Acceptability and Long-Term Compliance with Drug Treatment for Hypercholesterolemia in Japanese Male Workers: I. Acceptability of Drug Treatment

Transcription

1 J Occup Health 2002; 44: Journal of Occupational Health Acceptability and Long-Term Compliance with Drug Treatment for Hypercholesterolemia in Japanese Male Workers: I. Acceptability of Drug Treatment Masayuki TATEMICHI 1, Tsutahiro HAMAGUCHI 2, Munetaka HASHIRA 3, Takeshi HAYASHI 4, Masato ITO 5, Atsushi NAKATANI 4, Hisanori HIRO 6, Koji MORI 7, Yasushi OoKUBO 8, Takafumi EZAKI 9, Hiroki SUGIMORI 10 and Katsumi YOSHIDA 10 (S-LIT Study Group) 1 Department of Environmental and Occupational Health, Toho University School of Medicine, 2 IBM Japan Medical Services, 3 Health Management Division of Aizuwakamatsu Factory, Fujitsu Corp., 4 Hitachi Health Care Center, 5 Health Management Division of Matsushita Electronics Corp., 6 Center for Occupational Health, Tsurumi, NKK, 7 Medicine & Occupational Health, ExxonMobil Business Services Private Ltd., 8 Department of Occupational and Environmental Medicine Chiba University School of Medicine, 9 Kyoto Industrial Health Association and 10 Department of Preventive Medicine, St. Marianna University School of Medicine, Japan Abstract: Acceptability and Long-Term Compliance with Drug Treatment for Hypercholesterolemia in Japanese Male Workers: I. Acceptability of Drug Treatment: Masayuki TATEMICHI, et al. Department of Environmental and Occupational Health, Toho University School of Medicine We performed a prospective study to assess the acceptability and longterm compliance with drug treatment for hypercholesterolemia in Japanese male workers aged 40 to 57 yr with hypercholesterolemia detected in legally required periodical physical examinations. This paper describes the acceptability of statin-based drug treatment. All eligible workers who had not improved after 3-month diet therapy were recommended for drug treatment with simvastatin, a 3 hydroxy-3- methylcoenzyme A reductase inhibitor. Subsequently, according to each individual preference, the subjects were divided to two groups: a group which accepted drug treatment (DT) and a group which rejected it. The subjects who did not accept drug treatment were further assigned to two groups according to their preference: a group which accepted only dietary intervention (DI) and a non-treatment group (NT). Among 1,278 workers recruited, 656 (51.3%) were placed in the DT group, 414 (32.4%) in the DI group and 208 (16.3%) in the NT group. In the DT group, mean serum total cholesterol Received Jan 8, 2001; Accepted Feb 25, 2002 Correspondence to: M. Tatemichi, Department of Environmental and Occupational Health, Toho University School of Medicine, Omori-nishi, Otaku, Tokyo , Japan (TC) was the highest [272 ± 30 (SD) mg/dl]. The acceptability of drug treatment was significantly associated with the serum TC level, pre-existing medical history of hypertension and a family history of ischemic heart disease (IHD). The occupational issues such as type of job, overtime worked, sleeping time and psychological status were different among the three groups, and habits of taking breakfast and daily exercise were associated with acceptability of drug treatment. These data suggest that acceptability of the drug treatment is low and associated with the workand lifestyle- related issues as well as IHD risk factors. (J Occup Health 2002; 44: ) Key words: Hypercholesterolemia, Treatment, Patient acceptance, Prospective study Previous epidemiological studies have established the risk of ischemic heart disease (IHD) conferred by high serum concentrations of total cholesterol (TC) and lowdensity lipoprotein (LDL) cholesterol 1 3). The treatments recommended were published by National Cholesterol Education Program Expert Committee 4, 5). For Japanese subjects, guidelines for lipid-lowering programs were published by the Japanese Arteriosclerosis Society (JAS) 6), in which the serum TC level for starting lipid lowering intervention was 220 mg/dl in patients with coronary risk factors or 240 mg/dl in patients without them. In large-scale trials, treatment with 3-hydroxy-3-

2 Masayuki TATEMICHI, et al.: Patient Acceptance of Drug Treatment for Hypercholesterolemia 167 methyl-glutaryl coenzyme A (HMG-CoA) reductase inhibitors, which had better levels of patient compliance than other lipid-lowering drugs, was reported to produce excellent reductions in TC and LDL, ranging from 20% to 40% 7, 8), and to result in superior primary and secondary prevention of IHD. Thus, drug treatment with HMG- CoA reductase inhibitors became a key intervention method for lowering serum cholesterol levels. Since 1990, lipid-screening programs have been a part of the legally required periodical physical examination in the Japanese workplace and many patients have been identified as hypercholesterolemic. Management of hypercholesterolemia, which is the most prevalent disease in workers over 40 yr of age 9), has the higher priority in workplace health. Although the development of effective health