You are here

Article Text

Article menu

Download PDFPDF

Workplace
Original article
Night shift work at specific age ranges and chronic disease risk factors
  1. Cody Ramin1,
  2. Elizabeth E Devore1,
  3. Weike Wang2,
  4. Jeffrey Pierre-Paul3,
  5. Lani R Wegrzyn2,
  6. Eva S Schernhammer1,2,4,5
  1. 1Channing Division of Network Medicine, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
  2. 2Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA
  3. 3Massachusetts College of Pharmacy and Health Sciences, Boston, Massachusetts, USA
  4. 4Division of Sleep Medicine, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA
  5. 5LBI-ACR & ACR-ITR VIEnna/CEADDP, Vienna, Austria
  1. Correspondence to Eva S Schernhammer, Channing Laboratory, 181 Longwood Avenue, Boston, MA 02115, USA; eva.schernhammer{at}channing.harvard.edu

Abstract

Objectives We examined the association of night shift work history and age when night shift work was performed with cancer and cardiovascular disease risk factors among 54 724 women in the Nurses’ Health Study (NHS) II.

Methods We calculated age-adjusted and socioeconomic status-adjusted means and percentages for cancer and cardiovascular risk factors in 2009 across categories of night shift work history. We used multivariable-adjusted logistic regression to estimate odds ratios (ORs) and 95% CIs for key risk factors among 54 724 participants (72% ever shift workers). We further examined these associations by age (20–25, 26–35, 36–45 and 46+ years) at which shift work was performed.

Results Ever night shift workers had increased odds of obesity (body mass index ≥30 kg/m2; OR=1.37, 95% CI 1.31 to 1.43); higher caffeine intake (≥131 mg/day; OR=1.16, 95% CI 1.12 to 1.22) and total calorie intake (≥1715 kcal/day; OR=1.09, 95% CI 1.04 to 1.13); current smoking (OR=1.30, 95% CI 1.19 to 1.42); and shorter sleep durations (≤7 h of sleep/day; OR=1.19, 95% CI 1.15 to 1.24) compared to never night shift workers. These estimates varied depending on age at which night work was performed, with a suggestion that night shift work before age 25 was associated with fewer risk factors compared to night shift work at older ages.

Conclusions Our results indicate that night shift work may contribute to an adverse chronic disease risk profile, and that risk factors may vary depending on the age at which night shift work was performed.

https://doi.org/10.1136/oemed-2014-102292

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    What this paper adds

    • Recent studies have suggested that working at night is associated with an adverse risk profile for chronic diseases.

    • Adding to this data, we evaluated the effects of night shift work and timing by specific age ranges on chronic disease risk factors among 54 724 women in the Nurses’ Health Study II.

    • Our findings indicate that targeted lifestyle interventions for night shift workers should potentially take into account age when shift work is performed, as risk factors may vary over the course of a woman's shift work career.

    Introduction

    Shift work is common in many occupations in modern society, particularly in service industries including healthcare, manufacturing and transportation. According to the Current Population Survey,1 approximately 18% of the US labour force works alternative shifts that fall outside of a traditional day shift. While studies indicate that shift work (a proxy for light exposure at night and chronodisruption) may have adverse effects on acute health-related outcomes (eg, accidents, reproductive factors, gastrointestinal malfunction and sleep),2–5 associations between night shift work and chronic diseases are less clear and difficult to study given the need for detailed exposure assessment and long follow-up to ascertain chronic disease outcomes.6 ,7 A recent study8 of over 40 000 UK women found that working at night was associated with higher odds of smoking, obesity and being in the lowest third of socioeconomic status (SES), suggesting a more adverse risk profile for chronic diseases among shift workers.

    Adding to this data, we examined the distribution of risk factors for cancer and cardiovascular disease—the most common chronic diseases and leading causes of mortality in the USA9—by night shift work history in the Nurses’ Health Study (NHS) II, a large cohort of women with substantial shift work exposure, long-term follow-up, and a large number of reported health and lifestyle characteristics. Given the consistency of night shift work from young adulthood through older age among NHS II participants, we are in a unique position to evaluate night shift work and timing by specific age ranges in relation to chronic disease risk factors.

