences, mean inter-group differences in change, and the effect size of each secondary outcome measure. LiSMa

Control n

Mean differenced

95%

CI

p

n

Mean differenced

−1.50 −0.48 −1.78 −3.5 −3.6 −3.2 −9.4 −17.6 −0.44 0.08 −0.31 −1.90 −0.47 −11.1 −8.5 0.4 −46.3

−0.10 −0.03 0.51 2.0 0.1 0.6 −1.1 2.8 0.05 0.22 6.72 1.96 0.65 4.0 4.2 5.3 5.5

0.03 0.03 0.27 0.56 0.06 0.19 0.02 0.15 0.12 0.00 0.07 0.98 0.75 0.35 0.49 0.03 0.12

48 48 47 48 48 48 48 48 47 47 48 44 44 48 48 48 48

−2.14 (2.68) −0.74 (0.94) −1.43 (4.14) −1.4 (11.9) −2.9 (8.9) −4.4 (7.1) −11.2 (18.0) −11.1 (37.5) −0.27 (0.75) 0.07 (0.15) −1.6 (9.6) −3.73 (9.72) −1.04 (3.04) −9.8 (21.5) −5.8 (17.6) 2.0 (6.0) −30.1 (54.4)

b Weight(kg) Body mass index Umbilical circumference (cm) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg) Aspartate aminotransferase (IU/L) Alanine aminotransferase (IU/L) γ-glutamyl transferase (IU/L) Uric acid (mg/dL) Hemoglobin A1c (%) Fasting plasma glucose (mg/dL) Fasting insulin (μU/L) HOMA-IRc Total Cholesterol (mg/dL) LDL-cholesterol (mg/dL) HDL-cholesterol (mg/dL) Triglycerides (mg/dL)

39 39 39 39 39 39 39 39 39 39 39 39 39 39 35 39 39

−0.80 (2.2) −0.26 (0.69) −0.63 (3.53) −0.8 (8.5) −1.8 (5.7) −1.3 (6.0) −5.3 (12.9) −7.4 (31.4) −0.19 (0.77) 0.15 (0.21) 3.2 (10.8) 0.03 (5.96) 0.09 (1.72) −3.5 (23.3) −2.2 (19.6) 2.8 (7.6) −20.4 (80.0)

Inter-group difference 95%

CI

p

Mean differencee

95%

CI

p

−2.91 −1.01 −2.65 −4.8 −5.5 −6.5 −16.4 −21.9 −0.49 0.03 −4.4 −6.68 −1.97 −16.1 −11.0 0.2 −45.9

−1.36 −0.47 −0.22 2.0 −0.3 −2.3 −6.0 −0.2 −0.04 0.11 1.2 −0.77 −0.11 −3.6 −0.8 3.7 −14.3

0.00 0.00 0.02 0.42 0.03 0.00 0.00 0.05 0.02 0.00 0.25 0.02 0.03 0.00 0.03 0.03 0.00

−1.29 −0.47 −0.80 −0.5 −1.4 −2.6 −4.9 −6.3 −0.10 −0.09 −5.2 −2.09 −0.71 −3.1 −0.2 −1.0 −14.1

−2.32 −0.82 −2.48 −5.1 −4.6 −4.6 −9.3 −19.3 −0.39 −0.16 −9.3 −4.10 −1.28 −11.8 −8.1 −3.9 −39.6

−0.27 −0.11 0.88 4.0 1.9 −0.6 −0.4 6.6 0.18 −0.01 −1.0 −0.08 −0.15 5.6 7.6 2.0 11.4

0.01 0.01 0.35 0.82 0.40 0.03 0.10 0.35 0.47 0.05 0.02 0.04 0.00 0.48 0.96 0.52 0.28

effect sizef

b 0.13 0.15 0.02 0.02 0.06 0.53 0.60 0.29 0.28 0.20 0.31 0.73 0.77 0.22 0.17 0.36 0.25

a: LiSM: Life Style Modification Program. b: Values are expressed as means (SD). c: HOMA-IR: homeostasis model assessment of insulin resistance. d: Mean intra-group differences were calculated by subtracting pre- from post-intervention values. e: Mean inter-group differences in change between the control and LiSM groups. f: The effect size of the LiSM group versus the control group was calculated by dividing the difference between the two adjusted group means by the underlying standard deviation of all subjects. Tokyo, Japan, December 2006–May 2007.