guidance and diet-exercise treatment is essential for workplace health, basically, hypercholesterolemic workers should be treated according to the JAS guidelines. However, even in drug treatment with HMG- CoA reductase inhibitors, which have fewer side effects than other treatments, the difficulties in obtaining patients acceptance and, in turn, compliance were reported 10, 11). In Japan the effectiveness of HMG-CoA reductase inhibitors in primary and secondary prevention of IHD was examined in studies such as J-LIT (Japan-Lipid Intervention Trial) and KLIS (Kyushu Lipid Intervention study) 12, 13). However, there has not been any report on the rates of acceptability and compliance of drug therapy in Japanese workers, or on information regarding occupational issues that might affect these rates. We therefore conducted a prospective study to assess 1) the acceptability of and long-term compliance with drug treatment for hypercholesterolemia in Japanese male workers, diagnosed in a legally required periodical physical examination and 2) the factors, particularly jobrelated factors, which affect acceptability and compliance with drug treatment. We recommended treatment with simvastatin, an HMG-CoA reductase inhibitor, in addition to a diet-based treatment program, to all eligible workers with hypercholesterolemia, who had not been improved after 3-month diet therapy. Subsequently, according to each individual preference, subjects were divided to two groups: a group which accepted drug treatment (DT) and a group which rejected it. The subjects who did not accept drug treatment were further divided into two groups according to their preference: a group accepted only dietary intervention (DI) and a non-treatment group (NT). We then followed them up for three years. We did not randomly classify the participants, since we believe that the acceptability rate for drug treatment could be assessed by examining the percentage of subjects in the DT group, and factors which affected acceptability could be estimated by comparing the baseline characteristics of the three groups. This paper focuses on the acceptability of drug treatment. Other information will be presented in a separate article. Methods Study design Subjects Male workers aged 40 to 57 yr old with hypercholesterolemia (serum TC, >= 220 mg/dl) were randomly selected after their periodical physical examination. From among them, study subjects were recruited when they had serum TC above 220 mg/dl despite their receiving 3-month health guidance and dietary therapy for hypercholesterolemia. The subjects were recruited from November 1994 to March All subjects gave informed consent. Health guidance, including lifestyle modifications relating to diet and exercise, consisted of sessions lasting more than 15 min, which were conducted by industrial physicians or nurses using a standardized information brochure 14). At first, general health guidance was given following the brochure. The need for drug treatment for hypercholesterolemia was explained in addition to the information on the relationship between hypercholesterolemia and IHD, as well as other risk factors for IHD, including hypertension, diabetes mellitus, a family history of IHD, obesity and smoking. Secondly, an ideal diet was recommended by health professionals, according to the information in the brochure. Finally, the worker s lifestyle was analyzed and the steps for improvement of the lifestyle were recommended. Treatment with simvastatin, an HMG-CoA reductase inhibitor, was recommended to all eligible workers with hypercholesterolemia by an industrial physician in addition to a diet-based treatment program. Subsequently, according to each individual preference, the subjects were divided into two groups: a group which accepted drug treatment (DT) and a group which rejected it. The subjects, who did not accept simvastatin treatment, were further assigned to two groups according to their preference: a group which accepted only dietary intervention (DI) and a non-treatment group (NT). Inclusion and exclusion criteria The subjects were male Japanese workers aged 40 to 57 yr with primary hypercholesterolemia. This population was selected because IHD is known to increase in people over 40 yr of age, according to the National Cholesterol Education Program-Adult Treatment Panel II (NCEP- ATPII) 5) and workers aged 57 yr could be followed up for at least 3 yr before they reached retirement age (60 yr). Patients already taking a cholesterol-lowering drug were included after they underwent a washout period of one month, but they were excluded from the analysis for patient acceptance. Exclusion criteria were as follows: familial or secondary hypercholesterolemia, uncontrolled