    Materials and methods

    NHS II was initiated in 1989 when 116 430 female registered nurses, aged 25–42 years and living in 14 US states, completed an initial questionnaire on their medical history, health and lifestyle. Since 1989, similar questionnaires have been completed biennially to update this information, with follow-up rates at approximately 90%. This study was approved by the Institutional Review Board (IRB) of Brigham and Women's Hospital (Boston, Massachusetts, USA).

    Population for analysis

    In 2009, 90 482 women completed the NHS II cohort questionnaire; of these, 71 997 recalled information on primary work schedule for each prespecified age range (from ages 20 to 25, 26–35, 36–45 and 46+ years). Thus, we have shift work information on each time period during a woman's working life, as it was recalled in 2009. Women who provided work schedule information were generally similar to women who did not provide work schedule information (eg, mean age=55.2 vs 54.2 years, mean body mass index=27.8 vs 27.7 kg/m2, respectively). To create unambiguous exposure groups and reduce possible misclassification, we restricted the population for analysis to women whose responses to multiple questions about work schedule histories were internally consistent; thus, we excluded 16 877 women who reported a primary non-night shift schedule with ≥1 night shifts per month during the specified age range, and 396 women who reported a primary night shift schedule and no night shifts per month during the specified age range. As a result, there were 54 724 women left in our analytic sample.

    Assessment of shift work

    Participants reported detailed information on occupational history by age periods of their working life, including their primary shift work schedule, in 2009. Work schedule for each specific age range (20–25 years, 26–35 years, 36–45 years and 46+ years) was queried by asking women to report: ‘Your primary work schedule during each age range (Consider your schedule ‘day/evening’ if most work hours were between 7:00–15:00, or 15:00–23:00, ‘night’ if 23:00—7:00; and ‘early morning’ if 4:00—9:00)’. Response categories were given as ‘day/evenings only, nights only, early mornings only, rotating with nights, rotating with no nights, or other/didn't work’. In 2009, women also recalled, for each age range, the type of nursing occupation, years worked at occupation, full-time/part-time work, total years of rotating night shifts and average night shifts per month.

    Covariates of interest

    Potential cancer and cardiovascular disease risk factors were derived from the main NHS II questionnaires and were determined a priori based on evidence from published literature. We grouped potential risk factors into the following categories: modifiable, non-modifiable and health-related risk factors.

    Modifiable risk factors, including body mass index (BMI, in kg/m2), physical activity (MET-hours per week), smoking status, oral contraceptive use and postmenopausal hormone use were assessed in 2009; BMI at age 18 was assessed in 1989. Average hours of sleep over a 24-hour period were reported by age range of shift work in 2009. Dietary factors including alcohol (g/day), caffeine consumption (mg/day) and total energy intake (kcal/day) were assessed in 2007 because this was the cohort questionnaire most proximal to the reporting of shift work history. The Alternative Healthy Eating Index score (AHEI) was also assessed in 2007 and reflects adherence to a diet pattern based on foods and nutrients most predictive of disease risk; total scores range from 0 (non-adherence) to 110 (perfect adherence).10

    Non-modifiable risk factors, including age (years), menopausal status, age at menopause (years), parity, chronotype, living alone, and allergies were assessed in 2009; age at menarche was assessed in 1989. Nurses also reported their spouse/partner's education level in 1999 (high school or less; 2 or 4-year college; graduate school; or not married/missing) as an indicator of SES.

    Health-related risk factors, including medication use (ie, antihypertensives and aspirin), medical visits in the past 2 years (ie, mammography, physical examination and colonoscopy/sigmoidoscopy), and hypercholesterolaemia (blood cholesterol ≥200 mg/dL) were assessed in 2009; multivitamin use was assessed in 2007. History of diabetes, angina and high blood pressure, and family history of myocardial infarction and/or cancer, except non-melanoma skin cancer, were assessed in 1989 and information was updated subsequently on most of the biennial follow-up questionnaires.