16

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17

Table 5 Comparison of efficacy-evaluated normalized changes in secondary outcome measures between the control and LiSM groups. LiSMa

Control

Body mass index Umbilical circumference Systolic blood pressure Diastolic blood pressure Aspartate aminotransferase Alanine aminotransferase γ-glutamyl transferase Uric acid Hemoglobin A1c Fasting plasma glucose Fasting insulin HOMA-IRb Total cholesterol LDL-cholesterol HDL-cholesterol Triglycerides

p

Abnormal at baseline

Normalized (%)c

Abnormal at baseline

Normalized (%)c

25 36 19 20 8 8 19 13 6 4 5 8 14 10 4 15

1 (4.0) 4 (11.1) 2 (10.5) 4 (20.0) 3 (37.5) 3 (37.5) 3 (15.8) 4 (30.8) 0 (0.0) 2 (50.0) 1 (20.0) 2 (25.0) 6 (42.9) 3 (30.0) 1 (25.0) 4 (26.7)

29 34 27 23 12 10 16 13 3 5 5 13 21 23 2 22

7 (24.1) 5 (14.7) 5 (18.5) 9 (39.1) 9 (75.0) 7 (70.0) 5 (31.3) 8 (61.5) 1 (33.3) 3 (60.0) 5 (100.0) 10 (76.9) 8 (38.1) 8 (34.8) 2 (100.0) 13 (59.1)

0.04 0.65 0.46 0.17 0.09 0.17 0.28 0.12 0.13 0.76 0.10 0.02 0.78 0.79 0.08 0.05

a: LiSM: Life Style Modification Program. b: HOMA-IR: homeostasis model assessment for insulin resistance. c: The data are the percentage of those with abnormal values at baseline who converted to normal values at the end of intervention. Tokyo, Japan, December 2006–May 2007.

In our previous study of employees in a factory, we succeeded in increasing leisure time exercise energy expenditure and maximum oxygen uptake (Arao et al., 2007). As the data obtained on physical activity were inadequate, we could not assess the effects of this part of the intervention. All subjects decided to count steps every day and increase the daily number of steps, but only 44 subjects posted data on their web pages. Yon et al. reported more frequent self-monitoring to correlate with weight loss in their weight control program (Yon et al., 2007). In addition, only 10 subjects commented on weekend sports at the goal setting sessions. The reasons for the lack of exercise might reflect conditions such as climate, allergy to pollens, chronic fatigue, and lack of time and recreational facilities. To increase physical activities of white-collar workers in an inner-city area of Tokyo, modification of working schedules and environmental arrangements providing easier to access exercise facilities should be considered. The major limitation of our study was that the rate of completion of self-reports was low, with only 75% and 50% of participants in the LiSM and control groups, respectively, returning useable records. We could not obtain sufficient data at completion of the intervention for analysis by the intention to treat method. Thus, our results might include non-response errors such as over- or under-estimation of the results. In our control subjects, especially those who had no information about health benefits, the post-data collection completion rates were lower than those at baseline. Our previous face-to-face program achieved high compliance, i.e. 86.7% for self-monitoring of steps and 54.7% for dietary targeted activities on a specified sheet (Arao et al., 2007). A number of on-site and face-to-face programs have been found to be effective (Burke et al., 1997; Van Horn and Kavey, 1997), though there is a large gap between the development of effective interventions and their extensive use in industry or public health practice. One of the most significant obstacles has been the high cost and large time demands on both staff and participants (Glasgow and Emmons, 2007). We used a website to allow participants to describe their practices and the counseling received, taking into consideration the convenience of this approach for our study participants. Successful programs have employed e-mail and websites. These programs frequently appeal to participants not to drop out by sending e-mail reminders or, if there are no subsequent log-ons, study staff contact the participant and encourage him/her to return to the website (Van Wier et al., 2006; Ware et al., 2008; Svetkey et al., 2008; Stevens et al., 2008; Sternfeld et al., 2009). Continued personal follow-up, with an actual human being, may be required for

long-term maintenance in individuals attempting lifestyle behavioral changes. Conclusions Participation in the LiSM10!®, an individually tailored behaviorchange-oriented program, in the workplace can result in significant improvements in metabolic parameters relating to insulin resistance in middle-aged male white-collar workers. Generalized and relatively simple lifestyle changes are encouraged by a counselor and appear to be useful for preventing metabolic disorders. Further refinement of both personal contact and interactive technology based interventions is necessary to confirm long-term effects.

