Paper | Published:

Effect of shift work on body mass index: results of a study performed in 319 glucose-tolerant men working in a Southern Italian industry

International Journal of Obesity volume 27, pages 13531358 (2003) | Download Citation



OBJECTIVE: To examine the influence of shift work on metabolic and cardiovascular risk factors in subjects working in an industry sited in Apulia, Southern Italy.

DESIGN: Cross-sectional study of metabolic effects of shift work in glucose tolerant workers in a chemical industry in southern Italy.

SUBJECTS: The subjects included 319 glucose tolerant male individuals, aged 35–60 y.

MEASUREMENTS: Anthropometric parameters (body mass index (BMI) and waist-to-hip ratio (WHR)), fasting concentrations of glucose, insulin, and lipids (total cholesterol, HDL-cholesterol, triglycerides), the sum of glucose levels during 75 g-oral glucose tolerance test (Σ-OGTT), and systolic and diastolic blood pressure (SBP and DBP, respectively).

RESULTS: The prevalence of obesity was higher among shift workers compared to day workers, whereas body fat distribution was not different between the two groups. Shift workers had higher BMI than day workers, and shift working was associated with BMI, independently of age and work duration. Shift workers had significantly higher SBP levels, which were independently influenced by BMI, but not by shift work, thus suggesting that the difference in SBP may well be mediated by the increased body fatness.

CONCLUSION: In workers of an industry sited in Southern Italy, shift work may be directly responsible for increased body fatness and is indirectly associated with higher blood pressure levels and some features of metabolic syndrome.


Shift work is extremely frequent in several services and industries, in order to meet the needs for flexibility of the workforce, necessary to optimize the productivity and the business competitiveness in developed countries, where the proportion of shift workers is estimated to represent >20% of the entire working population. 1 Shift work is associated with several health problems, possibly due to an impairment of biological rhythms. In particular, an increased risk of coronary heart disease (CHD) has been reported in several studies performed in shift workers,2,3,4,5,6,7,8,9,10,11,12 with a direct association between relative risk (RR) for CHD and time of exposure to shift work.4 However, this significant correlation between shift work and CHD mortality has not been found by other groups dealing with this issue.13 Therefore, further insight is needed to elucidate the effect of shift work on cardiovascular risk.

Obesity is a well known independent cardiovascular risk factor14 and, interestingly, it has been shown to be more prevalent among shift workers;15 similarly, a more marked weight gain has been reported in shift working subjects.16,17,18,19 Moreover, abdominal fat accumulation is a cardiovascular risk factor stronger than obesity itself,14 and it is noteworthy that shift workers have been shown to have higher waist-to-hip ratio (WHR) (ie more central fat) than day workers, independently of body mass index (BMI).20

Among several well known cardiovascular risk factors, total cholesterol20,21,22 and triglycerides5,21,23,24 have been shown to be higher in shift workers than in day workers, independently of other lifestyle factors.20 Significant elevations in the serum levels of cholesterol, glucose, uric acid, and potassium have been reported during the first week after a night shift, and this impairment could not be explained by changes in dietary habits or other lifestyle variables.22 Interestingly, shift workers on counter-clockwise rotation are characterized by higher systolic blood pressure (SBP) levels, urinary catecholamines excretion, and plasma levels of triglyceride and glucose compared to shift workers on clockwise rotation.23

All European studies examining the relationship between shift work and risk factors for health problems have been performed in northern and in central Europe, whereas very few studies have been carried out in the Mediterranean area. Therefore, the present study was aimed to evaluate the influence of shift work on metabolic and cardiovascular risk factors in blue collar workers employed in an industry sited in Apulia, Southern Italy. Noteworthy, all subjects enrolled in the study were Caucasians, born and living in Apulia.

Subjects and methods

A total of 718 subjects, aged 35–60 y, were enrolled at the beginning of the study. These subjects were randomly selected among workers involved in the production process of a chemical industry sited in Apulia, representing about 50% of the workforce in the same industry. Before recruitment, all workers were previously informed by a letter about the aims and the methods of the study. All individuals gave written informed consent to be included into the study, which was approved by the Institutional Review Board of the University of Bari School of Medicine and was performed in accordance to the guidelines proposed in the Declaration of Helsinki. All subjects were asked about their work, family and clinical history, and their smoking habits. Since one of the inclusion criteria was to have at least 5 y of working age, that is, a reasonable latency for shift work to exert its effects on cardiovascular risk,19 33 of 718 subjects were excluded from the study since their working age was lower than 5 y. Among the remaining 685 workers, further 32 individuals were excluded because of documented clinical history of diabetes mellitus (n=28) or impaired glucose tolerance (IGT) (n=4).