    Additional covariates assessed for secondary analyses include work stress and social support. Work stress was assessed in 1993 and 1997 with the 27-item Karasek Job Content Questionnaire,11 which measures the psychological work load ‘demands’ and level of ‘control’ to manage the workload. Based on the job demand/control model, jobs are categorised into four classes of work stress: (1) jobs that are high demand/low control (‘high strain’), (2) jobs that are low demand/high control (‘low strain’), (3) jobs that are high demands/high control (‘active’) and (4) jobs that are low demand/low control (‘passive’).12 ,13 In 1993, social support was assessed when participants were asked if they had a close confidant (eg, someone who they can share confidences and feelings with).

    Statistical analysis

    Among the 54 724 women in our analysis, 15 391 women reported never working night shifts and 39 333 women reported working night shifts at some point during their career. Women who reported a primary shift schedule with rotating night shifts or nights only (ie, permanent nights) were categorised as ‘ever’ night shift workers. We calculated means and proportions of self-reported occupational characteristics across prespecified age ranges at which shift work was performed (age 20–25, 26–35, 36–45 and 46+; table 1). We then estimated age-adjusted (in 5-year increments) and SES-adjusted means and percentages of cancer and cardiovascular risk factors (modifiable, non-modifiable and health-related factors) for ever vs never night shift work. To evaluate which risk factors differed significantly by shift work status, we calculated two-sided p values using PROC GLM for continuous covariates and PROC CATMOD for categorical covariates (table 2). To correct for multiple comparisons, we used the Bonferroni adjustment on the basis of the number of tested covariates. P<αc (Bonferroni-corrected α) were considered statistically significant.

    View this table:
    Table 1

    Occupational characteristics of 54 724 women in the Nurses’ Health Study II across key age periods of their working life*,†,‡

    View this table:
    Table 2

    Modifiable, non-modifiable and health-related risk factors in 2009 across ever versus never night shift workers in Nurses’ Health Study II (N=54 724)†

    Next, we used multivariable-adjusted logistic regression models to estimate odds ratios (ORs) and 95% CIs for modifiable risk factors whose age-adjusted and SES-adjusted means or percentages differed significantly by shift work status, dichotomising continuous outcomes (eg, caffeine and calorie intake) along the median of never night shift workers (table 3). To determine the importance of age at which night shift work occurred, we first examined these associations overall according to night shift work (ever/never) (table 3) and then separately examined them among rotating and nights only shift work for each respective age range (tables 4 and 5). Logistic regression models were adjusted for age and SES for comparability to the Million Women Study8 (model 1); and additional covariates that produced a 5% or greater change of the β coefficient representing the exposure effect14 (model 2). To evaluate the association between shift work at specific ages and obesity, we additionally adjust for BMI at age 18 to provide estimates that are independent of earlier BMI (model 3). Based on the hypothesis that never night shift workers might have the least amount of chronodisruption, these women were the reference group in all analyses.

    View this table:
    Table 3

    ORs and 95% CIs of modifiable risk factors in 2009 across ever vs never night shift work in Nurses’ Health Study II (N=54 724)

    View this table:
    Table 4

    Odds ratios and 95% CIs of modifiable risk factors in 2009 across age ranges of rotating night shift work (as recalled in 2009) in Nurses’ Health Study II

    View this table:
    Table 5

    Odds ratios and 95% CIs of modifiable risk factors in 2009 across age ranges of night only shift work (as recalled in 2009) in Nurses’ Health Study II

    We conducted several secondary analyses. Given that psychosocial factors have been associated with adverse health effects, we adjusted for work stress (high strain, low strain, passive and active job) and social support (eg, close confidant) (yes, no). We also adjusted for type of nursing occupation (inside hospital, outside hospital, non-nursing/other) since work demand may vary by type of nursing occupation. To assess whether associations may be explained by either shift work intensity or shift work duration, we also conducted separate models that adjusted for average number of night shifts reported per month (0, 1–2, 3–4, 5–6, 7–8, 9–10, 11–15, 16–20, 21+ night shifts per month) and total years of night shift work (0, 1–<10, 10–<20, 20+ years).

