Summary
Cardiovascular diseases remain a major public health problem. The involvement of several occupational factors has recently been discussed, notably the organization of work schedules, e.g. shift work. To analyse the progress of knowledge on the relationship between cardiovascular risk factors and shift work. A review of English-language literature dealing with the link between cardiovascular factors and shift workers (published during 2000–2010) was conducted. Studies published in the past 10 years tend to document an impact of shift work on blood pressure, lipid profile (triglyceride levels), metabolic syndrome and, possibly, body mass index. However, the consequences on glucose metabolism are unclear. These results are not yet firmly established, but are supported by strong hypotheses. Some advice could reasonably be proposed to guide the clinical practitioner.
Résumé
Les maladies cardiovasculaires demeurent un problème majeur de santé publique. Ainsi récemment, l’implication des plusieurs facteurs professionnels a été évoqué et notamment l’organisation des horaires de travail (travail posté). Analyser l’avancée des connaissances sur les relations entre le travail posté et les facteurs de risque cardiovasculaire. Une revue de la littérature anglaise traitant du lien les facteurs de risque cardiovasculaire et le travail posté a été menée durant la période de 2000–2010. Les études les plus récentes tendent à documenter des effets du travail posté sur la pression sanguine artérielle, sur le profil lipidique (notamment sur les taux de triglycérides), sur le syndrome métabolique et probablement sur l’indice de masse corporelle. Les conséquences sur le métabolisme glucidique restent à préciser. Les résultats ne sont pas encore strictement établit, mais plusieurs hypothèses physiopathologiques les supportent et des conseils pourraient être raisonnablement proposés aux praticiens.
Background
Among the various causes of mortality, deaths attributable to CVDs are the most widespread worldwide, and forecasts suggest they will still rank first in 2030 (World Health Statistics, 2008). The factors implicated in CVD have inspired the development of various prevention strategies over the past 40 years. Although some of these factors are well proven, others remain uncertain. Among those currently recognized, non-modifiable risk factors (e.g. age and gender) are set apart from modifiable ones (e.g. high BP, dyslipidaemia and diabetes). However, despite improvements in therapeutic management, people remain at risk of CVD. This poses the question as to whether undiscovered or unrecognized factors could have a role to play in better overall risk management.
Some occupational factors are now suspected to be related to CVD. Among them, the management of work schedules (shift work) is becoming an increasingly important one. Directive 93/104/EC broadly defines shift work as ‘any method of organizing work in shifts whereby workers succeed each other at the same work stations according to a certain pattern, including a rotating pattern, and which may be continuous or discontinuous, entailing the need for workers to work at different times over a given period of days or weeks’. Typically, shift work can be performed in two shifts with a break in the late afternoon and on weekends (2 × 8), in three shifts with a break on weekends (3 × 8) or in four or five shifts to ensure working round the clock. This mode of operation may vary depending on the rotation cycle (number of days between two identical sequences), the direction of rotation (clockwise or counterclockwise) and the stability of the time slots planned (permanent night work). Shift work is therefore organized in a wide range of possible schedules. This way of managing work schedules contrasts with a more standard pattern (‘daytime work pattern’).
In the working world, shift work is a very common mode of operation to serve obvious economic and social goals. In the US, sources from the Bureau of Labor Statistics in 2004 stated that 15% of US employees did shift work. According to the fourth report on working conditions in Europe, issued in 2005, shift work represented an important mode of operation to address the economic circumstances of modern society (15–20%).
A link between shift work and cardiovascular disease has been hypothesized and highlighted increasingly in recent years, but cannot be firmly asserted. A meta-analysis of 17 studies, published in 1999, noted a 40% higher relative risk of CVD among shift workers compared to day workers, for both men and women . A recently published overview of the literature focusing on ischaemic heart disease and based on 16 studies (1972–2008) did not conclude with certainty that shift work has an impact . Broadly, similar results were seen recently from a 22-year period of follow-up of a Finnish cohort which analysed mortality due to coronary heart disease in both genders .
The difficulties in analysing to the consequences of shift work on cardiovascular risks remain for several reasons: heterogeneous definitions of shift work; heterogeneity in the confounding factors included in studies; and the pathological and physiological mechanisms considered. We wanted to contribute to this research by proposing a survey of the literature that deals with the impact of shift work on CVD risk factors during 2000–2010. Pertaining to the group of factors deemed modifiable, the work schedules could, if firmly implicated, be a major target for public and individual efforts to prevent CVD.
