Intervention on Coronary Risk Factors by Adapting a Shift Work Schedule to Biologic Rhythmicity

Transcription

1 Intervention on Coronary Risk Factors by Adapting a Shift Work Schedule to Biologic Rhythmicity KRISHNA ORTH-GOMER, MD The effects of a new shift schedule on coronary risk factors was assessed in a short-term intervention trial in 45 volunteer policemen that were divided into two groups. They worked 4 weeks each on their customary schedule (counter-clockwise ) and on the new schedule (clockwise ). Serum lipids, glucose, uric acid, blood pressure, nocturnal urinary excretion of catecholamines, the quality and quantity of sleep, and tobacco consumption were assessed before,, and after each schedule. During clockwise, serum levels of triglycerides (but not cholesterol), and of glucose (but not uric acid) were significantly lower than counter-clockwise. Sleep was reported to be longer and better with clockwise, but tobacco consumption did not differ between the two schedules. After clockwise, systolic (but not diastolic) blood pressure and urinary excretion of catecholamines were significantly lower than after counter-clockwise shift s. The results suggest that adapting shift to biological circadian rhythms has a favorable short-term effect, not only on subjective well being but also on risk factors for ischemic heart disease. Shift work is associated with an increased strain on the worker with both psychologic and physical consequences (1). Shift workers are known to have an increased prevalence of gastric and intestinal dysfunctions as well as sleep disturbances (2 4). Recent population studies indicate that risk factor levels for ischemic heart disease (IHD) are higher among shift workers than among day workers. In the Troms0 heart study of northern Norway, serum cholesterol was found to be higher and smoking was more common in shift workers than in other workers (5). In the German study of the Cottbus area, risk factor levels among night workers were higher than among day workers (6). In a Swedish prospective study of occupa- From the National Institute for Psychosocial Factors and Health, Stockholm, Sweden. Address reprint requests to: Dr. Kristina Orth-Gomer, IPM, Box , Stockholm, Sweden. Received for publication May 17, Accepted for publication November 15, tional stress and IHD, shift work was found to increase the incidence of acute myocardial infarction, but the excess risk was found in multivariate analyses to be due largely to excess smoking (7). The pathogenic mechanisms involved in shift work are likely to be related to interference with diurnal rhythms. Most biological functions, e.g., heart rate, blood pressure, and catecholamine excretion rates as well as many behavioral patterns such as sleep and wakefulness exhibit a 24-h rhythmicity, which is a circadian variation (8). This variation is partly dependent on external time cues such as daylight, clocks, work, and social arrangements, and partly on internal regulating mechanisms (9). Changing from day to night work has been found to be accompanied by a distortion of circadian rhythmicity, by an elevation of serum lipids, glucose, and uric acid, and by an increased urinary excretion of catecholamines (10). Furthermore, laboratory experiments have shown that, total deprivation of external time synchronizers, the circa- Psychosomatic Medicine Vol. 45, No. 5 (October 1983) 407 Copyright 1983 by the American Psychosomatic Society, Inc Published by Elsevier Science Publishing Co., Inc 52 Vanderbilt Ave., New York, NY /83/$3 00

2 KRISHNA ORTH-GOMER dian period is spontaneously prolonged to around 25 h (9). Thus, it is easier to adjust to an experimental prolongation than to a shortening of the circadian period. These findings were applied in a pilot study, which attempted to improve shift work conditions with simultaneous evaluation of the short-term effects on coronary risk factors and subjective well being. Night and shift work is usually included in the work schedule of patrolling policemen. Among Stockholm policemen, approximately 1800 men belong to the patrolling police force. In a survey study of 747 randomly selected patrolling policemen 52% were found to be dissatisfied with their working conditions, in particular with their workshift schedules, which included both evening and night work (11). In the intervention trial it was not possible to eliminate work irregular hours, but organization of the shift schedule was amenable to change. The customary schedule of made the men go to bed a few hours earlier on each subsequent day of the work period. They started the period with a night shift, continued with an evening shift, then switched to a mid-day shift, and finished with a morning shift (Fig. 1). In the new schedule, the of shifts was reversed from counter-clockwise to clockwise; thus, it was theoretically better adapted to spontaneous circadian rhythmicity. The effects of the new schedule on coronary risk factors and on subjective well being, sleep, and urinary excretion of catecholamines were evaluated and compared to the customary schedule. METHODS All male patrolling policemen were invited by letter to participate in the intervention trial: 57 men volunteered. The low response rate was probably due partly to the fact that only 15 days could be allowed for a decision, that the experimental period extended over a major holiday, and that participation made vacations impossible that time. Two groups of 23 men each were selected to match by age and professional experience. On physical examination, all men were found to be free from SHIFT ROTATION SCHEDULE COUNTER-CLOCKWISE ROTATION Day3(lW2) iay3(u-22) Fig. 1. Schedule of shift for Swedish policemen. Ordinary (left) and experimental (right) schedule. 408 Psychosomatic Medicine Vol. 45, No. 5 (October 1983)