3 168 J Occup Health, Vol. 44, 2002 hypertension or diabetes, severe liver or renal disease, any malignancy, and IHD. Treatment methods and goals Subjects in DT and DI groups received dietary advice from an industrial physician or a nurse for more than 15 min every 6 to 12 months with a standardized information brochure 14). Subjects in the DT group received simvastatin 5 mg daily. If the serum TC level did not decrease to the target level described below, the dose of simvastatin was increased to 10 mg, and if necessary, another drug was added after a 1-month evaluation. The targeted level for total cholesterol was less than 220 mg/ dl in patients without risk factors for IHD and less than 200 mg/dl in patients with risk factors for IHD according to the JAS guidelines. Data collection During the study, patient follow-up was conducted at 1, 3, 6, 12, 24, 30, and 36 months. Industrial physicians classified compliance with drug treatment into 4 grades based on interview data: always took the drug according to the prescription (complete); occasionally forgot to take the drug (good); frequently forgot to take the drug (poor); and always forgot to take the drug (very poor). The results of health guidance, including data from the patient interview, were also classified into 4 grades: lifestyle was almost completely improved after health guidance; lifestyle was moderately improved in more than half of the items in health guidance recommendations; lifestyle was slightly improved in less than half of the items in recommendations; and lifestyle was hardly changed. Complete hematology, blood chemistry, blood pressure and body weight were assessed. Electrocardiograms were obtained at the entry. Blood chemistry was determined at 1, 3, 6, 12, 24, 30 and 36 months after the start of treatment. A self-administered questionnaire asking about attitudes, health beliefs and job profiles was answered every year. Detailed medical and family histories and a food frequency questionnaire for dietary assessment were also included in the questionnaire. Clinical events documented during patient follow-up, included 1) diagnosed coronary heart disease, myocardial infarction, unstable angina pectoris, or heart failure, 2) diagnosed vascular disease other than coronary disease, cerebral vascular disease or arteriosclerosis obliterans (ASO), 3) abnormalities on ECG at resting, 4) death due to malignancy, and 5) other outcomes including longterm suspension from work. This information was obtained from a medical certificate submitted to each company. Quality control for laboratory data For the purpose of quality control for conducting a multi-centered study, we had chosen the laboratories that participated in the Zenkoku Roudou Dantai Rengou-Kai or certified by quality control programs, such as the Clinical Examination Quality Control Investigation conducted by the Japan Medical Association. In addition, we confirmed the assay methods and the use of international units for GOT, GPT and γ -GTP concentrations at laboratories. Research organization Data obtained were consolidated every year by an independent data center, and information regarding patient recruitment and survey results were discussed by a monitoring committee from both ethical and medical viewpoints. A management committee, analysis committee, and an evaluation of events committee were organized to manage this study. For proper enforcement of the study protocol, investigators had meetings twice a year during the study period. Statistical analysis The statistical analyses were performed using SPSS software (SPSS Japan, Inc., Tokyo). Mean values were compared with one-way ANOVA and multiple comparisons using Bonferroni s method. Nominal qualitative variables were compared with the chi-squared test. The odds ratio of the variables in acceptability for treatment was calculated by logistic regression analysis. Results Participation and acceptability rates Thirty-one industrial physicians in 27 clinics or health management divisions in industries from 19 companies in Japan participated in this study. Most of the industries belonged to the fields of electronics, automobile, steel and heavy metal production. In our preliminary crosssectional study, of the male workers, 21.5% over 40 yr and 25.6% over 50 yr of age were hypercholesterolemic. The total number of male workers in these companies was estimated to be more than 50,000. For 3 years, 1,563 male workers participated in this study. Of the 1,563, 1,278 remained in the study, but 285 did not meet the entry criteria (Fig. 1). Of 1,278 workers, 429 (34%) worked in metropolitan areas such as Tokyo, Yokohama, Kyoto and Osaka, and 849 (66%) in other areas. The six hundred and fifty-six (51.3%) subjects accepted drug treatment (DT), 414 (32.0%) accepted dietary intervention alone (DI), and 208 (16.1%) did not accept any treatment (NT). Acceptability did not differ significantly according to the location of the workplace. According to each center s data collection rate (actual number of data collected to the targeted number of data collected at each center), study centers were divided into three groups: low (30%), moderate (>= 30 and <80%), and high (>= 80%). There were 4 centers which had a low data collection rate of 1.2% (15 subjects). There