Methods
Literature search
Searches were conducted using the following electronic bibliographies and repositories, to analyse the link between shift work and cardiovascular risk factors: PubMed, Cochrane library and Embase. The following key words were included as MeSH terms: shift work, night work shift workers, and night workers. Each of these was combined individually with the following: (1) cardiovascular risk factors; (2) hypertension; (3) BMI (obesity, overweight); (4) lipids (triglyceride, cholesterol, HDL-C); (5) diabetes; and (6) metabolic syndrome).
Selected articles
From a review published in 1999 , we have updated the knowledge of this field by including: (1) original articles published between January 2000 to December 2010; (2) written in English and published in an international peer-reviewed journal; and (3) concerning adult subjects. Review articles and those with no empirical results informing the link between shift work and cardiovascular risk factors were excluded.
Classification of articles
The publications were classified by cardiovascular risk factors. We have also described systematically the type of study (cross-sectional, longitudinal, prospective or retrospective), the number, age and gender of subjects examined, and their distribution between day workers and shift workers, and the type of shift work.
Our purpose was to classify how the different types of work schedules studied in these articles are associated with different cardiovascular risk factors, taking care of confounding parameters. Summary tables are available at the end of this overview, organized by study design in each category of cardiovascular factor and by year of publication.
Results
Included articles
Initially, 215 articles were recovered from the search of the three databases, but after reviewing titles and abstracts and removing duplicates, 74 articles remained and have been studied in this review ( Fig. 1 ).
Shift work and hypertension
The role of the working environment on the pathogenesis of hypertension is not clear, but are there any grounds for so-called ‘idiopathic’ hypertension ( Table 1 )? The review of the literature published in 1999 by Bøggild and Knutsson showed no link between shift work and hypertension in most of the studies included. However, research carried out over the past decade, including more longitudinal studies and the use of the new hypertension definition (SBP > 140 mmHg and/or DBP > 90 mmHg and/or taking antihypertensive medication) tends to show an impact of shift work on threshold values of arterial BP.
| Reference | Study type; years; recruitment | n gender (age [mean or range]) | SW schedule | Adjustment parameters | BP parameters | Main results SW vs DW (OR [95% CI]) unless otherwise specified |
|---|---|---|---|---|---|---|
| Longitudinal studies | ||||||
| Sakata et al., 2003 | Prospective; 1991–2001; workers in Japanese steel company | 3022 ♂ DW 2316 ♂ SW | 3 × 8; irregular shift work: 5 days; 2 rest days; 5 evenings; 1 rest day; 5 nights; 2 rest days; clockwise rotation | Age, BMI, drinking and smoking habits, physical activity, TC, creatinine, γGTP, HbA1c, uric acid | Hypertension (SBP ≥ 140 mmHg and/or DBP ≥ 90 mmHg and/or treatment) | Increased risk of onset of hypertension (1.10 [1.01–1.20]) |
| van Amelsvoort et al., 2004 | 1 y; nurses and incinerator plant workers | 150 ♂ ♀ DW 227 ♂ ♀ SW | Counter-clockwise rotation (night, afternoon, morning); clockwise rotation (morning, afternoon, night) | Educational level, gender, age, JSI variables, variables under study at baseline | BP | No significant change in BP after 1 y in SW |