3 CORONARY RISK FACTORS clinically overt IHD or other chronic illness. One man acquired an acute infection the trial and had to be excluded. The mean ages of the men in the two groups were 30.4 (± 1.0) and 30.8 (± 1.2) years. They had been working as policemen for 8.6 ( ± 0.9) and 7.9 ( ± 1.0) years, respectively. In the first group, 32% of the men were smokers and, in the second group, 26% smoked; 5% and 4%, respectively, were smoking 20 cigarettes or more a day. The random sample of patrolling policemen who participated in the survey study had a mean age of 30.1 (± 0.3) years and had been working as policemen for 7.7 ( ± 0.3) years. Thirty-eight percent of them were smokers and five percent smoked 20 cigarettes or more daily (12). Thus, in these respects there were no major differences between the two study groups and the random sample of patrolling policemen. However, in the two study groups, 59% and 48%, respectively, had been working nights for more than 5 years, whereas in the random sample only 15% had had night work for that period of time. The proportion of men who said they wanted work-free days to catch up on sleep was similar; 91% in each of the study groups and 87% in the random sample. Both study groups worked for 4 weeks on each schedule. Group 1 started with the counter-clockwise, and group 2 with the clockwise,. Before starting the experiment, baseline levels of coronary risk factors were assessed, and the length and quality of night sleep was estimated; catecholamine excretion rates at rest were measured simultaneously. After 3 weeks of work the assessments were repeated and an estimation of work load and work strain was added. Finally, 3 days after the completion of each schedule the initial procedure was repeated. Schedules were then exchanged between groups and the same procedure was followed for an equal period of time. All examinations were performed in a quiet room at police headquarters. Fasting blood samples were obtained after a night's sleep at 8 AM by an experienced nurse. Blood pressure was measured with a sphygmomanometer on the right arm after 5 min of rest with the subject in a sitting position. Urinary samples were collected after sleep at night. The men were asked not to consume citrus fruits, bananas, chocolate, or alcohol 12 h before and each Collection period. They were allowed to consume tobacco and coffee as usual, but were asked to note all consumption in a protocol. Urinary epinephrine and norepinephrine were measured by a semiautomatic fluorophotometric method (13). Serum analyses were performed by an Autochemist (AGA). Physical and psychological strain at work were estimated by self-rating scales, which were completed twice after day work and twice after night work. The items were added up to compile an index of work strain ranging from 0 to 50 (14). Work load was estimated by the number of police actions taken each work shift as reported in routine protocols. The duration and quality of sleep were rated by each man after both day work with night sleep and night work with day sleep. Ratings of ease of falling asleep, ease of staying asleep, and of fitness after sleep were added up to an index of quality of sleep ranging from 1 to 5.25 (15). After each 4-week period, attitudes to work schedules and opinions about health status were assessed. The significance of differences between mean group values on coronary risk factors, catecholamine excretion rates, and subjective ratings at baseline examination were tested by Student's t-test (two-tailed). In order to compare changes in physiological parameters the two work schedules, ratios between work levels and baseline levels were calculated. To account for interindividual variability in, e.g., catecholamine excretion rates, individual ratios were formed and mean group deviations from baseline values were compared. The significance of differences between mean percent deviations from baseline values clockwise and counter-clockwise was tested by Student's t-test (twotailed). RESULTS Baseline Measurements No statistically significant initial differences in risk factor levels were found between the two groups of policemen. Relative body weight, fasting serum cholesterol, triglycerides, glucose, and uric acid were within normal limits and were very similar in the two groups. Systolic (124.5 ± 2.5 and ± 2.5 mm Hg) and diastolic blood pressure (83.3 ± 1.9 and 81.5 ± 2.3 mm Hg) were also similar. Night sleep was almost equally long (8,0 ± 0.2 and 8.1 ± 0.2 h) and was given the same qualitative rating (3.4 ± 0.2) in both groups. Psychosomatic Medicine Vol. 45, No. 5 (October 1983) 409