Conflict of interest statement No author has financial interest in the subject matter, materials, or equipment connected with this research.

Acknowledgments This study was supported by a grant from the International Life Sciences Institute Japan. We appreciate the collaborative efforts of the Meiji Dairies Corporation, Suntory Holdings Limited and Nichirei Foods Inc. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ypmed.2010.04.008. References Appel, L.J., Moore, T.J., Obarzanek, E., et al., 1997. A clinical trial of the effects of dietary patterns on blood pressure. N. Engl. J. Med. 336, 1117–1124. Arao, T., Oida, Y., Maruyama, C., et al., 2007. Impact of lifestyle intervention on physical activity and diet of Japanese workers. Prev. Med. 45, 146–152. Bandura, A., 2001. Social cognitive theory: an agentic perspective. Annu. Rev. Psychol. 52, 1–26. Brug, J., Oenema, A., Campbell, M., 2003. Past, present, and future of computer-tailored nutrition education. Am. J. Clin. Nutr. 77 (4 Suppl), 1028S–1034S. Burke, L.E., Dunbar-Jacob, J.M., Hill, M.N., 1997. Compliance with cardiovascular disease prevention strategies: a review of the research. Ann. Behav. Med. 19, 239–263. Cope, M.B., Erdman Jr., J.W., Allison, D.B., 2008. The potential role of soyfoods in weight and adiposity reduction: an evidence-based review. Obes. Rev. 9, 219–235.

C. Maruyama et al. / Preventive Medicine 51 (2010) 11–17 Cullen, K.W., Baranowski, T., Smith, S.P., 2001. Using goal setting as a strategy for dietary behavior change. J. Am. Diet. Assoc. 101, 562–566. Dansinger, M.L., Gleason, J.A., Griffith, J.L., Selker, H.P., Schaefer, E.J., 2005. Comparison of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA 293, 43–53. de Castro, T.G., Gimeno, S.G., Ferreira, S.R., Cardoso, M.A., Japanese-Brazilian Diabetes Study Group, 2006. Association of dietary fiber with temporal changes in serum cholesterol in Japanese-Brazilians. J. Nutr. Sci. Vitaminol. (Tokyo) 52, 205–210. Fujii, H., Muto, T., 2009. Diffusion of health education programs with reference to health behavior theories in Japanese workplaces: present status and future plans. J. Occup. Health 51, 84–90. Glanz, K., Patterson, R.E., Kristal, A.R., et al., 1994. Stages of change in adopting healthy diets: fat, fiber, and correlates of nutrient intake. Health Educ. Q. 21, 499–519. Glasgow, R.E., Emmons, K.M., 2007. How can we increase translation of research into practice? Types of evidence needed. Annu. Rev. Public Health 28, 413–433. Japan Association of Diabetes Care and Education, 2003. Food exchange lists. Dietary guidance for persons with diabetes. Japan Diabetes Society (Eds). Bunkodo Co., Ltd., Tokyo, pp. 76–77. Johnson, S.S., Paiva, A.L., Cummins, C.O., et al., 2008. Transtheoretical model-based multiple behavior intervention for weight management: effectiveness on a population basis. Prev. Med. 46, 238–246. Kaplan, B.H., Cassel, J.C., Gore, S., 1977. Social support and health. Med. Care 15 (5 SUPPL), 47–58. Kawada, T., Suzuki, S., 2008. Physical symptoms and psychological health status by the type of job. Work 31, 397–403. Lakka, H.M., Laaksonen, D.E., Lakka, T.A., et al., 2002. The metabolic syndrome and total and cardiovascular disease mortality in middle-aged men. JAMA 288, 2709–2716. Lichtenstein, A.H., Appel, L.J., Brands, M., et al., 2006. Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation 114, 82–96. Ornish, D., Scherwitz, L.W., Billings, J.H., et al., 1998. Intensive lifestyle changes for reversal of coronary heart disease. JAMA 280, 2001–2007. Park, S.K., Tucker, K.L., O'Neill, M.S., et al., 2009. Fruit, vegetable, and fish consumption and heart rate variability: the Veterans Administration Normative Aging Study. Am. J. Clin. Nutr. 89, 778–786. Portnoy, D.B., Scott-Sheldon, L.A., Johnson, B.T., Carey, M.P., 2008. Computer-delivered interventions for health promotion and behavioral risk reduction: a meta-analysis of 75 randomized controlled trials, 1988–2007. Prev. Med. 47, 3–16. Prochaska, J.O., DiClemente, C.C., Norcross, J.C., 1992. In search of how people change. Applications to addictive behaviors. Am. Psychol. 47, 1102–1114. Serra-Majem, L., Roman, B., Estruch, R., 2006. Scientific evidence of interventions using the Mediterranean diet: a systematic review. Nutr. Rev. 64, S27–S47. Shimazu, T., Kuriyama, S., Hozawa, A., et al., 2007. Dietary patterns and cardiovascular disease mortality in Japan: a prospective cohort study. Int. J. Epidemiol. 36, 600–609.