Therefore, 653 subjects underwent clinical examination and measurement of anthropometric parameters. Waist circumference was measured as the midway between the lower-rib margin and the superior anterior iliac spine. Hip circumference was measured as the widest circumference over the greater trochanter. WHR was also calculated. SBP and diastolic blood pressure (DBP) readings were recorded to the nearest 2 mmHg as the mean of two measurements with the subjects seated and using a mercury manometer with an appropriate cuff size. Hypertension was defined according to WHO criteria.25

A fasting 0800 blood sample was drawn from an antecubital vein for the measurement of blood glucose. In all, five individuals were found to have fasting blood glucose higher than 125 mg/dl and, therefore, they were excluded from the study.

On the day after the first blood sample collection, 648 subjects showing fasting blood glucose levels lower than 126 mg/dl and reporting no clinical history of diabetes mellitus or IGT, underwent a new fasting blood sample collection for measurement of glucose and lipid (total and HDL cholesterol and triglycerides) levels.

Thereafter, a standard 75-g oral glucose tolerance test (OGTT) was performed in these 648 subjects. Normal glucose tolerance (NGT), IGT, and diabetes were defined according to the Recommendations of the American Diabetes Association (ADA) Expert Committee on the Diagnosis Classification and of Diabetes Mellitus.26 After performing OGTTs, one new case of diabetes mellitus and 11 new cases of IGT were diagnosed and were excluded from the study. Therefore, 636 out of 718 workers initially enrolled, free from DM and IGT, were admitted to the next phase of the study. For economic reasons, fasting insulin levels were also measured only in 319 subjects randomly selected among these 636 subjects.

According to the aim of the study, regarding the metabolic characteristics of employees of a chemical industry involved in shift working, the results from the entire group of 636 workers and those from the subgroup of 319 workers with insulin measurement were almost overlapping. Therefore, we decided to present only the results of the subgroup in whom insulin levels have been measured.

Plasma concentrations of insulin were measured by radioimmunological assay, using a commercially available kit (Behring, Scoppitto, Italy). Blood glucose was determined by the glucose-oxidase method (Sclavo, Siena, Italy). Total cholesterol, high-density lipoprotein (HDL) cholesterol, and triglycerides were measured using enzymatic assays (Boehringer Mannheim, GMbH Diagnostica Mannheim, Germany).

Regarding the work organization, 134 of 319 (42%) subjects worked daytime (0700–1600), whereas 185 individuals (58%) worked on three regular rotating shifts. The direction of rotation was counter clockwise (night, night, afternoon, afternoon, morning, morning, rest, rest, rest). Furthermore, the combined effect of shift work and working age (class of working age × shift) was also evaluated.

Statistical analysis was performed using the SAS® program for Windows, SAS Institute Inc. (1995) software (Cary, NC, USA). Results are presented as mean, standard deviation, and range (minimum–maximum) for all parameters. Pearson's correlation coefficients were used to quantify the univariate associations among variables. Comparison of the means of continuous variables between shift workers and day workers was performed by analysis of variance (ANOVA). χ2-test was used to evaluate the differences between the two groups for the categorical variables. Multiple regression analyses were carried out to test the joint effect of different continuous variables on continuous-dependent variables (metabolic and cardiovascular parameters). Variables with a skewed distribution (plasma levels of triglycerides and insulin) were logarithmically transformed before analysis to improve the approximation to a Gaussian distribution. Analysis of covariance was used to evaluate the influence of independent continuous and categorical variables on dependent variables (cardiovascular and metabolic parameters). The minimal statistical significance was defined for P<0.05.


The prevalence of overweight and obesity found in our study sample was 199/319 (62.4%) and 50/319 (15.7%), respectively.

Table 1 shows the differences of continuous variables between day workers and shift workers. BMI, SBP, Σ-OGTT were significantly higher in shift workers. We found no differences between the two groups for the other parameters investigated.

Table 1: Comparison of continuous variables between day workers and shift workers

Obesity was more prevalent in shift workers (37/185=20.0%) than in day workers (13/134=9.7%) (χ2 test=8.64; P<0.05), whereas body fat distribution was not different between the two groups. Shift workers were more frequently smokers (74/185=40.0%) or nonsmokers (71/185=38.4%) than day workers (46/134=34.3 and 43/134=32.1%, respectively), even though this difference did not reach statistical significance (χ2 test=5.7; P=0.058). The prevalence of hypertension was not different between day workers (15/134=11.2%) and shift workers (26/185=14.1%) (χ2 test=0.57; NS).