4 Masayuki TATEMICHI, et al.: Patient Acceptance of Drug Treatment for Hypercholesterolemia 169 Fig. 2. Association between acceptability and TC level. DT = drug treatment group, DI = dietary intervention group, NT = non-treatment group. 220 to 240 mg/dl, and 40.2% in those with TC values from 240 to 260 mg/dl. The percent of subjects in the NT group decreased in patients with cholesterol concentrations over 240 mg/dl. Fig. 1. Subject profiles. TC = total cholesterol, IHD = ischemic heart disease. were 13 centers which had a moderate collection rate of 36.6% (468 subjects), and 10 centers with a high collection rate of 62.2% (795 subjects). The acceptability rates in DT, DI and NT groups were 53.0%, 33.5% and 13.5% in those centers with a moderate collection rate, and 49.4%, 32.3% and 18.2% in the centers with a high collection rate, respectively. There was no significant difference between those centers with moderate and high collection rates in the rate of acceptability. Characteristics of the three groups Baseline characteristics are shown in Table 1. There was no significant difference among the three groups in age or body mass index (BMI), but lipid profiles and the incidence of pre-existing illness were significantly different. Acceptability and TC value The acceptability of drug treatment was closely associated with the serum total cholesterol concentration (Fig. 2). It was 22.7% in patients with TC values from Acceptability and risk factors of IHD The difference in baseline characteristics among the three groups is suggested to be the factors associated with acceptability. Predictors of the acceptability rate coincided with major risk factors of IHD. These were hypercholesterolemia (TC >= 260 mg/dl), hypertension, diabetes mellitus, family history of IHD, obesity (BMI >= 24.0 kg/cm 2 ), and smoking status. The mean (± SD) number of IHD risk factors was significantly (p<0.001) higher in the DT group (2.1 ± 1.0) than in the DI group (1.7 ± 1.0) or in the NT group (1.5 ± 1.1) (Fig. 3). We performed logistic regression analysis to assess which IHD risk factors could be associated with acceptability of drug treatment. Analytical results revealed that hypercholesterolemia (TC >= 260 mg/dl), a pre-existing medical history of hypertension, and a family history of IHD were significantly associated with increased acceptability of the drug treatment, whereas obesity and diabetes mellitus did not influence it (Table 2). In contrast, current smoking reduced the rate of accepting drug treatment. Acceptability and job profile We examined the association between acceptability and job profile, including type of job, job-style, and daily working time. In addition, the effect of the stress level and sleeping time on acceptability of drug treatment was also assessed in this study (Table 3-1). The stress level

5 170 J Occup Health, Vol. 44, 2002 Table 1. Characteristics of study patients During Treatment Dietary Intervention Non-Treatment n=656 n=414 n=208 p Age c) 48.3 ± ± ± 4.5 NS a) BMI (kg/m 2 ) c) 24.4 ± ± ± 2.4 NS a) Lipid profile c) TC (mg/dl) ± 28.4*** ϕϕϕ ± 20.0 ϕϕϕ ± 21.5 <0.001 a) TG (mg/dl) ± ϕϕϕ ± ϕϕ ± <0.001 a) HDL-c (mg/dl) 53.1 ± ± ± 14.3 NS a) LDL-c (mg/dl) ± 32.3*** ϕϕϕ ± 23.9 ϕ ± 22.5 <0.001 a) Liver function c) GOT (IU/dl) 28.3 ± 16.8 ϕϕ 26.3 ± ± a) GPT (IU/dl) 34.7 ± 25.9 ϕϕ 32.5 ± ± a) γ-gtp (IU/dl) 64.4 ± 81.0** ϕϕ 47.7 ± ± 43.7 <0.001 a) FBS (mg/dl) c) ± 29.5** ϕ ± ± a) SBP (mmhg) c) ± 16.8*** ϕ ± ± 15.0 <0.001 a) DBP (mmhg) c) 81.8 ± 11.2* 79.7 ± ± a) Present Medical History (%) Hypertension 16.6% 9.2% 7.2% <0.001 b) Diabetes Mellitus 9.6% 4.8% 5.3% b) Hyper-uremic acidemia 9.9% 3.1% 2.9% <0.001 b) Smoking habit 50.7% 61.4% 56.5% b) NS: not significant, ***p<0.001, **p<0.01, *p<0.05, compared to Dietary Intervention group, ϕϕϕ p<0.001, ϕϕ p<0.01, ϕ p<0.05, compared to Non-Treatment group, a: one-way ANOVA, b: χ 2 -test, c: Data are the mean ± SD. TC = total cholesterol, TG = triglyceride, HDL = high density lipoprotein, LDL = low density lipoprotein, FBS = fasting blood sugar, SBP = systolic blood pressure, DBP = diastolic blood pressure. was classified into 11 grades, from 0=low to 10=high. There was little difference in acceptability among types of jobs in DT and DI groups, except that the percentage of manufacturing workers in the NT group was lower than in the DT and DI groups. The rates of acceptability in shift workers among treatment groups were not significantly different. Mean work overtime and stress level were the lowest in the DI group. In the NT group sleep time was the shortest and stress level was the highest. Risk factors for IHD, especially hypercholesterolemia (TC level >= 260 mg/dl), presence of hypertension or both, were significantly associated with acceptance of Fig. 3. Comparison of mean numbers of risk factors for IHD in the three groups. The total number of risk factors defined as TC >= 260 mg/dl, hypertension, DM, family history of IHD, obesity (BMI >= 24.0 kg/m 2 ), and current smoking was determined and its mean was calculated for the three groups: DT = drug treatment group, DI = dietary intervention group and NT = non-treatment group.