17

Sofi, F., Cesari, F., Abbate, R., Gensini, G.F., Casini, A., 2008. Adherence to Mediterranean diet and health status: meta-analysis. BMJ 337, a1344. Sternfeld, B., Block, C., Quesenberry Jr., C.P., et al., 2009. Improving diet and physical activity with ALIVE: a worksite randomized trial. Am. J. Prev. Med. 36, 475–483. Stevens, V.J., Funk, K.L., Brantley, P.J., et al., 2008. Design and implementation of an interactive website to support long-term maintenance of weight loss. J. Med. Internet Res. 10, e1. Svetkey, L.P., Stevens, V.J., Brantley, P.J., et al., 2008. Comparison of strategies for sustaining weight loss: the weight loss maintenance randomized controlled trial. JAMA 299, 1139–1148. Tahara, Y., Shima, K., 1995. Kinetics of A1C, glycated albumin, and fructosamine and analysis of their weight functions against preceding plasma glucose level. Diabetes Care 18, 440–447. Teramoto, T., Sasaki, J., Ueshima, H., et al., 2007. Executive summary of Japan Atherosclerosis Society (JAS) guideline for diagnosis and prevention of atherosclerotic cardiovascular diseases for Japanese. J Atheroscler Thromb 14, 45–50. The National Health and Nutrition Survey in Japan, 2005. Ministry of Health, Labour and Welfare of Japan (Eds). Daiichi Shuppan, Tokyo, pp.152–153. Tuomilehto, J., Lindström, J., Eriksson, J.G., et al., Finnish Diabetes Prevention Study Group, 2001. Prevention of type 2 diabetes mellitus by changes in lifestyle among subjects with impaired glucose tolerance. N. Engl. J. Med. 344, 1343–1350. Van Horn, L., Kavey, R.E., 1997. Diet and cardiovascular disease prevention: what works? Ann. Behav. Med. 19, 197–212. Van Wier, M.F., Ariëns, G.A., Dekkers, J.C., et al., 2006. ALIFE@Work: a randomised controlled trial of a distance counselling lifestyle programme for weight control among an overweight working population [ISRCTN04265725]. BMC Public Health 6, 140. Villareal, D.T., Miller III, B.V., Banks, M., Fontana, L., Sinacore, D.R., Klein, S., 2006. Effect of lifestyle intervention on metabolic coronary heart disease risk factors in obese older adults. Am. J. Clin. Nutr. 84, 1317–1323. Wadden, T.A., Berkowitz, R.I., Womble, L.G., et al., 2005. Randomized trial of lifestyle modification and pharmacotherapy for obesity. N. Engl. J. Med. 353, 2111–2120. Ware, L.J., Hurling, R., Bataveljic, O., et al., 2008. Rates and determinants of uptake and use of an internet physical activity and weight management program in office and manufacturing work sites in England: cohort study. J. Med. Internet Res. 10, e56. Wing, R.R., Phelan, S., 2005. Long-term weight loss maintenance. Am. J. Clin. Nutr. 82 (1 Suppl), 222S–225S. Yamaoka, K., Tango, T., 2005. Efficacy of lifestyle education to prevent type 2 diabetes: a meta-analysis of randomized controlled trials. Diabetes Care 28, 2780–2786. Yon, B.A., Johnson, R.K., Harvey-Berino, J., Gold, B.C., Howard, A.B., 2007. Personal digital assistants are comparable to traditional diaries for dietary self-monitoring during a weight loss program. J. Behav. Med. 30, 165–175. Zimmet, P., Magliano, D., Matsuzawa, Y., Alberti, G., Shaw, J., 2005. The metabolic syndrome: a global public health problem and a new definition. J. Atheroscler. Thromb. 12, 295–300.