Table 2 shows the values of BMI distributed by shift and class of working age in day workers and shift workers.

Table 2: Means and s.d. of BMI distributed for classes of working age in day workers and shift workers (319 subjects), in whom insulinaemia was measured

Analysis of covariance was performed to verify the effect of shift work and working age on study parameters, with each variable being controlled for all the others that could significantly influence the results. In particular, only statistical models having at least one significant relationship between the dependent variable and the independent variables are presented. Shift work, working age, and class of working age × shift are always shown in each statistical model; concerning the other independent variables, only those having a significant relationship with the dependent variables are shown.

Table 3 shows the results of the analysis of covariance. In particular, we found a significant relationship between shift work (but not working age) and BMI, even after taking into account fasting insulin levels. Both SBP and DBP were significantly influenced by BMI, working age, and smoking, whereas only SBP was influenced by Σ-OGTT. Finally, a direct relationship between working age and triglyceride plasma levels was found, also after controlling for fasting insulin.

Table 3: Analysis of covariance in the 319 workers in whom fasting insulinaemia was measured


At the best of our knowledge, this is the first study performed in Italy examining simultaneously the effect of shift work on BMI, insulin levels, and metabolic and cardiovascular risk factors in men. Furthermore, this is the first study examining these relationships in a Mediterranean area.

We show that shift workers have higher BMI than day workers, and shift working is associated with BMI, independently of age and work duration. This is an important finding since a higher BMI is associated not only with an increased risk for morbidity and mortality, but also with accidents and injuries at work.14,27 Our results are consistent with previous studies showing that the prevalence of obesity is higher in shift workers15 and that weight gain occurs in late-shift workers (evening and night),16 cleanroom workers changing from an 8 to a 12-h shift,17 and offshore workers under continued exposure to day–night shift work.18 Other evidence highlights a positive relationship between BMI and duration of shift work exposure.19 In particular, employees involved in shift working for more than 5 y had significantly higher BMI than those with no shift work experience.19 Weight gain in shift workers has been explained by several mechanisms, such as higher calorie intake,16 changes both in dietary habits (such as eating fewer meals and more snacks) and in the circadian distribution of food intake,16,28,29 lower physical exercise,16,30 and changes in sleeping habits.16

Moreover, rotating shift work has been shown to increase the daily cortisol secretion31 and to dissociate the cortisol circadian rhythm, with a progressive rise in cortisol levels during sleep (that is, normally, a resting phase).32,33,34 This seems to be an important point, if it is taken into account that central obesity has been suggested to have a hypothalamic neuroendocrine origin, with increased ACTH and cortisol secretion.35,36

Unfortunately, we examined neither cortisol levels nor dietary habits and alcohol intake in the present survey; therefore, no clear-cut conclusion can be drawn with this respect. However, a subgroup of subjects were investigated by a dietitian (data not shown); this interview clearly showed that the common meal consumed during the night shift was poor in fibres and rich in animal proteins, saturated fat (cheese, ham, eggs, etc), and foods with high glycaemic index (white bread, sweets, etc), that is, a kind of food that is well known to predict an increase of body fat in working individuals.30

This is one of the rare studies examining insulin levels in workers, and it should be noted that the relationship between BMI and shift work was maintained also after including insulin levels in the statistical model, thus possibly excluding that weight gain in shift workers may be mainly mediated by increasing insulin levels.

Even though BMI was higher in shift workers, WHR was not different between the two groups. Noteworthy, since the mean BMI and WHR values were quite high in both the groups, it may well be that WHR is not a sensitive parameter to detect possible small changes in body fat distribution in a population with such a high prevalence of overweight and obesity.

A further significant finding of this study was that shift workers had a higher Σ-OGTT. Σ-OGTT was independently associated with BMI, insulinaemia, and triglycerides. Since increased insulin and triglyceride levels are established features of the insulin resistance syndrome, higher glucose levels in shift workers are likely to reflect such a condition. However, since shift work was not associated with Σ-OGTT, the impairment of glucose metabolism in shift workers is likely to reflect a condition of insulin resistance, rather than a direct effect of shift work in itself. The importance of these findings derives from the fact that both body fatness and insulin resistance accelerate the development of atherosclerosis independently of each other.37,38

On the other hand, insulin levels were significantly associated with BMI, but not with shift work. Therefore, in our opinion, it seems possible that shift work is responsible for higher BMI and, in turn, higher body fatness causes insulin sensitivity to decrease and glucose metabolism to be altered. Again, increased cortisol secretion may play a central role, since higher cortisol production can cause insulin resistance in muscles.35,36

Even though the prevalence of hypertension was not different between shift workers and day workers, the former had significantly higher SBP levels. However, SBP was influenced by BMI, but not by shift work, in our study population; this result seems to suggest that the difference in blood pressure levels between shift workers and day workers may well be mediated by the increased body fatness. However, we cannot exclude that the stress induced by shift work, increased cortisol, and/or catecholamine secretion, higher fat and salt intake, and sedentary habits may contribute to the increase in blood pressure levels.