    For all analyses, we report multivariable-adjusted results, which were similar to age-and-SES adjusted results. All analyses were conducted with SAS software, V.9.3 (SAS Institute Inc, Cary, North Carolina, USA).

    Results

    Table 1 describes occupational characteristics across age ranges that shift work was performed among 54 724 women in NHS II. Qualitatively, there were several important differences across these age ranges; specifically, as age at shift work increased, women were more likely to work in nursing careers outside of the hospital (3.3% among women aged 20–25 vs 16.5% among women aged 46+), more likely to have a primary day/evening work schedule (44.6% among women aged 20–25 vs 80% among women aged 46+), and less likely to work full-time (84% among women aged 20–25 vs 72.7% among women aged 46+). Additionally, more women worked rotating night shifts earlier in life compared to later in life (36.6% among women aged 20–25 vs 5.3%, among women aged 46+).

    After adjusting estimates for age and SES and applying a Bonferonni correction, we observed several significant differences across chronic disease risk factors comparing ever vs never night shift workers (table 2). Among modifiable risk factors, ever night shift workers were more likely to have a higher BMI (28.1 vs 27 kg/m2), higher daily caloric (1822 vs 1772 kcal/day) and caffeine intake (184 vs 167 mg/day), shorter sleep duration (67% vs 63% ≤7 h/day), and to currently smoke (7% vs 5%) compared to never night shift workers. Among non-modifiable risk factors, compared to never night shift workers, ever night shift workers were more likely to be nulliparous (19% vs 17%), live alone (13% vs 11%) and less likely to be morning chronotypes (33% vs 38%). Among health-related factors, ever night shift workers were also more likely to have a history of diabetes (8% vs 6%), angina (3% vs 2%) and high blood pressure (36% vs 33%); regularly use antihypertensive medication (34% vs 32%) and aspirin (11% vs 9%), and less likely to have had a mammography in the past 2 years (90% vs 92%), compared to never night shift workers (table 2).

    In table 3, we observed that night shift workers had significantly higher odds of obesity (defined as BMI ≥30 kg/m2; OR=1.37, 95% CI 1.31 to 1.43), higher caffeine (defined as ≥131 mg) (OR=1.16, 95% CI 1.12 to 1.22) and total calorie intake (defined as ≥1715 kcal per day; OR=1.09, 95% CI 1.04 to 1.13), less sleep (defined as ≤7 h per day; OR=1.19, 95% CI 1.15 to 1.24), and being a current smoker (OR=1.30, 95% CI 1.19 to 1.42) compared to never night shift workers. Further, odds of modifiable risk factors were generally stronger among ever nights only shift workers at any age range compared to ever rotating night workers, and varied depending on age at which night work was performed (tables 4 and 5). Our data suggests that rotating night work was significantly associated with BMI earlier in life (age 26–35) and later in life (age 46+) and smoking behaviour earlier in life (age 26–35) and mid-life (age 36–45; table 4). Nights only shift work was associated with BMI and smoking across almost all age ranges of working night only shifts; however, total calories, caffeine intake and sleep were generally not associated with night only work at any age ranges (table 5). In general, odds of risk factors were lower among women who worked rotating or night only shifts early in life (age 20–25), although this pattern was not entirely consistent.

    In models that additionally adjusted for work stress, social support, type of nursing occupation and total years of night shift work, results did not substantially change (data not shown). Higher levels of average night shifts per month were significantly associated with an increased risk of obesity (eg, multivariable OR (95% CI) for 1–2 nights/month vs 0 nights/month=1.13 (0.70 to 1.83); 3–4 nights/month vs 0 nights/month=1.38 (0.85 to 2.22), 5–6 nights/month vs 0 nights/month=1.65 (1.02 to 2.66), 7–8 nights/month vs 0 nights/month=1.85 (1.15 to 2.99), 9–10 nights/month vs 0 nights/month=1.85 (1.14 to 3.00), 11–15 nights/month vs 0 nights/month=2.25 (1.39 to 3.64), 16–20 nights/month vs 0 nights/month=2.28 (1.39 to 3.75), 21+ nights/month vs 0 nights/month=3.42 (1.95 to 6.03); p-trend<0.0001). Furthermore, the increased risk of obesity was present across age categories among rotating and night only shift workers.

    Discussion

    In this study of US female nurses, women were more likely to report a primary rotating night-shift schedule, full-time work and a nursing occupation within a hospital setting earlier in life compared to later in life. We also found that night shift workers tended to have a more adverse risk profile for chronic diseases compared to never night shift workers, and risk factors varied depending on age at which night work was performed.

    Epidemiological studies, including previous studies in the NHS cohorts,15–17 have observed differences in several established cancer and cardiovascular risk factors among those with versus without a history of shift work, including a higher BMI,18–24 dietary intake25 and smoking,21 ,26–29 as well as shorter duration of sleep.30 ,31 Exact mechanisms between shift work and these key risk factors are not well established. Several putative mechanisms have been suggested, including increased circadian rhythm disruption and disturbed sociotemporal patterns5 ,18 ,32 ,33; both can lead to behavioural changes in dietary intake, smoking and duration of sleep. These key risk factors may lead to unfavourable metabolic disturbances contributing to increased obesity among night shift workers. In line with our results, a recent study8 that examined characteristics of 41 652 women in the Million Women Study, a cohort with 13.2% of participants reporting night shift work, found that women with a history of night shift work had a higher BMI and were more likely to be current smokers compared to those without a history of night shift work (mean BMI=27.3 vs 26.6 kg/m2; percent smokers=8.6% vs 6.2%, respectively). This study did not assess the association between night shift work and daily caffeine or total calorie intake.

    Interestingly, we found that higher intensity of night shift work was associated with an increased risk of obesity, suggesting that increasing average number of night shifts worked per month (possibly indicative of more severe circadian disruption) might be an important risk factor for obesity among shift workers. This is also reflected in the stronger ORs for obesity we observed among women who reported night only shift work as their primary work schedule (table 5), compared to those with rotating night shift work as their primary work schedule (table 4). Additional studies are warranted to confirm our finding, and to also explore the role of individual chronotype in these associations; that is, whether a shift schedule that is in line with a person's chronotype (eg, night work carried out by evening types) may less severely affect obesity and related outcomes than a shift schedule opposing chronotype (eg, early morning work schedules for morning types).

    While it appears plausible that the age at which an individual performed shift work may modulate the effects of shift work on health outcomes, no other prior study, to our knowledge, has examined this particular aspect of shift work among women. Previous data from the NHS cohorts indicate that longer durations of shift work are associated with higher cancer and CVD risk, whereas shorter durations do not appear to increase these risks.15–17 ,34 ,35 This data could provide indirect evidence for a higher risk associated with shift schedules worked earlier in life since duration could simply be an indicator of exposure earlier in life. However, in our current study, although adverse risk profiles appeared to vary slightly depending on the age at which a woman worked night shifts, and were stronger among women with night work only, versus those with rotating night shift worker, no clear pattern emerged in support of this.

    The observed differences we found in chronotype among shift workers (ie, a lower proportion of morning chronotypes among night shift workers) were consistent with previous publications.8 ,36 ,37 Recent studies36 ,37 have provided evidence that evening or ‘neither’ (ie, neither morning nor evening) chronotype may be associated with an increased risk of breast cancer compared to morning chronotype. While further research is warranted to validate these findings, it is nevertheless interesting that we observed a lower prevalence of morning types and less favourable health characteristics among night shift workers.

    Strengths of this study include its size, wide variety of health and lifestyle characteristics, and detailed shift work information specific to the ages when work was performed. There are also several limitations of this study. First, these analyses are cross-sectional and do not necessarily have a causal interpretation. However, our results are intended to inform future prospective studies in exploring associations between night shift work and chronic disease risk. Second, our analyses utilise information on women's risk factor profiles from later in life, and therefore shorter term versus Longer term effects of shift work are not differentiated. Third, assessment of shift work was self-reported and ascertained retrospectively; hence, this exposure is susceptible to random misclassification. To alleviate some of this concern, we excluded women with inconsistent shift work histories in our analysis. If any misclassification is present, it is likely non-differential; however, given the nature of shift work, others have also reported the possibility of differential misclassification.38 ,39 Finally, the study population consists primarily of white female nurses and may not be generalisable to other populations, such as men and different racial groups. Future studies should explore associations between shift work and chronic disease risk factors in diverse populations.

    Taken together, our findings suggest that night shift work may be correlated with an adverse risk profile for chronic disease and that differences in risk profiles may exist, depending on the age range at which night shift work occurs. Differences in obesity risk suggested the most clinical relevance. Future prospective research is needed to confirm if timing of night shift work may impact adverse health effects. If confirmed, these findings may indicate that targeted lifestyle interventions for night shift workers should take into account the age at which night shift work starts and/or occurs, as risk factors may vary over the course of a woman's shift work career.

    Acknowledgments

    The authors would like to thank the participants and staff of the Nurses’ Health Study II cohort for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, WY.

    References

    1. ↵
      1. McMenamin TM
      . A time to work: recent trends in shift work and flexible schedules. Monthly Lab Rev 2007;130:3–15.
      OpenUrl
    2. ↵
      1. Akerstedt T
      . Psychological and psychophysiological effects of shift work. Scand J Work Environ Health 1990;16:67–73.
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      1. Bonzini M,
      2. Palmer KT,
      3. Coggon D, et al
      . Shift work and pregnancy outcomes: a systematic review with meta-analysis of currently available epidemiological studies. BJOG 2011;118:1429–37.
      OpenUrlCrossRefPubMed
    4. ↵
      1. Gold DR,
      2. Rogacz S,
      3. Bock N, et al
      . Rotating shift work, sleep, and accidents related to sleepiness in hospital nurses. Am J Public Health 1992;82:1011–14.
      OpenUrlCrossRefPubMedWeb of Science
    5. ↵
      1. Knutsson A
      . Health disorders of shift workers. Occup Med 2003;53:103–8.
      OpenUrlAbstract
    6. ↵
      1. Knutsson A
      . Methodological aspects of shift-work research. Chronobiol Int 2004;21:1037–47.
      OpenUrlCrossRefPubMed
    7. ↵
      1. Stevens RG,
      2. Hansen J,
      3. Costa G, et al
      . Considerations of circadian impact for defining ‘shift work’ in cancer studies: IARC Working Group Report. Occup Environ Med 2011;68:154–62.
      OpenUrlAbstract/FREE Full Text
    8. ↵
      1. Wang XS,
      2. Travis RC,
      3. Reeves G, et al
      . Characteristics of the Million Women Study participants who have and have not worked at night. Scand J Work Environ Health 2012;38:590–9.
      OpenUrlCrossRefPubMedWeb of Science
    9. ↵
      1. Hoyert DL,
      2. Xu JQ
      . Deaths: preliminary data for 2011. Natl Vital Stat Rep 2012;61:1–51.
      OpenUrlPubMed
    10. ↵
      1. Chiuve SE,
      2. Fung TT,
      3. Rimm EB, et al
      . Alternative dietary indices both strongly predict risk of chronic disease. J Nutr 2012;142:1009–18.
      OpenUrlAbstract/FREE Full Text
    11. ↵
      1. Karasek R,
      2. Theorell T
      . Healthy work: stress, productivity, and the reconstruction of working life. New York, NY: Basic Books, 1990.
    12. ↵
      1. Kroenke CH,
      2. Spiegelman D,
      3. Manson J, et al
      . Work characteristics and incidence of type 2 diabetes in women. Am J Epidemiol 2007;165:175–83.
      OpenUrlAbstract/FREE Full Text
    13. ↵
      1. Schernhammer ES,
      2. Hankinson SE,
      3. Rosner B, et al
      . Job stress and breast cancer risk: the Nurses’ Health Study. Am J Epidemiol 2004;160:1079–86.
      OpenUrlAbstract/FREE Full Text
    14. ↵
      1. Greenland S
      . Modeling and variable selection in epidemiologic analysis. Am J Public Health 1989;79:340–9.
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      1. Kawachi I,
      2. Colditz GA,
      3. Stampfer MJ, et al
      . Prospective Study of Shift Work and Risk of Coronary Heart Disease in Women. Circulation 1995;92:3178–82.
      OpenUrlAbstract/FREE Full Text
    16. ↵
      1. Schernhammer ES,
      2. Kroenke CH,
      3. Laden F, et al
      . Night work and risk of breast cancer. Epidemiology 2006;17:108–11.
      OpenUrlCrossRefPubMedWeb of Science
    17. ↵
      1. Schernhammer ES,
      2. Laden F,
      3. Speizer FE, et al
      . Rotating night shifts and risk of breast cancer in women participating in the Nurses’ Health Study. J Natl Cancer Inst 2001;93:1563–8.
      OpenUrlAbstract/FREE Full Text
    18. ↵
      1. Antunes LC,
      2. Levandovski R,
      3. Dantas G, et al
      . Obesity and shift work: chronobiological aspects. Nutr Res Rev 2010;23:155–68.
      OpenUrlCrossRefPubMed
    19. ↵
      1. Di Lorenzo L,
      2. De Pergola G,
      3. Zocchetti C, et al
      . Effect of shift work on body mass index: results of a study performed in 319 glucose-tolerant men working in a Southern Italian industry. Int J Obes Relat Metab Disord 2003;27:1353.
      OpenUrlCrossRefPubMedWeb of Science
    20. ↵
      1. Morikawa Y,
      2. Nakagawa H,
      3. Miura K, et al
      . Effect of shift work on body mass index and metabolic parameters. Scand J Work Environ Health 2007;33:45–50.
      OpenUrlCrossRefPubMedWeb of Science
    21. ↵
      1. Pan A,
      2. Schernhammer ES,
      3. Sun Q, et al
      . Rotating night shift work and risk of type 2 diabetes: two Prospective Cohort Studies in Women. PLoS Med 2011;8:e1001141.
      OpenUrlCrossRefPubMed
    22. ↵
      1. Smith P,
      2. Fritschi L,
      3. Reid A, et al
      . The relationship between shift work and body mass index among Canadian nurses. Appl Nurs Res 2013;26:24–31.
      OpenUrlCrossRefPubMed
    23. ↵
      1. Suwazono Y,
      2. Dochi M,
      3. Sakata K, et al
      . A Longitudinal Study on the effect of shift work on weight gain in male Japanese workers. Obesity 2008;16:1887–93.
      OpenUrlCrossRefPubMedWeb of Science
    24. ↵
      1. van Amelsvoort LGPM,
      2. Schouten EG,
      3. Kok FJ
      . Duration of shiftwork related to body mass index and waist to hip ratio. Int J Obes Relat Metab Disorders 1999;23:973.
      OpenUrlCrossRefPubMedWeb of Science
    25. ↵
      1. Morikawa Y,
      2. Miura K,
      3. Sasaki S, et al
      . Evaluation of the effects of shift work on nutrient intake: a Cross-sectional Study. J Occup Health 2008;50:270–8.
      OpenUrlCrossRefPubMedWeb of Science
    26. ↵
      1. Bøggild H,
      2. Knutsson A
      . Shift work, risk factors and cardiovascular disease. Scand J Work Environ Health 1999;25:85–99.
      OpenUrlCrossRefPubMedWeb of Science
    27. ↵
      1. Puttonen S,
      2. Härmä M,
      3. Hublin C
      . Shift work and cardiovascular disease—pathways from circadian stress to morbidity. Scand J Work Environ Health 2010;36:96–108.
      OpenUrlCrossRefPubMedWeb of Science
    28. ↵
      1. van Amelsvoort LGPM,
      2. Jansen NWH,
      3. Kant I
      . Smoking among shift workers: more than a confounding factor. Chronobiol Int 2006;23:1105–13.
      OpenUrlCrossRefPubMedWeb of Science
    29. ↵
      1. Schernhammer ES,
      2. Feskanich D,
      3. Liang G, et al
      . Rotating night-shift work and lung cancer risk among female nurses in the United States. Am J Epidemiol 2013;178:1434–41.
      OpenUrlAbstract/FREE Full Text
    30. ↵
      1. Pilcher JJ,
      2. Lambert BJ,
      3. Huffcutt AL
      . Differential effects of permanent and rotating shifts on self-report sleep length: a meta-analytic review. Sleep 2000;15:155–63.
      OpenUrl
    31. ↵
      1. Sallinen M,
      2. Kecklund G
      . Shift work, sleep, and sleepiness—differences between shift schedules and systems. Scand J Work Environ Health 2010;36:121–33.
      OpenUrlCrossRefPubMedWeb of Science
    32. ↵
      1. Härmä M
      . Workhours in relation to work stress, recovery and health. Scand J Work Environ Health 2006;32:502–14.
      OpenUrlCrossRefPubMedWeb of Science
    33. ↵
      1. Wang X-S,
      2. Armstrong MEG,
      3. Cairns BJ, et al
      . Shift work and chronic disease: the epidemiological evidence. Occup Med 2011;61:78–89.
      OpenUrlAbstract/FREE Full Text
    34. ↵
      1. Schernhammer ES,
      2. Laden F,
      3. Speizer FE, et al
      . Night-shift work and risk of colorectal cancer in the Nurses’ Health Study. J Natl Cancer Inst 2003;95:825–8.
      OpenUrlAbstract/FREE Full Text
    35. ↵
      1. Viswanathan AN,
      2. Hankinson SE,
      3. Schernhammer ES
      . Night shift work and the risk of endometrial cancer. Cancer Res 2007;67:10618–22.
      OpenUrlAbstract/FREE Full Text
    36. ↵
      1. Hansen J,
      2. Lassen CF
      . Nested case–control study of night shift work and breast cancer risk among women in the Danish military. Occup Environ Med 2012;69:551–6.
      OpenUrlAbstract/FREE Full Text
    37. ↵
      1. Ramin C,
      2. Devore EE,
      3. Pierre-Paul J, et al
      . Chronotype and breast cancer risk in a cohort of US nurses. Chronobiol Int 2013;30:1181–6.
      OpenUrlCrossRefPubMed
    38. ↵
      1. Ijaz S,
      2. Verbeek JP,
      3. Seidler AP, et al
      . Night-shift work and breast cancer—a systematic review and meta-analysis. Scand J Work Environ Health 2013;39:431–47.
      OpenUrlCrossRefPubMedWeb of Science
    39. ↵
      1. Stevens RGP,
      2. Hansen JMP,
      3. Schernhammer ESMDD, et al
      . Response to Ijaz S, et al. “Night-shift work and breast cancer—a systematic review and meta-analysis”/Response to Letter to the Editor, re: Ijaz S, et al. “Night-shift work and breast cancer—a systematic review and meta-analysis”. Scand J Work Environ Health 2013;39:631–4.
      OpenUrlCrossRefPubMedWeb of Science

    Footnotes

    • Contributors CR, EED, WW, JP-P, LRW and ESS designed the research. CR and EED conducted statistical analyses. CR, EED, LRW and ESS drafted the manuscript. WW and JPP provided critical input in the writing of the manuscript. All authors have read and approved the final version of this manuscript.

    • Funding This manuscript was supported by UM1CA176726 and 5R01OH009803 (PI: Schernhammer E) from CDC—NIOSH. LRW was supported in part by training grant NIH R25 CA098566.

    • Competing interests None.

    • Ethics approval Institutional Review Board (IRB) of Brigham and Women's Hospital (Boston, Massachusetts, USA).

    • Provenance and peer review Not commissioned; externally peer reviewed.