6 Masayuki TATEMICHI, et al.: Patient Acceptance of Drug Treatment for Hypercholesterolemia 171 Table 2. Association between IHD risk factors and acceptability of drug treatment Risk Factors odds ratio 95% CI TC TC<260 (mg/dl) reference TC>=260 (mg/dl) 4.0 ( ) Hypertension non reference present 1.7 ( ) Family History non reference present 1.7 ( ) Diabetes Mellitus non reference present 1.6 ( ) Obesity BMI<24.0 (cm 2 /kg) reference BMI>=24.0 (cm 2 /kg) 1.1 ( ) Smoking non-smoker reference current smoker 0.7 ( ) Table 3-1. Acceptability and job associated factors Total Drug Dietary Noninformative Treatment Intervention Treatment cases (%) n (%) n (%) n (%) p Type of Job Office 193 (19.8) 70 (17.1) 78 (20.4) 45 (24.3) Manufacturing 382 (39.1) 159 (38.9) 140 (36.6) 83 (21.7) Sales 28 (2.9) 13 (3.2) 8 (2.1) 7 (3.8) Research and Development 46 (4.7) 20 (4.9) 21 (5.5) 5 (2.7) Management 192 (19.7) 81 (19.8) 77 (20.2) 34 (18.4) Others 135 (13.8) 66 (16.1) 58 (15.2) 11 (5.9) a) 976 (100) 409 (100) 382 (100) 185 (100) Shift work (%) non-shift work % 738 (72.7) 311 (72.8) 283 (71.1) 144 (75.8) shift work % 277 (27.3) 116 (27.2) 115 (28.9) 46 (24.2) NS a) 1015 (100) 427 (100) 398 (100) 190 (100) Work overtime (h/month) (mean ± SD) ± ± ± b) Stress level (low-high, 0-10) (mean ± SD) ± ± 2.1 ϕ 5.0 ± b) Sleeping time (h) (mean ± SD) ± 1.0** ϕϕ 6.6 ± ± 0.8 <0.001 b) NS: not significant, ***p<0.001, **p<0.01, *p<0.05, compared with Dietary Intervention group, ϕϕϕ p<0.001, ϕϕ p<0.01, ϕ p<0.05, compared to Non-Treatment group, a: χ 2 -test, b: one-way ANOVA. drug treatment as mentioned above. Therefore, with logistic regression analysis we calculated the odds ratio for the acceptance of drug treatment, adjusted by the presence of these IHD risk factors. The results of univariate regression analysis are summarized in Table 3-2. There were 456 subjects without IHD risk factors [mean age: 48.3 ± 4.7 (SD)] and 822 subjects with IHD risk factors (mean age 48.3 ± 4.9). Working in shifts was not associated with acceptance of drug treatment. Working overtime was significantly associated with acceptance of drug treatment. In subclass analysis, we further examined the interaction between IHD risk factors and job-related factors for the acceptability of drug treatment. No significant association was found between the variables in job types, work overtime, sleeping time and stress level. Only working in shifts was significantly related to IHD risk factors. Subclass analysis demonstrated that the odds ratio of shift workers to non-shift workers for accepting drug treatment was 0.5 (95% CI ) in subjects

7 172 J Occup Health, Vol. 44, 2002 Table 3-2. IHD risk factors* adjusted odds ratio of the variables concerning job associated factors for acceptance of drug treatment Odds ratio 95% CI Type of Job Office Reference Manufacturing Sales Research and Development Management Others Shift work non-shift work Reference shift work Work overtime Stress level (low-high) Sleeping time *Total cholesterol level >= 260 mg/dl or pre-existing drug treatment for hypertension. Table 3-3. Subclass analysis by IHD risk factors* Subjects without Subjects with IHD risk factor IHD risk factor Odds ratio 95% CI Odds ratio 95% CI Shift work non-shift work Reference Reference shift work *Total cholesterol level >= 260 mg/dl or pre-existing drug treatment for hypertension. without IHD risk factors, and it was 1.5 (95% CI, ) in subjects with IHD risk factors (Table 3-3). Acceptability and private lifestyle There were no significant differences among the groups in any factors regarding private lifestyles (Table 4-1). IHD risk factors adjusted odds ratio demonstrated that only taking breakfast every day was significantly associated with acceptability in the DT group (Table 4-2). The results of the interaction between IHD risk factors and private lifestyles demonstrated that daily exercise was significantly related to IHD risk factors for acceptability of drug treatment. The results of subclass analysis showed that the odds ratio of worker accepting drug treatment for subjects with daily exercise (>2 d/wk) to no-exercise was 2.3 (95% CI, ) in subjects without IHD risk factors and it was 0.8 (95% CI, ) in subjects with IHD risk factors. Discussion In this study, the overall rate of acceptability of drug treatment was about 50%. The acceptability rate was about 40% in subjects with TC over 240 mg/dl