| Oishi et al., 2005 | Prospective; 1991–2001; workers in Japanese steel company | 1560 ♂ DW (40.3 y) 1381 ♂ SW (40.5 y) All had mild hypertension | 3 × 8 irregular shift work: 5 days; 2 rest days; 5 evenings; 1 rest day; 5 nights, 2 rest days; clockwise rotation | Age, BMI, drinking and smoking habits, physical activity, TC, creatinine, γGTP, HbA1c, uric acid | Progression from mild hypertension (SBP 140–159 and/or DBP 90–99 mmHg); to severe hypertension (SBP ≥ 160 and/or DBP ≥ 100 mmHg) | SW significantly associated with progression from mild to severe hypertension (1.23 [1.05–1.44]) and severe diastolic hypertension (1.28 [1.07–1.52]) |
| Kivimäki et al., 2006 | Prospective; 2000–2004; nurses in 21 Finnish hospitals | 1999 ♀ DW (45.3 y) 5038 ♀ SW (41.1 y) | 3 × 8; 2 × 8; permanent nights | None | BP | Baseline BP NS; high BP in SW did not predict leaving organisation to get DW |
| Morikawa et al., 2007 | Prospective; 1993–2003; Japanese factory workers | 1993/2003: 712 ♂ DW/DW (36.4 y) 173 ♂ DW/SW (36.0 y) 210 ♂ SW/DW (36.2 y) 434 ♂ SW/SW (33.5 y) | 2 × 8, 3 × 8; counter-clockwise rotation; non-continuous system (5 days, 5 nights, 5 evenings with 2 weekend rest intercalated); continuous system (3–4 days, 3–4 nights, 3–4 evenings and 1 rest day intercalated) | Age, BMI, smoking, alcohol, physical activity | Increase in SBP and DBP over 10 y | NS |
| Virkkunen et al., 2007 | Prospective; 1982–1999; industrial workers; Helsinki heart study, Finland | 404 ♂ DW (52.5 y) 27 ♂ SW (52.6 y) | 2 × 8; 3 × 8; irregular work, night work (without noise or physical workload) | None | BP during 8-year follow-up | SBP and DBP NS |
| Nabe-Nielsen et al., 2008 | Prospective; 2004–2005; Danish social/healthcare assistants | 1483 ♂ ♀ DW (35.1 y) 482 ♂ ♀ SW1 (36.3 y) 124 ♂ ♀ SW2 (36.6 y) 474 ♂ ♀ SW3 (33.6 y) 307 ♂ ♀ SW4 (33.6 y) | SW1: evenings; SW2: nights; SW3: 2 × 8 without night work; SW4: 2 × 8 and 3 × 8 with night work | Age, gender, education, cohabitation, general self-efficacy, years of school, former experience in the eldercare sector | Hypertension history | SW1 (1.04 [0.66–1.63]); SW2 (0.70 [0.27–1.82]); SW3 (1.26 [0.79–2.02]); SW4 (1.18 [0.68–2.05]); all vs DW |
| Suwazono et al., 2008 | Prospective; 1991–2005; workers in Japanese steel company | 3963 ♂ DW (35.8 y) 2748 ♂ SW (36.7 y) | 3 × 8 irregular shift work: 5 days; 2 rest days; 5 evenings; 1 rest day; 5 nights; 2 rest days; clockwise rotation | Age, BMI, tobacco, alcohol, physical activity, TC, creatinine, γGTP, AST, HbA1c, uric acid | Increased SBP or DBP by ≥ 10%, ≥ 15%, ≥ 20%, ≥ 25%, ≥ 30% | ≥ 10%: SBP (1.15 [1.07–1.23]); DBP (1.19 [1.11–1.28]); ≥ 15%: SBP (1.21 [1.12–1.31)]; DBP (1.22 [1.13–1.33]); ≥ 20%: SBP (1.15 [1.04–1.28]); DBP (1.24 [1.13–1.37]); ≥ 25%: SBP (1.20 [1.06–1.37]); DBP (1.16 [1.03–1.30]); ≥ 30%: SBP (1.23 [1.03–1.47]); DBP (1.04 [0.89–1.22]) |
| Puttonen et al., 2009 | Prospective; 1980–2001; Young Finns Study | 668 ♀ DW 515 ♂ DW 831 ♀ SW 712 ♂ SW All 24–39 y | 2 × 8; 3 × 8; regular evening or night work | Age | BP | No association with SBP or DBP |
| De Bacquer et al., 2009 | Cohort study; 1995–2003; BELSTRESS, Belgium | 1220 ♂ DW (44.7 y) 309 ♂ SW (43.1 y) | 2 or 3 rotating shifts | Age, WC, DBP, HDL-C | SBP/DBP ≥ 130/85 mmHg and/or treatment | (1.31 [1.04–1.66]) |
| Lin et al., 2009 | Retrospective cohort; 2002–2007; employees of electronic manufacturing company, Taiwan | 125 ♀ DW (31.1 y) 160 ♀ past SW (34.7 y) 102 ♀ current SW (31.9 y) | 2 × 8; 6 days, 3 rest days, 6 nights, 3 rest days; 12-h shifts | Smoking, age, insulin status, metabolic syndrome, job, physical activity, snack before sleeping | BP ≥ 130/85 mmHg or treatment at 5-y FU | Elevated BP: DW: 15.2%; past SW 34.4%; current SW: 37.3% (both P < 0.05 vs DW) |
| Pietrouisty et al., 2010 | Prospective cohort, 2003–2007; nurses in 3 hospitals in Italy | 336 ♀ ♂ DW (37.9 y) 402 ♀ ♂ SW (38.9 y) | Night or rotating shifts with ≥ 4 nights/month | None | Incidence of BP ≥ 130/85 mmHg or treatment | DW: 11.9%; SW: 11.2%; NS |
| Hublin et al., 2010 | Cohort study; 1975–1981–2003; Finnish nationwide official register | 7698/7838 ♂/♀ DW 108/224 ♂/♀ nights 509/701 ♂/♀ SW-DW 562/610 ♂/♀ DW-SW 962/930 ♂/♀ SW | From questionnaire: mainly DW; mainly SW; mainly nights | Age, marital status, social class, education, smoking, alcohol, hypertension, BMI, physical activity, life satisfaction, sleep length, use of hypnotics and/or tranquilizers, physical load of work, working place | Hypertension | Hypertension incidence not significantly different for both genders |
| Thomas & Power 2010 | ‘1958 British birth cohort’ | 2710/1665 ♂/♀ DW 662/368 ♂/♀ nights 776/300 ♂/♀ mornings 2226/1357 ♂/♀evenings 1358/864 ♂/♀ weekends All 45 y | SW if ≥ 1/week outside 07:00–18:00; 4 types of SW: Summing the Number of shift work types (0–4) nights, mornings, evenings, weekends | Social class, total h per week, employee, physical activity, diet, smoking, alcohol, diabetes treatment | BP | NS for SBP and DBP for ♂ working several types of shift NS for SBP for ♀ in several types; lower DBP in ♀ type 4 |
| Cross-sectional studies | ||||||
| Ohira et al., 2000 | Japan | 26 ♂ DW (32 y) 27 ♂ SW (31 y) | 3 × 8; 2 days, 1 afternoon, 2 nights, 3 days off | BMI, alcohol, anger expression, physical activity | 24-h ABP (mean BP during awake, asleep, non-work awake, work) | SBP significantly higher during awake and work periods; but no difference in DBP |
| Munakata et al., 2001 | Nurses, Japan | 18 ♀ SW (29 y) | 2 days off, 2 nights, 1 day off, 2–3 days, 2 evenings | None | 24-h ABP | SBP lower during night shift than day shift ( P = 0.01); no difference in DBP |
| Ha et al., 2001 | Workers in manufacturing firm | 134 ♂ SW (29 y) | 3 × 8; counter-clockwise rotation; 3 days in the same shift, 1 day off | Age, JSI, occupational history, past medical history, family history, smoking, alcohol consumption, interaction age × duration of SW | BP (measured at each of the 3 shifts) and SW duration | SBP and DBP associated with duration of SW ( P < 0.05); BP fell morning to afternoon to night ( P = 0.031) |
| Karlsson et al., 2001 | VIP study, Sweden | 9857 ♀ DW 9719 ♂ DW 4632 ♀ SW 3277 ♂ SW All aged 30, 40, 50 or 60 y | Any shift or weekend work | None | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | Prevalence of hypertension: NS except for men included in group 40 years old 15.5% sw/12.3% dw, P < 0.01 |
| Kitamura et al., 2002 | Factory employees with untreated hypertension | 12 ♂ SW (53.6 y) | 4 days, 2 days off, 4 nights | None | 24-h ABP | Circadian BP pattern changed from dipper to non-dipper on 1st day of night shift; returns to dipper within a few days |
| Nagaya et al., 2002 | Gifu Prefectural Center; Health-check programme, Japan | 2824 ♂ DW (47.1 y) 826 ♂ SW (45.6 y) | Any mandatory night work (permanent night workers excluded) | BMI, job, drinking, smoking, exercise Stratified by age classes | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | 30–39 y (NS); 40–49 y (1.62 [1.17–2.24]); 50–59 y (NS) |
| Karlsson et al., 2003 | WOLF study, Sweden | 665 ♂ DW (44.3 y) 659 ♂ SW (44.2 y) | 3 rotating shifts | None | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | DW: 21.1%; SW: 16.9%; P = 0.052 |
| Murata et al., 2005 | Copper-smelting plant, Japan | 87 ♂ DW 153 ♂ SW | 4-shift forward rotation system | BMI, HDL-C, TG, haemoglobin, haematocrit, smoking, drinking, age, work duration | SBP ≥ 160 mmHg or DBP ≥ 95 mmHg | (1.06 [0.50–2.22]) |
| Ha & Park 2005 | Nurses and blue-collar workers | 134 ♂ SW (29.1 y) 226 ♀ SW (28.5 y) | 3 × 8; irregular rotating shifts including mornings, evenings, nights | Smoking, drinking, physical activity, JSI | BP | ♂ ≥ 30 y: SW duration associated with SBP ( P < 0.05); ♀ < 30 y: SW duration inversely associated with DBP ( P < 0.05) |
| Fialho et al., 2006 | Medical residents in ER | 56 ♂ ♀ DW and SW (25.4 y) | DW: common working day; SW: 24-h shift | None | Mean 24-h ABP | Higher SBP (117 vs 113 mmHg; P < 0.05) and DBP (73 vs 69 mmHg; P < 0.05) |
| Ghiasvand et al., 2006 | Repairs workshop | 266 ♂ DW (38.6 y) 158 ♂ SW (46.4 y) | Not normal daylight hours | None | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | DW: 11.7%; SW: 17.1%; NS |
| Sookoian et al., 2007 | 2005; factory workers in Buenos Aires | 877 ♂ DW (34 y) 474 ♂ SW (36 y) | Clockwise rotation; 2 × 8; 4 days, 4 nights, 3 rest days, 2 days, 3 rest days, 4 days | None | Mean BP | Mean SBP: NS; mean DBP: DW: 78 mmHg; SW: 76 mmHg ( P = 0.04) |
| Copertaro et al., 2008 | 2005; hospital staff, Italy | 77 ♂ ♀ healthcare DW (48.6 y) 70 ♂ ♀ healthcare SW (47.3 y) Subgroup: 41 ♀ DW nurses (45.8 y) 32 ♀ SW nurses (43.5 y) | SW healthcare 1–6 nights/month; nurses: 3 × 8; 1 day, 1 evening, 1 night, 2 rest days | None | BP | Overall: SBP and DBP NS. Nurses: SBP elevated ( P < 0.05); DBP NS |
| Haupt et al., 2008 | Study of Health In Pomerania (SHIP) | 760/1052 ♂/♀ DW (61.5 y) 506/192 ♂/♀ SW (62.3 y) | Ever worked shifts/nights | None | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | DW: 52.8%; SW: 57.0%; NS |
| Nazri et al., 2008 | Factory workers in Kota Bharu, Kelantan | 72 ♂ DW (31.6 y) 76 ♂ SW (32.3 y) | 2 × 8; alternating day- and night-time; 2 days, 2 nights, 3 rest days | BMI, nature of job | SBP ≥ 140 mmHg or DBP ≥ 90 mmHg or treatment | DW: 4.2%; SW: 22.4%; (9.1 [1.4–56.7]); P < 0.001 |
| Lo et al., 2008 | Nurses from municipal hospital, Taiwan | 8 ♀ DW (27 y) 16 ♀ SW (27 y) | SW: day, evening, night | None | 24-h ABP | SBP and DBP increased significantly during sleeping period after night/evening shift; SW associated with dipper/non-dipper status; BP not return to baseline after night shift ( P < 0.05) |
| Viitasolo et al., 2008 | Airline company, Finland | 84 ♂ SW: 40 changing to rapid forward rotation (47 y); 22 changing to flexible system (37 y); 22 staying on old system (44 y) | Old system: 3 evenings, 3 mornings, 3 nights, 2 rest days between each shift Rapid forward rotation:: morning, evening, night, 2 rest days; flexible system: direction and rotation comparable with? old system with 3 days off; possible change with rotas of the 3rd or 4th week | None | Mean SBP | SBP +6 mm Hg for flexible shift system, +2.5 mm Hg for rapid forward rotation; no change for old system; DBP NS for the 3 groups |
| Esquirol et al., 2009 | Industrial plant, France | 98 ♂ DW (48.8 y) 100 ♂ SW (46.5 y) | 3 × 8; clockwise rotation; 1–2 mornings, 1–2 afternoons, 1–2 nights, 3–4 rest days | None | BP | SBP or DBP; NS |
| Sfreddo et al., 2010 | Hospital nursing personnel, Brazil | 276/35 ♀/♂ DW (33.1 y) 159/23 ♀/♂ night SW (36.4 y) | Night work | Age, gender, skin colour, marital status, y at school, tobacco, alcohol, sleeping hours, BMI | Hypertension: ≥ 140/90 mmHg or treatment; pre-hypertension: ≥ 120–139/ ≥ 80–89 mmHg | Hypertension or pre-hypertension (RR 1.0 [0.8–1.3]) |
| Chen et al., 2010 | Manufacturing factory, Taiwan | 401 ♀ office workers (33.5 y) 220 ♀ DW (33.6 y) 656 ♀ day SW (34.9 y) 561 ♀ night SW (32.7 y) | Day SW: 07:00–19:00; night SW: 19:00–07:00; 2 days work, 2 days off | Age, smoking, drinking, educational level, duration of employment | SBP ≥ 130 mmHg or DBP ≥ 85 mmHg or treatment | DW (1.6 [0.8–3.0]); day SW (1.1 [0.6–2.1]); night SW (2.3 [1.2–4.4]); all vs office |
Among workers in a Japanese steel factory, a significantly higher risk of developing hypertension (odds ratio [OR] 1.10, 95% confidence interval [CI] 1.01–1.20) has been found for shift workers compared to day workers ; and a higher risk of progression from mild to severe hypertension (OR 1.23, 95% CI 1.05–1.44) . Moreover, some studies have implicated shift work as a possible cause for raised systolic or diastolic pressure , but this has not been confirmed by other longitudinal studies, based on baseline results or during the monitoring periods . The disparities in these findings may relate to the way other co-variates or confounders, such as career development, have been handled in the analyses. Similarly, the findings of cross-sectional studies have been mixed: some report a link between shift work and BP , while others refute it .
Some authors have analysed the effects of shift work on hypertension, while taking into account age and duration of exposure. In one study, a higher risk was found among shift workers aged 40–49 years, but not in those aged 30–39 and 50–59 years, suggesting a ‘healthy workers effect’ . A significant increase in arterial SBP was also noted in men aged > 30 years exposed to shift work over 1–10 years, but these results have not been confirmed for women involved in shift work . A first year of exposure to shift work seems to have no effect on arterial BP, in both genders combined .
Partial adjustment of the circadian rhythms of arterial BP could be an explanation, with insufficient decreases in arterial BP during a night awake and relative increases during a sleep period after night or evening work compared to sleep during the night for day workers. An evolving non-dipper status, complete or partially reversible after a day off, is recorded . The measurement of heart rate variability for evaluating the cardiac autonomic function has been used in some studies. Modification of sympathetic system responses is probably involved under the influence of level of physical activity during the sleep-wakefulness cycle . Twelve-hour night-shift work has been found to result in an elevated BP and heart rate, a decrease in heart rate variability, and a delay in BP recovery . In the shift system, the individual flexibility of work hours seems to have a less deleterious effect on BP than rapid forward rotations . Considering the sum of the number of shift work types worked at least once per week (night, evening, morning, weekend), the results only show an inverse relationship for DBP in women .
In summary, studies published in the past 10 years have highlighted a potential increase in the risk of developing hypertension for shift workers; and the duration of exposure could influence this link.
Shift work and lipid disturbances
The survey published by Bøggild and Knutsson in 1999 identified 27 studies correlating shift work and lipid factors ( Table 2 ).
| Reference | Study type; years; recruitment | n gender (age [mean or range]) | SW schedule | Adjustment parameters | Lipid parameters | Main results SW vs DW (OR [95% CI]) unless otherwise specified |
|---|---|---|---|---|---|---|
| Longitudinal studies | ||||||
| Sakata et al., 2003 | Prospective; 1991–2001; workers in Japanese steel company | 3022 ♂ DW 2316 ♂ SW | 3 × 8; irregular shift work: 5 days, 2 rest days, 5 evenings, 1 rest day, 5 nights, 2 rest days; clockwise rotation | None | Mean TC at baseline | NS |
| van Amelsvoort et al., 2004 | 1 y; nurses and incinerator plant workers | 150 ♂ ♀ DW 227 ♂ ♀ SW | Counter-clockwise rotation (night, afternoon, morning); clockwise rotation (morning, afternoon, night) | Educational level, gender, age, JSI variables, variables under study at baseline | Mean TC, HDL-C, LDL-C at baseline and 1 y | LDL-C:HDL-C lower ( P = 0.004); TC, LDL-C and HDL-C NS |
| Oishi et al., 2005 | Prospective; 1991–2001; workers in Japanese steel company | 1560 ♂ DW (40.3 y) 1381 ♂ SW (40.5 y) | 3 × 8; irregular shift work: 5 days, 2 rest days, 5 evenings, 1 rest day, 5 nights, 2 rest days; clockwise rotation | None | Mean TC at baseline | NS |
| Kivimäki et al., 2006 | Prospective; 2000–2004; nurses in 21 Finnish hospitals | 1999 ♀ DW (45.3 y) 5038 ♀ SW (41.1 y) | 3 × 8; 2 × 8; permanent nights | None | Self-reported high TC | At baseline: NS; high TC was not a predictor of leaving organisation to find day work |
| Morikawa et al., 2007 | Prospective; 1993–2003; Japanese factory workers | 1993/2003: 712 ♂ DW/DW (36.4 y) 173 ♂ DW/SW (36.0 y) 210 ♂ SW/DW (36.2 y) 434 ♂ SW/SW (33.5 y) | 2 × 8; 3 × 8; counter-clockwise rotation; non-continuous system (5 days, 5 nights, 5 evenings with 2 weekend rest intercalated); continuous system (3–4 days, 3–4 nights, 3–4 evenings and 1 rest day intercalated) | Age, BMI, smoking, alcohol, physical activity | Mean TC over 10 y | NS |
| Biggi et al., 2008 | Prospective; 1976–2007; Milan | 157 ♂ DW (42.3 y) 331 ♂ SW (47.0 y) | Permanent nights | Study period, job, age, lifestyle variables | TG > 150 mg/dL; TC > 200 mg/dL | High TG and high TC more common in night workers |
| Dochi et al., 2008 | Prospective; 1991–2005; steel company, Japan | 3263 ♂ DW 2247 ♂ SW | Clockwise rotation; 3 × 8; 5 days, 2 rest days, 5 evenings, 1 rest day, 5 nights, 2 rest days | Age, tobacco, alcohol, physical activity, BMI, laboratory data | TC ≥ 5.7 mmol/L | (1.1 [1.00–1.21]) |
| Lin et al., 2009 | Retrospective cohort; 2002–2007; electronic manufacturing company, Taiwan | 125 ♀ DW (31.1 y) 160 ♀ past SW (34.7 y) 102 ♀ current SW (31.9 y) | 2 × 8; 6 days, 3 rest days, 6 nights, 3 rest days; 12-h shifts | Smoking, age, insulin status, metabolic syndrome, job, physical activity, snack before sleeping | HDL-C < 40 (♂) or < 50 (♀) mg/dL; TG ≥ 150 mg/dL at 5-y FU | High TG: DW 9.6%; past SW 8.1%; current SW 11.8% (NS); low HDL-C: DW 11.2%; past SW 11.3%; current SW 14.7% (NS) |
| Dochi et al., 2009 | Prospective; 1991–2005; steel company, Japan | 4079 ♂ DW (36.2 y) 2807 ♂ SW (36.9 y) | Clockwise rotation; 3 × 8; 5 days, 2 rest days, 5 evenings, 1 rest day, 5 nights, 2 rest days | Age, tobacco, alcohol, physical activity, BMI, laboratory data | TC raised from baseline | ≥ 20% (1.16 [1.07–1.26]); ≥ 25% (1.16 [1.05–1.28]); ≥ 30% (1.11 [0.98–1.25]); ≥ 40% (1.30 [1.07–1.58]) |
| Puttonen et al., 2009 | Prospective; 1980–2001; Young Finns Study | 668 ♀ DW 515 ♂ DW 831 ♀ SW 712 ♂ SW All 24–39 y | 2 × 8; 3 × 8; regular evening or night work | Age | Mean LDL-C, HDL-C, TG at baseline | LDL-C and HDL-C both NS; TG (♀: 1.26 vs 1.14; P = 0.042; ♂: 1.69 vs 1.48; P = 0.017) |
| De Bacquer et al., 2009 | Cohort study; 1995–2003; BELSTRESS, Belgium | 1220 ♂ DW (44.7 y) 309 ♂ SW (43.1 y) | 2 or 3 rotating shifts | Age, WC, DBP, HDL-C | TG ≥ 220 mg/dL; HDL-C < 40 mg/dL | Low HDL-C (1.42 [1.02–1.99]); high TG (1.53 [1.22–1.92]) |
| Suwazono et al., 2010 | Prospective; 1991–2005; steel company, Japan | 4079 ♂ DW (36.2 y) 2807 ♂ SW (36.9 y) | Clockwise rotation; 3 × 8; 5 days, 2 rest days, 5 evenings, 1 rest day, 5 nights, 2 rest days | Age, tobacco, alcohol, physical activity, BMI, creatinine, GGT, AST | TC raised from baseline; BMD and BMDL were calculated | ≥ 20% (1.03 [1.02–1.04]); ≥ 45% (1.13 [1.09–1.09]); threshold y for SW at 5% risk of increased TC: age 40–49 y: ≥ 21 y; age ≥ 50 y: 22.7 y |
| Cross-sectional studies | ||||||
| Karlsson et al., 2001 | VIP study, Sweden | 9857 ♀ DW 9719 ♂ DW 4632 ♀ SW 3277 ♂ SW All aged 30, 40, 50 or 60 y | Any shift or weekend work | Age, socioeconomic group | TG > 1.7 mmol/L; HDL-C < 1.0 mmol/L (♀) or < 0.9 mmol/L (♂); mean TC | Low HDL-C: ♀ (1.26 [1.03–1.53]), ♂ (1.15 [0.96–1.38]); high TG: ♀ (1.13 [1.02–1.25]), ♂ (1.12 [1.01–1.24]); higher TC in SW for ♀ at age 40, 50, 60 y ( P < 0.05) and ♂ at 40 y ( P < 0.05) |
| Nagaya et al., 2002 | Gifu Prefectural Center; Health-check programme, Japan | 2824 ♂ DW (47.1 y) 826 ♂ SW (45.6 y) | Any mandatory night work (permanent night workers excluded) | BMI, job, drinking, smoking, exercise] | TG ≥ 1.7 mmol/L; HDL-C < 1.04 mmol/L Stratified by age classes | 40–49 y: high TG (1.65 [1.26–2.16]); 50–59 y: low HDL-C (0.59 [0.36–0.93]); all other associations NS |
| Karlsson et al., 2003 | WOLF study, Sweden | 665 ♂ DW (44.3 y) 659 ♂ SW (44.2 y) | 3 rotating shifts | Age, socioeconomic group, current smoking, physical activity, low social support, JSI | TC > 6.4 mmol/L; HDL-C < 0.9 mmol/L; TG ≥ 1.7 mmol/L; | TC > 6.4 mmol/L: DW 28.1%; SW 19% ( P < 0.0001); HDL-C < 0.9 mmol/L: (2.03 [1.18–3.48]); TG ≥ 1.7 mmol/L: (1.40 [1.08–1.83]) |
| Di Lorenzo et al., 2003 | Chemical industry, Italy | 134 ♂ DW (48.9 y) 185 ♂ SW (48.7 y) | 3 × 8; counter-clockwise; 2 nights, 2 afternoons, 2 mornings, 3 rest days | None | Mean TC, HDL-C, TG | All NS |
| Ha & Park 2005 | Nurses and blue-collar workers | 134 ♂ SW (29.1 y) 226 ♀ SW (28.5 y) | 3 × 8; irregular rotating shifts including mornings, evenings, nights | Smoking, drinking, physical activity, JSI, BMI | Mean TC | ♂ ≥ 30 y: TC elevated with SW duration (ß = 9.72; P < 0.05); ♀ ≥ 30 y: (ß = –2.82; P < 0.05) |
| Sookoian et al., 2007 | 2005; factory workers in Buenos Aires | 877 ♂ DW (34 y) 474 ♂ SW (36 y) | Clockwise rotation; 2 × 8; 4 days, 4 nights, 3 rest days, 2 days, 3 rest days, 4 days | None | Mean TC, HDL-C, LDL-C, TG | SW had elevated TG ( P = 0.003); TC, HDL-C, LDL-C all NS |
| Lavie & Lavie 2007 | Electric company, Israel | 207 ♂ DW (55.1 y) 154 ♂ SW (56.3 y) | 6 mornings, 3 afternoons, 3 nights; 1–2 rest days intercalated | Age, sleep disturbed | Mean TC, HDL-C, LDL-C, TG | All NS |
| Haupt et al., 2008 | Study of Health in Pomerania (SHIP) | 760/1052 ♂/♀ DW (61.5 y) 506/192 ♂/♀ SW (62.3 y) | Ever worked shifts/nights | None | Mean TC, LDL-C:HDL-C | LDL-C:HDL-C higher in SW (3.0 vs 2.9; P < 0.05); TC NS |
| Copertaro et al., 2008 | 2005; hospital staff, Italy | 77 ♂ ♀ healthcare DW (48.6 y) 70 ♂ ♀ healthcare SW (47.3 y) Subgroup: 41 ♀ DW nurses (45.8 y) 32 ♀ SW nurses (43.5 y) | SW healthcare 1–6 nights/month; nurses: 3 × 8; 1 day, 1 evening, 1 night, 2 rest days | None | Mean HDL-C or TG | Overall, SW had lower HDL-C ( P < 0.05) and higher TG ( P < 0.01). In nurses, SW had lower HDL-C (NS) and higher TG (NS) |
| Esquirol et al., 2009 | Industrial plant, France | 98 ♂ DW (48.8 y) 100 ♂ SW (46.5 y) | 3 × 8; clockwise rotation; 1–2 mornings, 1–2 afternoons, 1–2 nights, 3–4 rest days | None | Mean TG, TC, LDL-C, HDL-C | In SW: decreased HDL-C ( P < 0.028); elevated TG ( P < 0.039); TC and LDL-C NS |
| Chen et al., 2010 | Manufacturing factory, Taiwan | 401 ♀ office workers (33.5 y) 220 ♀ DW (33.6 y) 656 ♀ day SW (34.9 y) 561 ♀ night SW (32.7 y) | Day SW: 07:00–19:00; night SW: 19:00–07:00; 2 days work, 2 days off | Age, smoking, drinking, educational level, duration of employment | TG ≥ 150 mg/dL or 1.295 mmol/L; HDL < 40 mg/dL or 1.695 mmol/L | High TG (vs office): DW (0.9 [0.5–1.8]); day SW (1.7 [0.9–3.1]); night SW (0.7 [0.4–1.5]); low HDL-C (vs office): DW (1.1 [0.7–1.8]); day SW (0.9 [0.6–1.4]); night SW (0.7 [0.4–1.1]) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