4 KRISHNA ORTH-GOMER Finally, the urinary excretion of epinephrine (8.4 ± 0.8 and 10.2 ± 1.0 pmol/ min; and norepinephrine (92.4 ± 7.6 and 93.0 ± 6.0 pmol/min) corresponding to the night sleep period did not differ significantly between the two study groups. Effects on Serum Levels of Lipids, Glucose, and Uric Acid The effects on fasting serum cholesterol, triglycerides, glucose, and uric acid are shown as a percent change in Figure 2. Serum glucose (t = 4.65, p < 0.001) and triglycerides (t = 2.95, p = 0.005) but not cholesterol (t = 1.33, p = 0.191) levels were significantly lower clockwise as compared to counter-clockwise. Serum uric acid levels also were lower clockwise (t = 1.68, p = 0.100), but the difference did not reach statistical significance. Values of these measurements made 3 days after each schedule revealed no statistically significant difference between the two schedules. Effects on Blood Pressure Blood pressure was measured 3 days after the completion of each work schedule but, due to a shortage of personnel, not 0/ Cholesterol chonge 16 U Tn glycendes ** IF Uric acid [Clockwise r- -Crui'l. 1! ( ::(>* s- 1 ' ' ; <:: v. after ofter 4A t after after Fig. 2. Percent change in serum cholesterol, triglycerides, glucose, and uric acid with clockwise and counter-clockwise s. 410 Psychosomatic Medicine Vol. 45, No. 5 (October 1983)

5 CORONARY RISK FACTORS the ongoing work schedule. Mean systolic blood pressure was significantly lower after clockwise (111.6 ± 11.2 mm Hg] as compared to counter-clockwise (115.6 ±10.8mmHg,t = 2.52,p < 0.05). Mean diastolic blood pressure did not differ significantly between the two schedules (75.2 ± 9.9 mm Hg and 75.8 ± 8.3 mm Hg, (t = 0.41). As the initial blood pressure measurement was made by a different investigator, it was not meaningful to calculate the percent change from baseline to work level. Effects on Subjective Assessments The consumption of tobacco (cigarettes, cigars, and pipe smoking as well as snuff consumption) was recorded four work shifts on both schedules. The mean tobacco consumption among smokers' was similar clockwise and counterclockwise s (4.7 ± 2.2 versus 4.4 ± 2.0 cigarettes, cigars, or pipes per 8-h work shift, t = 0.91), as was the mean snuff consumption among snuffers (3.7 ± 0.5 versus 4.1 ± 0.6 snuffings per shift, t = 1.14). Reports of work load and work strain did not differ significantly between the two shift s. During day work, a mean of 3.9 (± 0.3) actions were reported the clockwise and a mean of 3.5 (± 0.4) actions the counter-clockwise (t = 0.87). At night more actions were taken, but the mean number clockwise did not differ from that counter-clockwise (5.1 ± 0.4 versus 4.7 ± 0.3; t = 0.87). Indices of work strain were quite similar under all four conditions; 18.5 (± 0.9) day work with a clockwise and 17.8 (± 0.8) with a counter-clockwise (t = 0.91) ( ± 0.7) night work with a clockwise and 17.9 (±0.7) with a counter-clockwise (t = 1.72). Ratings of length and quality of sleep at night and the day are shown in Table 1. Sleep at night after day work was reported to be significantly longer and better clockwise as compared to counter-clockwise. The differences decreased after termination of the schedules. Day sleep after night work was reported to be considerably shorter and of inferior quality as compared to night sleep, but significant differences between the two schedules were not found. Subjective assessments made after each schedule are compared in Table 2. Dissatisfaction with the schedule was stronger after the counter-clockwise as compared to the clockwise, but the difference did not reach statistical significance. Dissatisfaction with having too little time for rest between work shifts, however, was significantly stronger after counter-clockwise than after clockwise. Appraisal of own health was also significantly higher after the clockwise schedule. Effects on Catecholamine Excretion Rates The urinary excretion of epinephrine and of norepinephrine are illustrated in Figure 3. The excretion rates of both compounds rose the first weeks of clockwise, but fell towards the end of the observation period. During the counter-clockwise, however, the excretion rates of both compounds rose, with only one exception, throughout the period. Only at the last assessment, 3 days after termination of either schedule, did the difference between schedules reach statistical significance (t = 1.91 for epinephrine; p = if a one-tailed t-test is applied, and t = 2.19 for norephinephrine; p =0.035). Psychosomatic Medicine Vol. 45, No. 5 (October 1983) 411