and only 60% in those with TC over 280 mg/dl, and the mean TC level in the DT group was over 270 mg/dl. These results suggest that the acceptability of drug treatment for hypercholesterolemia is low in Japanese male workers with hypercholesterolemia diagnosed in legally required periodical physical examination. Each worker with hypercholesterolemia selected the treatment method according to his individual preference. The baseline characteristics of the participants were significantly different among the three groups and analysis revealed that there were factors which affected patients choice of the management programs, equally, the acceptability of the treatment. The main aim of this study was to investigate the relationships between acceptability of drug treatment and occupational conditions, including workers lifestyle and job style, such as type of job, working style, working time, and psychological status in work as well. We believe that the results of this study would be useful information when physicians provide acceptable management programs for hypercholesterolemic middle-aged Japanese male

8 Masayuki TATEMICHI, et al.: Patient Acceptance of Drug Treatment for Hypercholesterolemia 173 Table 4-1. Acceptability and private lifestyle factors Total Drug Dietary Noninformative Treatment Intervention Treatment cases (%) n (%) n (%) n (%) p Marital status married 915 (91.5) 385 (91.2) 361 (91.9) 169 (91.4) single 85 (8.5) 37 (8.8) 32 (8.1) 16 (8.6) NS a) Living style living with family 902 (91.3) 378 (90.6) 354 (91.2) 170 (92.9) alone 86 (8.7) 39 (9.4) 34 (8.8) 13 (7.1) NS a) Eat breakfast occasionally or never 166 (16.4) 61 (14.3) 72 (18.1) 33 (17.5) every day 846 (83.6) 365 (85.7) 325 (81.9) 156 (82.5) NS a) Frequency of <1 d 428 (44.5) 189 (44.7) 146 (41.7) 93 (49.5) eating dinner out 1 2 d 329 (34.2) 141 (33.3) 126 (36.0) 62 (33.0) (d/wk) >2 d 204 (21.2) 93 (22.0) 78 (22.3) 33 (17.6) NS a) Exercise (d/wk) <1 d 558 (55.8) 236 (56.2) 216 (54.5) 106 (57.6) 1 2 d 313 (31.3) 126 (30.0) 122 (30.8) 65 (35.3) >2 d 129 (12.9) 58 (13.8) 58 (14.6) 13 (7.1) NS a) Frequency of drinking alcohol (d/wk) C ± ± ± 1.1 NS b) NS: not significant, a: χ 2 -test, b: one-way ANOVA, c: Data represent mean ± SD. Table 4-2. IHD risk factors adjusted odds ratio for acceptance of drug treatment Odds ratio 95% CI Marital status married Reference single Living style living with family Reference alone Eat breakfast occasionally or never Reference every day Frequency of <1 d Reference eating dinner out 1 2 d (d/wk) >2 d Exercise (d/wk) <1 d Reference 1 2 d >2 d Frequency of drinking alcohol (d/wk) % CI: 95% confidence interval. workers. The outcome of the analysis as to acceptability and compliance generally depended on the characteristics of the study population. We randomly selected subjects who received the legally required periodical physical examination at the workplace. The subject population was neither the participants in health events, wellness programs, or others, where the participants were highly health conscious, nor the workers who were persuaded by our health guidance. In an attempt to draw a generalized conclusion from the results of this study which can be applied to diverse corporate environments, we conducted a multi-centered study and selected companies from wider field of industries, including electronics, automobile, steel, and heavy metal production. These are representatives of Japanese corporations. By comparing the baseline data in three groups, factors

9 174 J Occup Health, Vol. 44, 2002 affecting the acceptability of drug treatment were identified as mentioned above. We determined which were risk factors for IHD affected workers acceptance of the drug treatment. It is important for workers to recognize the necessity of drug treatment to prevent IHD, since hypercholesterolemia is clinically asymptomatic. Multivariate analyses showed that hypertension and a family history of IHD, in addition to a high cholesterol level, but not obesity or smoking, were independent factors that affected subjects acceptance of drug treatment. This finding suggests that workers do understand the risk of a high serum total cholesterol level and the presence of certain illnesses, particularly hypertension which leads to IHD, but they do not recognize lifestyle-related factors, such as obesity and smoking, as risk factors for IHD. Although our study population