Total cholesterol, triglyceride, and HDL-cholesterol concentrations were not different between shift workers and day workers. These results are apparently in contrast with previous studies showing higher plasma levels of total cholesterol20,21,22 and triglycerides5,21,23,24 in shift workers. Since this is the first study performed in a Mediterranean area, it may well be that different dietary habits or genetic factors linked to geographical origin of the study subjects may be responsible for the lack of this difference.

Both triglyceride and HDL-cholesterol levels were independently influenced by insulin levels, thus confirming the tendency for insulin resistance to cluster with higher triglyceride and lower HDL-cholesterol concentrations (metabolic syndrome).

In synthesis, our study shows that shift work acts as a risk factor for obesity and metabolic and cardiovascular diseases even in the Mediterranean area, that is, theoretically protected by more healthy dietary habits compared to Northern Europe. Therefore, our findings, along with those from other authors, strengthen the need for preventive educational programmes concerning the changes in lifestyle (diet and physical activity), performed by occupational physicians, to be applied in all workers, and particularly in shift workers, of industrialized countries.

In conclusion, the present study, performed only in blue collar men working in an industry of Southern Italy, shows that shift work may be directly responsible for increased body fatness and is indirectly associated with higher blood pressure levels and some features of the metabolic syndrome.


  1. 1.

    , , , . The prevalence and health impact of shiftwork. Am J Publ Health 1986; 76: 1225–1228.

  2. 2.

    , . Job characteristcs of occupations and myocardial infarction risk: effect of possible confounding factors. Soc Sci Med 1983; 17: 1497–1503.

  3. 3.

    , , . Type of occupation and near-future hospitalization for myocardial infarction and some other diagnosis. Int J Epidemiol 1985; 14: 378–388.

  4. 4.

    , , , . Increased risk of ischaemic heart disease in shift workers. Lancet 1986; 2: 89–92.

  5. 5.

    , , . Prevalence of risk factors for coronary disease among day and shift workers. Scand J Work Environ Health 1988; 14: 317–321.

  6. 6.

    . Shift work and coronary heart disease. Scand J Soc Med 1989; 44(Suppl): 1–36.

  7. 7.

    . Cardiovascular diseases and work environment. A critical review of the epidemiologic literature on non chemical factors. Scand J Work Environ Health 1989; 15: 165–179.

  8. 8.

    , , , . Risk indicators of ischemic heart disease among male professional drivers in Sweden. Scand J Work Environ Health 1993; 19: 326–333.

  9. 9.

    , , , , , , . Prospective study of shift work and risk of coronary heart disease in women. Circulation 1995; 92: 3178–3182.

  10. 10.

    , , , , . Shift work, occupation and coronary heart disease over 6 years of follow-up in the Helsinki Heart Study. Scand J Work Environ Health 1997; 23: 257–265.

  11. 11.

    , , , , . Shift work and myocardial infarction: a case–control study. Occup Environ Med 1999; 56: 46–50.

  12. 12.

    . Working hours and ischemic heart disease in Danish men: a 4-year cohort study of hospitalization. Int J Epidemiol 1993; 22: 215–221.

  13. 13.

    , , , , , . Shiftwork and mortality from ischemic heart disease. Occup Environ Med 1996; 53: 367–373.

  14. 14.

    . Obesity: a chronic disease with alarming prevalence and consequences. J Intern Med 1998; 244: 267–269.

  15. 15.

    , , . Is there an association between shift work and having a metabolic syndrome? Results from a population based study of 27485 people. Occup Environ Med 2001; 58: 747–752.

  16. 16.

    , , , , . Work-shift period and weight change. Nutrition 2000; 16: 27–29.

  17. 17.

    , , , , , . Excessive fatigue and weight gain among cleanroom workers after changing from an 8-hour to a 12-hour shift. Scand J Work Environ Health 2001; 27: 318–326.

  18. 18.

    . Shift work and age as interactive predictors of body mass index among offshore workers. Scand J Work Environ Health 2002; 28: 64–71.

  19. 19.

    , , . Duration of shiftwork related to body mass index and waist to hip ratio. Int J Obes Relat Metab Disord 1999; 23: 973–978.

  20. 20.

    , , , , , , , , , . Shift work and risk factors for coronary heart disease in Japanese blue-collar workers: serum lipids and anthropometric characteristics. Occup Med 1997; 47: 142–146.

  21. 21.

    , , , . The Tromso heart study. Acta Medica Scand 1976; 200: 107–118.

  22. 22.

    , . Day and night work: changes in cholesterol, uric acid, glucose and potassium in serum and circadian patterns of urinary catecholamine excretion. Acta Med Scand 1976; 200: 47–53.

  23. 23.

    . Intervention or coronary risk factors by adapting a shift work schedule to biological rhythmicity. Psychosom Med 1983; 45: 407–415.

  24. 24.

    , , , , , , . Increased triglyceride levels in shift workers. Am J Med 1992; 93: 259–262.

  25. 25.

    , , , , , , , , , . World Health Organization — International Society of Hypertension Guidelines for the Management of Hypertension. Guidelines sub-committee of the World Health Organization. Clin Exp Hypertens 1999; 21: 1009–1060.

  26. 26.

    The Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care 2002; 25(Suppl 1): S5–S20.

  27. 27.

    , , , , . Industrial accidents are related to relative body weight: the Israeli CORDIS study. Occup Environ Med 1996; 53: 832–835.

  28. 28.

    . Do eating and drinking habits interact with work schedule variables? Work Stress 1990; 4: 203–211.

  29. 29.

    , , . Shift related dietary intake in day and shift workers. Appetite 1995; 25: 253–265.

  30. 30.

    , , , , , . Predictors of weight change over two years among a population of working adults: the Healthy Worker Project. Int J Obes Relat Metab Disord 1994; 18: 145–154.

  31. 31.

    . Alteration of circadian rhythm in shift working ambulance personnel. Monitoring of salivary cortisol rhythm. Ergonomics 1992; 35: 1331–1340.

  32. 32.

    , , , , , , . Effect of shift work on the night-time secretory patterns of melatonin, prolactin, cortisol, and testosterone. Eur J Appl Physiol Occup Physiol 1990; 60: 288–292.

  33. 33.

    , , . Effects of total sleep deprivation on urinary cortisol self rated arousal and mood in depressed patients. Psichiatr Res 1990; 34: 149–162.

  34. 34.

    , , , , . Internal dissociation of the circadian markers of the cortisol rhythm in night workers. Am J Physiol 1996; 270: E608–E613.

  35. 35.

    , . Hypothalamic origin of the metabolic syndrome X. Ann NY Acad Sci 1999; 892: 297–307.

  36. 36.

    , . Neuroendocrine abnormalities in visceral obesity. Int J Obes Relat Metab Disord 2000; 24(Suppl 2): S80–S85.

  37. 37.

    , , , , , , , . Plasma leptin is independently associated with the intima-media thickness of the common carotid artery. Int J Obes Relat Metab Disord 2001; 25: 805–810.

  38. 38.

    , , , , , , , . Lower insulin sensitivity as an independent risk factor for carotid wall thickening in normotensive, non-diabetic, non-smoking normal weight and obese premenopausal women. Int J Obes Relat Metab Disord 2000; 24: 825–829.

Download references

Author information


  1. Occupational Health, Department of Internal Medicine and Public Health, University of Bari, Bari, Italy

    • L Di Lorenzo
    • , N L'Abbate
    • , A Basso
    •  & L Soleo
  2. Internal Medicine, Endocrinology, and Metabolic Diseases, Department of Emergency and Organ Transplantation, University of Bari, Bari, Italy

    • G De Pergola
    • , N Pannacciulli
    •  & R Giorgino
  3. Health Directorate, Lombardy Region, Milan, Italy

    • C Zocchetti
  4. Institute of Endocrinology, University of Foggia, Foggia, Italy

    • M Cignarelli


  1. Search for L Di Lorenzo in:

  2. Search for G De Pergola in:

  3. Search for C Zocchetti in:

  4. Search for N L'Abbate in:

  5. Search for A Basso in:

  6. Search for N Pannacciulli in:

  7. Search for M Cignarelli in:

  8. Search for R Giorgino in:

  9. Search for L Soleo in:

Corresponding author

Correspondence to G De Pergola.

About this article

Publication history






Further reading