6 KRISHNA ORTH-GOMER TABLE 1..ength of sleep schedule (h) After schedule Quality and Length of Sleep During and After Each Rotation Schedule* Clockwise (n =45) 9.0 ± ±0.22 Counter-clockwise [n = 45) Night sleep 8.4 ± ±0.28 t-value of difference 2.04 b 1.05 n.s. Quality of sleep schedule (index ) After schedule 3.7 ± ± ± c 1.76 n.s. Day sleep 5.6 ± ± 0.22 Length of sleep schedule (h) After schedule 5.2 ± ± n.s n.s. Quality of sleep schedule (index ) After schedule a Mean ± SEM. b p < c p < ± ± ± ± n.s n.s. DISCUSSION Large-scale intervention trials against risk factors for IHD usually attempt to induce changes of life style by either public campaigns or individual counselling on diet and smoking, and by pharmacologic treatment of high blood pressure (16, 17, 18). The possibility that working conditions may influence risk factor levels for IHD have been extensively investigated (7, 19, 20, 21) but successful efforts to reduce risk factor levels by improving working conditions have not been published. The attempt in this pilot study to adapt shift work to biological circadian rhythmicity was accompanied by a reduction of some of the standard risk factors for IHD as well as by changes in psychologic factors. Serum triglycerides, glucose, and uric acid were significantly lower clockwise as compared to counter-clockwise, but the differences disappeared after termination of the schedules. Systolic blood pressure, which was only measured after schedules, also decreased with the clockwise. Habits of tobacco and snuff consumption remained unchanged throughout the two schedules. Although it was not possible to keep dietary habits under strict control, it seems reasonable to assume that they did not differ between schedules to any greater extent than smoking habits. However, if any policeman used eating as a 412 Psychosomatic Medicine Vol. 45, No. 5 (October 1983)

7 CORONARY RISK FACTORS % change Unnory epinephnne Urinary norepinephnne I 1 Clockwise [ ] Counter-clockwise weeks 3 weeks i rototion 2 weeks 3 weeks Fig. 3. Percent change in urinary excretion of epinephrine and norepinephrine with clockwise and counter-clockwise s. TABLE 2. Subjective Assessments After Each Rotation Schedule' Clockwise Counter-clockwise t-value of difference n = 45 n = 45 Dissatisfaction with schedule 2.56 ± ± n.s. Dissatisfaction with amount of leisure time between workshifts 1.44 ± c Opinion of own general health status 1.84 ± ± b "Mean ± SEM. p <0.01. p < Range of scale 1-5. Higher scale score = more dissatisfied. Psychosomatic Medicine Vol. 45, No. 5 (October 1983) 413

8 KRISHNA ORTH-GOMER means to cope with stress and fatigue, an improved work schedule may have improved dietary habits and thus accounted for changes of, e.g., lipids and glucose. As anticipated in the study design, work load estimated by the number of reported police actions did not differ between the two schedules. Subjective ratings of work strain also were similar. The general sense of well being, however, improved with the clockwise, perhaps best demonstrated in the change of sleeping habits. Sleep at night after day work was reported to be longer and better with the clockwise than with the counterclockwise. The effect was more prominent than after termination of schedules. Although the number of hours worked were the same on both schedules, the resting periods between work shifts were prolonged from approximately 12 to 20 h on the clockwise. This may have accounted for part of the beneficial effects on sleep. As a result of the prolonged rest between shifts, weekends after s were shortened with the clockwise. This disadvantage may partly explain why satisfaction with the new schedule was not as unanimous as the subjective improvement of general health. Contrary to what may have been expected from earlier reports (22, 23), there was no direct relationship between the changes of catecholamine excretion rates and those of plasma lipids, glucose, and uric acid or length and quality of sleep. The expected decrease clockwise as compared to counter-clockwise did not appear until after termination of the schedules. It is possible that endogenous biological rhythmicity of catecholamine production and excretion may have counteracted and delayed the expected decrease clockwise. A latency period of several weeks, when changes in diurnal patterns of catecholamine excretion rates were induced by changes in sleep/wake patterns, thus was demonstrated (1Q). It is also possible that the rather sharp increase, especially the first 2 weeks, arose from the effects of novelty (24). The clockwise was a new experience to the participants, but the counter-clockwise was not. The results of this pilot study suggest that it is possible to improve both the subjective sense of general good health and physiologic health indicators by improving working conditions. Caution is needed, however, in the interpretation of short-term changes in a small study group. Further studies are also needed to examine whether the same changes persist in a long-term trial, and whether they have an impact on the incidence or mortality of ischemic heart disease. The study was supported by a grant from the Swedish Work Environment Fund. REFERENCES 1. Rutenfranz J, Knauth P, Angersbach D, Shift work research issues, in Biological Rhythms, Sleep and Shift Work. Edited by LC Johnson, WP Colquhoun, DI Tepas, MC Colligan, New York, SP Medical & Scientific Books, 1981, pp Costa G, Apostoli P, d'andreas F, Gaffur E: Gastrointestinal and neurotic disorders in textile shift workers, in Night and Shift Work. Biological and Social Aspects. Edited by A Reinberg, N Vieux, P Andlauer, Oxford, Pergamon Press, 1981, pp Psychosomatic Medicine Vol. 45, No. 5 (October 1983)

9 CORONARY RISK FACTORS 3. Rutenfranz J, Colquhoun WP, Knauth P, Ghata JN: Biomedical and psychosocial aspects of shift work. Scand J Work Environ Health 3: , Akerstedt T, Gillberg M: Sleep disturbances and shift work, in Night and Shift Work. Biological and Social Aspects. Edited by A Reinberg, N Vieux, P Andlauer. Oxford, Pergamon Press, Thelle D, Rorde OH, Try K, Leymann EH: The Tromso Heart Study. Methods and main results of the crosssectional study. Acta Med Scan 200: , Piorkowski P, Giinther KH, Harig H, Handreg W, Braun H: Social factors correlation with coronary heart disease risk in a rural community of the GDR -Model Cottbus. Abstract, International Symposium, Psychophysiological risk factors of cardiovascular diseases. Psychosocial stress, personality and occupational specificity, Karolovy Vary, Sept. 7 11, AlfredssonL, KarasekR, Theorell T: Myocardial infarction risk and psychosocial work environment An analysis of the male Swedish working force. Soc Sci Med 4: , Halberg F: Biological rhythms, in Advances in Experimental Medicine and Biology, Vol. 4. Biological rhythms and endocrine function. Edited by LW Hedlund, JM Franz, AD Kenny. New York: Plenum Press, Akerstedt T: Altered sleep/wake patterns and circadian rhythms. Laboratory and field studies of sympatho-adreno-medullary and related variables. Acta Physiol Scand Suppl 469, Theorell T, Akerstedt T, Day and night work. Changes in cholesterol, uric acid, glucose and potassium in serum and in circadian patterns of urinary catecholamine excretion. Acta Med Scand 200:47 53, Zamore de Alencar K, OlivegSrd Landen R, Olsson Vikstrom A: Patrolling policemen in Stockholm. I.I. Introduction to the study, social background and work organization. Report No. 142, Laboratory for Clinical Stress Research, Stockholm, Orth Gomer K, OlivegSrd Landen R: Patrolling policemen in Stockholm. 1:5. Health status. Report No. 145, Laboratory for Clinical Stress Research, Stockholm, Andersson B, Hovmoller S, Karlsson CG, Svensson S: Analysis of urinary catecholamines: an improved auto-analyzer fluorescence method. Clin Chim Acta 51:13 28, Kroes W, Hurrel J: Job stress and the police officer. Identifying stress reduction techniques. US Dept. of Health Education and Welfare. Publication Service. HEW -Publication No National Institute for Occupational Safety and Health, Akerstedt T, Torsvall L: Shift work. Shift-dependent well-being and individual differences. Ergonomics, 24: , Salonen JT, Puska P, Mustamiemi H: Cardiovascular morbidity and mortality changes the comprehensive community programme for the control of cardiovascular diseases in North Karelia, Finland, Br Med J 2: , World Health Organization European Collaborative Group: Multifactorial trial in the prevention of coronary heart disease: 2. Risk factor changes at two and four years. Eur Heart J 3: , Hjermann I, Holme I, Velve Byre K, Leren P: Effect of diet and smoking intervention on the incidence of coronary heart disease. Report from the Oslo Study Group of a randomized trial in healthy men. Lancet 12: , Hinkle L, Whitney LH, Lehman EW, Dunn J, Benjamin B, Kling R: Occupation, education and coronary heart disease. Science , Karasek R, Baker A, Marxer F, Ahlbom A, Theorell T: Job decision latitude, job demands and cardiovascular disease. A prospective study of Swedish men. Am J Publ Health 71:694, Orth Gomer K: Ischemic heart disease and psychological stress in Stockholm and New York. J Psychosom Res 23: , Carlson LA, Levi L, Ord L: Stressor-induced changes in plasma lipids and urinary excretion of catecholamines and their modification by nicotine acid. Acta Med Scand Suppl 528:191, Rahe R, Rubin RT: Measures of subjects motivation and affect correlated with their serum uric acid, cholesterol and cortisone. Arch Gen Psychiatry 26: , Frankenhauser M: Behaviour and circulating catecholamines. Brain Res 31: ,1971 Psychosomatic Medicine Vol. 45, No. 5 (October 1983) 415

10