had had greater access to education, via daily health guidance by industrial physicians and nurses, than the general population, our results showed that acceptability of drug treatment was still low. This indicates the difficulties in educating workers to accept drug treatment for hypercholesterolemia. As the next step, we examined factors affecting acceptability for drug treatment. Although private issues such as marital status and living style did not affect acceptability, some job-related factors did significantly affect it. Since risk factors for IHD, such as a TC level >= 260 mg/dl or hypertension, significantly affected acceptability of drug treatment, we examined IHD risk factor adjusted odds ratios of job and private-lifestyle related factors for accepting drug treatment. It is possible that some job types are associated with acceptability of drug treatment, but the concise job type was unknown. The association between working in shifts and acceptability for drug treatment depended on the presence of the IHD risk factors. Shift workers without IHD risk factors were not likely to accept drug treatment. On the other hand, shift workers with IHD risk factors, easily accepted drug treatment. These findings suggest that working in shifts may primarily discourage patients from accepting drug treatment. However, in shift workers with IHD risk factors, awareness of the risk of IHD may be increased by working in shifts. Mean work overtime and the stress level were lowest in the DI group. This leads us to speculate that diet-based treatment may require extra free time and patient psychological commitment. In the NT group, the mean sleeping time was shortest, and the mean stress level was highest. Workers in the habit of taking breakfast every day or daily exercise, suggesting higher consciousness of health, tended to accept drug treatment. From these results we consider that workrelated issues and consciousness of health rather than individual private issues such as marital status and living style affect acceptability of drug treatment. This study has several limitations. 1) acceptability depends on the participants baseline characteristics. It is possible that variability and selection bias caused by the multi-center study method decreased the sensitivity of the results, because the characteristics of workers are thought to differ considerably from company to company. In addition, in collecting data, several variations might have occurred. In 4 centers, 100% of subjects preferred drug treatment and their data collecting rate was low as < 30%. The percentage of workers in the DT group differed considerably from center to center with low, moderate and high collecting rates, indicating the existence of a selection bias. In this study we consolidated the data and analyzed them together, since the data from the centers with a low collecting rate accounted for only 1.2 % (15 subjects) of the total. 2) Many confounding factors might be present in interpreting the data, because this study used an observational approach. Therefore, although we discussed the data focusing on the correlation with risk factors for IHD, interpretation of the results might still be limited. 3) Our study subjects were workers who accepted re-examination of the serum total cholesterol level after diet intervention, so that, workers, who had refused to be re-examined or even refused to hear the results after the physical examination, were not included. Our data may have been obtained from subjects who had a relatively higher concern for their health condition although, even though, at first the study population was selected randomly. 4) Although we attempted quality control among laboratories by selecting with a certain standard, strict quality control was not performed according to international quality control programs 15). Therefore, our clinical data require careful evaluation for quantitative values. In conclusion, the present study revealed the low acceptability of statin-based treatment for hypercholesterolemia in workers detected at the time of a legal periodical examination, particularly in hypercholesterolemic workers without other risk factors of IHD. In order to improve the acceptability of the drug treatment, we believe that physicians should be required to take into consideration the work-related issues as well as the health consciousness of the individual worker. S-LIT Study Group Management Committee Tsutahiro Hamaguchi (IBM Japan) Munetaka Hashira (Fujitsu Ltd) Hisanori Hiro (NKK) Toshiaki Miyamoto (Nippon Steel Corporation KIMITSU works) Analysis Committee Masayuki Tatemichi* (Toho University) Takafumi Ezaki (Kyoto Industrial Health Association) Jiro Moriguchi (Kyoto Industrial Health Association)

10 Masayuki TATEMICHI, et al.: Patient Acceptance of Drug Treatment for Hypercholesterolemia 175 Hiroto Imai (Kyoto Industrial Health Association) Takashi Hayashi (Hitachi Health Care Center) Tsutomu Odahara (Hitachi Health Care Center) Atsushi Nakatani (Hitachi Health Care Center) Mitsuhide Taira (Benesse Corporation) Katsutoshi Tanaka (Toshiba Corporation) Kako Takei (Mito General Hospital, Hitachi Ltd.) Masato Ito (Matsushita Electronics Corporation) Tetsutaro Sugiura (Matsushita Electric Industries Co. Ltd.) Masato Tanaka (Toyota Motor Kyushu, Inc.) Yasutsugu Kudo (TDK Corporation) Monitoring Committee Yasushi Okubo* (Chiba University) Toshiyuki Matsuda (Sumitomo Metal Industries, Ltd.) Kaori Matsuda (Sumitomo Metal Industries, Ltd.) Keiichiro Yarita (Mazda Motor Corporation) Event Committee Koji Mori* (Exxon Mobil Business Services Private Ltd.) Masaiwa Inoue (Nippon Paper Industries) Shuji Nakano (Toshiba Corporation Yanagicho Complex) Hiroki Takahashi (Toshiba Corporation) Yoshiaki Tomioka (Sumitomo Metal Industries, Ltd.) Evaluation Committee Toro Takeda* (Lion Corporation) Tetsuya Asaeda (Kyoto Industrial Health Association) Satomi Nagai (Matsushita Electric Works, Ltd.) Masahiro Munaka (NISSAN Motor, Ltd. Kyushu plant) Koichi Chiba (VICTOR Company of Japan, Ltd.) Supervisors Yoshinori Shouji (Chiba Occupational Health Center) Katsumi Yoshida, Hiroki Sugimori (St. Marianna University) * members of management committee. References 1) The Lipid Research Clinics Coronary Primary Prevention Trial results. I. Reduction in incidence of coronary heart disease. JAMA 1984; 251: ) The Lipid Research Clinics Coronary Primary Prevention Trial results. II. The relationship of reduction in incidence of coronary heart disease to cholesterol lowering. JAMA 1984; 251: ) Kannel WB, Neaton JD, Wentworth D, et al. Overall and coronary heart disease mortality rates in relation to major risk factors in 325,348 men screened for the MRFIT. Multiple Risk Factor Intervention Trial. Am Heart J 1986; 112: ) Report of the National Cholesterol Education Program Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. The Expert Panel. Arch Intern Med 1988; 148: ) Summary of the second report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel II). JAMA 1993; 269: ) Investigating committee of guideline for diagnosis and treatment of hyperlipidemia, Japan Atherosclerosis Society: Guideline for diagnosis and treatment of hyperlipidemia in adults. Jpn Atherosclerosis 1997; 25: ) Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995; 333: ) Randomised trial of cholesterol lowering in 4444 patients with coronary heart disease: the Scandinavian Simvastatin Survival Study (4S). Lancet 1994; 344: ) Industrial Health Division, Industrial Safety & Health Department, Labor Standards Bureau. Results of legal medical check-up in Roudou Eisei Kannri, 2000; 11: ) Maenpaa H, Manninen V, Heinonen OP. Comparison of the digoxin marker with capsule counting and compliance questionnaire methods for measuring compliance to medication in a clinical trial. Eur Heart J 1987; 8: ) Frick MH, Elo O, Haapa K, et al. Helsinki Heart Study: primary-prevention trial with gemfibrozil in middleaged men with dyslipidemia. Safety of treatment, changes in risk factors, and incidence of coronary heart disease. N Engl J Med 1987; 317: ) Matsuzawa Y, Oyokawa S, Saitoh T, et al. Design and Baseline Results of Hyperlipidemic Patients in the Japan Lipid Intervention Trial (J-LIT) A 6 years Follow-up, Nationwide Multiclinic Study- Rinshou Iyaku 1994; 10: ) A coronary primary intervention study of Japanese men: study design, implementation and baseline data. The Kyushu Lipid Intervention Study Group. J Atheroscler Thromb 1996; 3: ) Ozono Y, Fukushima,S. Hyperlipidemia and diet for hyperlipidemia. Iyaku Koukoku-sha, Tokyo, 1994: ) Myers GL, Cooper GR, Winn CL, Smith SJ. The Centers for Disease Control-National Heart, Lung and Blood Institute Lipid Standardization Program. An approach to accurate and precise lipid measurements. Clin Lab Med 1989; 9: