Disruption of circadian rhythms promotes alcohol use: a systematic review

Abstract

This systematic review investigates the bidirectional relationship between alcohol consumption and disrupted circadian rhythms. The goal of this study was to identify (i) the types of circadian rhythm disruptors (i.e. social jet lag, extreme chronotypes, and night shift work) associated with altered alcohol use and (ii) whether sex differences in the consequences of circadian disruption exist. We conducted a search of PubMed, Embase, and PsycINFO exclusively on human research. We identified 177 articles that met the inclusion criteria. Our analyses revealed that social jet lag and the extreme chronotype referred to as eveningness were consistently associated with increased alcohol consumption. Relationships between night shift work and alcohol consumption were variable; half of articles reported no effect of night shift work on alcohol consumption. Both sexes were included as participants in the majority of the chronotype and social jet lag papers, with no sex difference apparent in alcohol consumption. The night shift research, however, contained fewer studies that included both sexes. Not all forms of circadian disruption are associated with comparable patterns of alcohol use. The most at-risk individuals for increased alcohol consumption are those with social jet lag or those of an eveningness chronotype. Direct testing of the associations in this review should be conducted to evaluate the relationships among circadian disruption, alcohol intake, and sex differences to provide insight into temporal risk factors associated with development of alcohol use disorder.

Short Summary: This systematic review focuses on the effects of disrupted circadian rhythms on alcohol consumption. Our analyses revealed that social jet lag and the extreme chronotype referred to as eveningness were consistently associated with increased alcohol consumption (86%). In contrast, relationships between night shift work and alcohol consumption were variable.

Introduction

Alcohol consumption is increasing in the United States. In 2018, within a 1-month period, 139.8 million Americans consumed alcohol, 67.1 million (nearly 25%) met the criteria for binge drinking, and 16.6 million were characterized as engaging in heavy drinking (Mental Health Services Administration Health 2019). In addition, the economic burden of alcohol misuse was estimated to be 250 billion dollars in the United States and accounted for approximately 5.3% of global deaths annually (World Health Organization, 2019). The personal, economic, and global burdens of alcohol overuse is likely to grow in the coming years as the Coronavirus Disease 2019 (COVID-19) substantially increased the frequency, quantity, and severity of substance use, including alcohol (Schmidt et al. 2021).

Disruption of circadian rhythms is a relatively unexplored potential factor influencing alcohol use. Circadian rhythms are internal cycles synchronized to precisely 24-h in order to achieve alignment with the solar day (Vitaterna et al. 2001). Circadian rhythms in humans underlie the typical pattern of being active during the light of day and sleeping during the dark of night. However, when common disruptive factors such as social jet lag, extreme chronotypes (preferences for early mornings or late nights), and night shift work force a mismatch between the body’s internal clocks and environmental rhythms, disruption of endogenous circadian rhythms can occur with a wide range of consequences for physical and mental health (Lee et al. 2021). Furthermore, the neurobiology of natural, motivated behaviors such as food seeking (Antons et al. 2020) overlaps significantly with the neurobiology of alcohol-seeking behaviors (Volkow and Morales 2015), and so it is conceivable that endogenous circadian fluctuations contribute to alcohol-related drive states, behaviors, and physiological responses (Meyrel et al. 2020; Tamura et al. 2021).

Social jet lag occurs when there is a difference of ≥2 h in sleep midpoint between work and nonwork days. This shift in schedule causes circadian misalignment. Among the cognitive and behavioral changes that are potentially relevant to alcohol use, social jet lag has been shown to heighten impulsivity (Mcgowan et al. 2020).

An individual’s chronotype reflects the time of day that they naturally choose to sleep and when they feel most energetic (Roenneberg et al. 2019), but also reflects the timing of peak affective, cognitive, and physical performance (Minz and Pati 2021). As reported in twin studies, ~50% of variance in morningness–eveningness is attributable to genetics (Barclay et al. 2014). Males are more likely to have an evening phenotype than females, but this sex difference becomes less pronounced as men and women age (Fabbian et al. 2016). Similar to social jet lag, evening chronotype is associated with delay discounting, and specifically a greater preference for small, immediate rewards than large, delayed rewards (Evans and Norbury 2021). Furthermore, delay discounting in evening chronotype phenotype is also associated with risky alcohol consumption (Acheson 2020).

Night shift work also is associated with numerous physical and mental health disorders (Costa 1996), including reports of higher intensity and riskier drinking patterns when individuals are working night shift compared to day shift (Cheng et al. 2021). The effects of night shift on impulsivity and other cognitive processes that may contribute to alcohol use have not been extensively studied. However, when the various causes of circadian rhythm disruption, including social jet lag, extreme chronotypes, and working the night shift, are considered together, there is evidence to suggest the presence of cognitive changes that have previously been associated with increased alcohol use.

The intensity of alcohol craving in those who regularly drink fluctuates in accordance with a 24-h rhythm, exhibiting low-intensity craving in the morning and the greatest craving in the evening. An individual’s activity and sleep timing patterns influence the 24-h craving rhythm, as observed by a later peak of craving among those of later chronotypes (Hisler et al. 2021). In addition, drinking patterns may be related to daily reward, wanting, and risk-taking rhythms (Hasler et al. 2014; Byrne et al. 2017). The first aim of this systematic review is to investigate which types of circadian rhythm disruption (i.e. social jet lag, extreme chronotypes, and working the night shift) are associated with increased alcohol use in the general population.

Historically, alcohol use has been reported to be more prevalent in men than women (Erol and Karpyak, 2015); however, female participants are also often underrepresented in alcohol use studies, and sex has less frequently been studied as a biological variable (SABV) affecting alcohol use. This underrepresentation of women in studies involving alcohol use and circadian rhythm disruption is troubling, as sex differences in risk and vulnerability to AUD have been established (Inkelis et al. 2020). Therefore, this systematic review also aims to investigate SABV in the effects of the disruption of circadian rhythms that may precipitate alcohol use. Finally, the relationship between age and alcohol consumption will be assessed between circadian rhythm disruptions.

Further research on the influence that alcohol has on circadian rhythms and how the circadian system affects alcohol consumption is needed to clarify the proposed bidirectional relationship between circadian disruption and alcohol use. A recent systematic review highlighted how circadian systems are disrupted by excess alcohol consumption (Meyrel et al. 2020), whereas we test the other side of the bidirectional hypothesis, specifically that disruption of circadian rhythms is associated with increased alcohol consumption. Although delayed sleep phase disorder (DSPD, delay in sleep–wake timing) and chronotypes (preference for a certain sleep–wake pattern) terminology are related, there are differences on neurobiology, comorbidities, and clinical features (Reid et al. 2012). However, for completeness, we have included delayed sleep phase (DSP) and DSPD in the current systematic review.

Materials and Methods

Research approach

This systematic literature review was conducted in January 2022 and is being reported using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al. 2009). The plan was pre-registered with Open Science Frame (OSF; osf.io/azf4).

Search methods: eligibility criteria and study selection

To identify relevant studies, the search strategies were used in PubMed (31 January 2022), Embase (26 January 2022), and PsycINFO (26 January 2022) (Table 1). Searches were limited to human research only, as well as requiring papers to be in English. Duplicates were removed. An a priori decision was made to exclude papers if they investigated the other side of the bidirectional relationship between alcohol and circadian rhythm disruption, by which alcohol disrupts circadian systems. Similarly, review papers, editorial comments, and opinion papers that did not include original data were excluded from the full-text screen. The first screen involved determination of appropriateness based on the title and abstract_. Specifically, this included forms of circadian disruption such as social jet lag, extreme chronotypes, and night shift work affecting alcohol consumption or abstinence. Then, full-text versions were obtained for all of the papers meeting the inclusion criteria based on the title and abstract screen (100%). The screen was performed independently by one primary reviewer (M.J.N.) and two secondary reviewers (P.S.S. and R.R.) who made decisions for inclusion or exclusion; when consensus could not be achieved, discrepancies were resolved by the consensus reviewer (C.L.H-K.). The included literature was divided into two subgroups involving disruption of circadian rhythms: (i) social jet lag and chronotype and (ii) night shift work.

Table 1.

Database search strategies.

PubMed (312 results) (("Circadian disruption" [Text Word] OR "morningness" [Text Word] OR "eveningness" [Text Word] OR "chronotype" [Text Word] OR "shift work" [Text Word] OR "night shift" [Text Word] OR "social jet lag" [Text Word] OR "non-24" [Text Word]) AND "alcohol" [Text Word]) NOT ("Animals"[MeSH Terms] NOT ("Animals"[MeSH Terms] AND "Humans"[MeSH Terms]))

Embase (select filters, human only, male and female, uncheck 4 boxes at search start) (237 results) ("circadian rhythm"/exp OR "circadian"/exp OR "chronotype"/exp OR "jet lag"/exp OR "morningness"/exp OR "eveningness"/exp OR (((disrupt* OR dysregulat*) NEAR/2 circadian):ti,ab,kw) OR "jet lag":ti,ab,kw OR jetlag:ti,ab,kw OR "disrupted sleep":ti,ab,kw OR "night shift":ti,ab,kw OR eveningness:ti,ab,kw OR morningness:ti,ab,kw OR chronotype:ti,ab,kw OR "non-24":ti,ab,kw) AND "alcohol"/exp NOT ([animals]/lim NOT [humans]/lim)

PsycINFO (266 results) S1 MA ("Alcohol Drinking" OR "Alcoholism" OR "Alcoholics" OR "Alcoholic Intoxication") OR DE ("Alcohol Use Disorder" OR Alcoholism) S2 TI ("alcohol use" OR "alcohol consumption" OR "alcohol addiction" OR "alcohol misuse" OR "alcohol abuse") OR AB ("alcohol use" OR "alcohol consumption" OR "alcohol misuse") S3 S1 OR S2 S4 DE Human Biological Rhythms OR MA ("Circadian Rhythm" OR "Shift Work" OR "Circadian Clocks" OR "Sleep Disorders, Circadian Rhythm" OR "Chronobiology Disorders") S5 TI (((disrupt* OR dysregulat*) N2 circadian) OR "sleep–wake" OR "jet lag" OR "disrupted sleep" OR "night shift" OR "night shift work*" OR eveningness OR morningness OR chronotype OR chronodisruption OR "non-24") OR AB (((disrupt* OR dysregulat*) N2 circadian) OR "sleep–wake" OR "jet lag" OR jetlag OR "disrupted sleep" OR "night shift" OR "night shift work*" OR eveningness OR morningness OR chronotype OR "non-24") S6 S4 OR S5 S7 S3 AND S6

Additional Limitations: English language, human population, document type: journal article

A “reverse snowballing” approach (screening of all cited references of papers identified in the initial screen described above) was conducted using Web of Science to identify additional relevant papers (September–November 2022). First, the entire list of cited papers was refined in Web of Science (Clarivate PLC) by removing review papers and then using a Boolean approach to identify those with a circadian component: circadian OR chronotype OR eveningness OR morningness OR jetlag OR jet lag OR shift OR night shift. Then, Adobe Acrobat search function was used to scan each of the remaining papers to determine whether drink or alcohol appears in the results section; if so, a full-text screen was performed to determine whether the paper met the criteria for inclusion in the systematic review.

Study characteristics, assessments, and alcohol outcomes

Among papers selected for inclusion, pertinent study information was extracted. We included the total number of participants, mean age and/or age range, and sex. In addition, we included the measures of alcohol use, findings related to alcohol consumption. Alcohol-related outcomes were reported in various ways across studies, including g/day, g/week, g/month, number of drinks within the last month, number of drinks per day, number of drinks per week, AUD Identification Test (AUDIT) score, or a simple yes/no question pertaining to alcohol use. Therefore, we reported the increase, no change, or decrease of alcohol-related behavior. Finally, a brief description of the study design of each included paper was provided.

Statistical approach

The effects of circadian disruption and SABV on general alcohol outcomes within each category were determined and subjected to a chi-square (χ²) analysis. As age was reported in a variety of ways across studies (e.g. mean age of the study population, mean age of each experimental group, median age, and age range for the study population) (Supplementary Tables 1–3). When mean age was provided for each experimental group in a paper, we combined the data into a single group by decomposing the mean from all groups (https://www.statstodo.com/CombineMeansSDs.php). The age was reported as the mean ± SD and analyzed using a Student’s t-test (unpaired), utilizing the Statistical Package for the Social Sciences (SPSS v.27).

Differences were considered statistically significant if P <.05. Figures were generated using GraphPad Prism (v.9).

Results

The PRISMA Flow Chart of Study Design is depicted in Fig. 1. The initial screening of PubMed, PsycINFO, and Embase yielded 741 unique references. The first round of screening was accomplished with the use of Covidence software for systematic review management. The titles and abstracts were screened for relevant mentions of circadian rhythm disruption and alcohol or lifestyle factors yielding 284 articles. A full-text screen of these articles identified 101 papers that described studies of disrupted circadian rhythms and how it influences alcohol consumption, hence meeting the criteria for inclusion in this systematic review. An additional 76 relevant papers were identified through the reverse snowball approach, for a total of 177 included papers. Of these papers, 66 examined the relationship between social jet lag or chronotype and alcohol use (Supplementary Table S1), 101 examined the relationship between night shift work and alcohol use (Supplementary Table S2), and 10 examined various other forms of circadian disruption, including DSP, DSPS, and light at night exposure (Supplementary Table S3).

Figure 1.

PRISMA flow chart of study design

Effect of circadian cycle disruption on alcohol consumption

Of the 66 articles that examined the effects of social jet lag and chronotype and alcohol use (Fig. 2A), there was a significant effect of these factors (χ² = 124.6, P < .01). A statistically significant increase in alcohol consumption was observed in the majority of articles (86.4%) among those with the social jet lag and eveningness chronotype. The remaining articles reported either no significant effect on alcohol consumption (10.6%) or a significant decrease in alcohol use (3.0%).

Figure 2.

Social jet lag, chronotype, shift work studies and distribution of male and female participants. (A) Social jet lag and chronotypes: there was high concurrence that these factors increase alcohol use (n = 66). (B) Night shift: there are disparate alcohol use outcomes depending on whether all papers are grouped together (a) versus grouped by the sex or sexes of the participants. The difference in the outcome patterns among papers that include both sexes (b; n = 45), females only (c; n = 22), or males only (c; n = 33) highlights potential differences between men and women in the effects of night shift work on alcohol use. Furthermore, 9% (4 of 45) of the papers that included both sexes reported a significant effect in women but not men (referred to as “split outcomes”). One paper did disclose the sexes of the participants. (C) Male and females: (a) >80% of social jet lag and chronotype papers included both males and females. (b) In contrast, the night shift studies are skewed toward males and ~44% of the studies include participants of both sexes

Of the 101 articles that examined the effects of night shift work and alcohol use (Fig. 2B), there was a significant difference among frequencies associated with the various alcohol outcomes (χ² = 63.19, P < .01). A portion of the studies showed a significant increase in alcohol use (28.7%); however, the majority of studies (50.5%) reported no effect of shift work on alcohol consumption. The remaining articles reported a statistically significant decrease (16.8%) in alcohol use among night shift workers relative to day shift workers.

Of the 10 articles (Table S3) that examined the effects of other forms of circadian disruption on alcohol use, three studied DSP or DSPD; DSP was associated with increased alcohol use in two of the papers and the third paper reported that adolescents with more severe DSPD symptoms were more likely to have consumed alcohol. Statistical analyses were not conducted due to the small number of papers.

Sex differences

Of the 66 articles investigating social jet lag and chronotype on alcohol consumption (Fig. 2C), there was a significant effect of inclusion of participants by sex (χ² = 108.8, P < .05). The majority of the articles (81.5%) incorporated participants of both sexes. The remaining articles involved either only males (6.2%) or only female (10.8%) participants, and one study (1.5%) did not report population information (Williams et al. 2021).

Of the 101 studies on night shift workers and alcohol use (Fig. 2C), there was a significant effect of inclusion of participants by sex (χ² = 11.91, P < .05). Less than half of the articles included both sexes (44.6%). The remaining articles included either men only (32.7%) or women only (21.8%). One study (0.9%) did not report the sex of the participants (Asare-Anane et al. 2015). Of the 45 studies that contain both male and female participants, there was a significant effect of night shift work on alcohol (χ² = 16.9, P < .05). A portion of the studies (28.9%) reported an increase in alcohol consumption. The remaining articles reported either no effect (44.4%) or a decrease (17.8%) in alcohol consumption. Other articles (8.9%) reported disparate effects based on sex where females reported either increased (two articles) or decreased (two articles) alcohol consumption, whereas there were no significant effects reported for males in any of these four articles. In 22 studies of shift work that include only women (Fig. 2C), there was a significant effect of alcohol use (χ² = 15.27, P < .05). Two studies reported an increase in alcohol consumption among participants (9.1%), whereas the majority of the remaining articles described no effect (63.6%) or reported a decrease (27.3%) in alcohol consumption. Among the 33 night shift studies that include only men (Fig. 2C), we also found a significant effect (χ² = 14.18, P < .05). A portion of the studies reported an increase in alcohol consumption (39.4%); however, about half of the articles reported no effect of shift work on alcohol use (51.2%). The remaining articles reported a decrease (9.1%) in alcohol intake.

Age

The mean age (±SD) of the study population was significantly younger [t(145) = −6.99, P < .001] for social jet lag and chronotype papers (28.55 ± 14.15 years) than the individuals included in the shift work papers (41.31 ± 7.89 years). The mean of a study’s population was provided in 41 social jet lag/chronotype papers and in 39 shift work papers; the mean of the study population was calculated for 18 social jet lag/chronotype papers and for 49 shift work papers that provided means strictly for individual groups. Therefore, we were able to aggregate the mean age (±SD) of the study population for 59 out of 66 social jet lag/chronotype papers and 88 out of 101 shift work papers.

Discussion and Conclusion

Consistent with our hypothesis, this systematic review revealed that circadian rhythm disruption increases alcohol consumption. The association between social jet lag or eveningness chronotypes and alcohol use was remarkably consistent; nearly all the studies reported an increase in alcohol use. Furthermore, the vast majority of these studies included both males and females, and there was no evidence of a sex difference in outcomes.

Individuals with the eveningness chronotype are likely to be awake much later in the evening and consumption of alcohol is a common night-time activity and possibly used initially to induce sleep (Koob and Colrain 2020). In contrast, early morning types may be less likely to consume alcohol early in the day because it is less socially acceptable and/or effects may interfere with daytime responsibilities. However, the consistency of the effect of social jet lag and chronotype on alcohol use suggests the likelihood of a fundamental biological mechanism as well. eveningness is associated with altered neural responses to reward; among individuals with late chronotypes a reduction in activation of the medial prefrontal cortex (mPFC) is associated with greater alcohol consumption (Hasler et al. 2013). A similar reduction in mPFC response to reward is induced by recent social jet lag (Hasler et al. 2022b). Furthermore, delayed phase rest–activity rhythms, which are a hallmark of eveningness, are associated with increased caudate dopamine 1 receptor (D1R) availability and increased subjective stimulant effects and craving (Zhang et al. 2021).

Social jet lag and eveningness also are associated with several affective and cognitive factors known to increase the risk for alcohol use, including depressive symptoms (Henderson et al. 2019; Ojio et al. 2020; Taillard et al. 2021; Jongte and Trivedi 2022) and shifts in attention, cognitive processing, emotional processing, impulsivity, decision-making, and risk-taking (Berdynaj et al. 2016; Correa et al. 2017; Smarr and Schirmer 2018; Smit et al. 2020; Evans and Norbury 2021; Beracci et al. 2022). Eveningness is associated with increased impulsivity at both the state and trait levels (Hasler et al. 2022b), which is a well-established risk factor for alcohol use (Dick et al. 2010). A recent study demonstrated that the association between eveningness and alcohol use is mediated by impulsivity (Evans and Norbury 2021).

Social jet lag and eveningness are consistently associated with reduced sleep duration and quality (Taillard et al. 1999; Wittmann et al. 2006; Merikanto et al. 2012; Merikanto and Partonen 2020; Jongte and Trivedi 2022). Sleep disturbances are common among individuals with AUD and several longitudinal studies have suggested that sleep disturbances among the general population predict increases in future alcohol use in both children and adults (Wong et al. 2004, 2010; Pieters et al. 2015; Hasler et al. 2016; Mike et al. 2016; Nguyen-Louie et al. 2018). However, whether sleep disruption is a direct contributor to increased alcohol use among individuals with social jet lag or late chronotypes is less clear. Some studies have demonstrated that eveningness predicts future increases in alcohol use (Nguyen-Louie et al. 2018; Hasler et al. 2022a); however, the association between eveningness and alcohol consumption remains significant after controlling for sleep quality (Evans and Norbury 2021).

In contrast to social jet lag and chronotype studies, evidence of alcohol use gathered from night shift studies was not as clear-cut. Approximately half the shift work studies report no effect on alcohol use. Among the remaining studies, slightly more reported an increase in alcohol use than a decrease in alcohol use associated with shift work. When analyzing the sex distribution among participants, it became clear that there was a discrepancy in shift work studies compared to studies investigating other forms of circadian rhythm disruption. Specifically, approximately half of the shift work studies included both men and women, and when represented only one sex, they were more often only men than only women. The outcome patterns for alcohol use were very similar when all of the studies were analyzed together and when studies including only men were analyzed. In contrast, when female-only studies were analyzed separately, only a small minority of them reported increased alcohol consumption. The predominant outcome was no effect, with a much larger proportion of female-only studies reporting a reduction in alcohol use among night shift workers than among male-only studies. The four night shift studies that did include SABV in the statistical analyses reported a significant effect on alcohol consumption observed in women, but not men. When considered together, the studies did not support the hypothesis that females are more vulnerable than males to the effects of circadian disruption on alcohol use. However, our data reinforce the importance of including both sexes in research and make a strong argument for statistically analyzing sex as an independent variable.

This systematic review draws attention to the question of why the studies of social jet lag and eveningness are highly consistent in reporting increased alcohol use, whereas the effects of night shift on alcohol use are inconsistent across studies. One possible explanation for this discrepancy is that the circadian peak in alcohol craving more often corresponds with alcohol availability among individuals with social jet lag and eveningness than among those working the night shift. Regardless of day of the week, alcohol craving typically peaks at 21:00 and then declines through the evening to reach a trough at 09:00, and does not appear to be altered by sleep timing (Hisler et al. 2022). Therefore, individuals with social jet lag or an eveningness chronotype may have more consistent access to alcohol in the evening hours when alcohol craving is elevated than people who work the night shift and presumably have limited access to alcohol while at work. This may be explained by diurnal rhythms in the biological, pharmacokinetics, and rewarding properties of alcohol revealed by studies demonstrating variation across the day in alcohol dehydrogenase (Salsano et al. 1990), aldehyde dehydrogenase (Brick et al. 1984), and dopamine signaling (Ferris et al. 2014) in rodents. Likewise, diurnal fluctuations in breath alcohol concentrations have been confirmed in adults consuming a fixed oral dose of alcohol at various times of day (Rukmini et al. 2021). Therefore, circadian rhythms may influence blood alcohol concentrations and neural responses and interact with timing of alcohol intake to create differential risk for development of AUD among people of eveningness versus morningness chronotype (Danel and Touitou 2004).

Individuals who experience social jet lag or have an eveningness chronotype also have daily schedules that are far more suitable for sustaining a normative social drinking life than individuals who are working the night shift. In other words, night shift workers may experience greater misalignment of work/wake and social schedules than individuals with eveningness or social jetlag because night shift workers are at work during the time period that drinking is most prevalent (Dawson 1996).

Another potentially important discrepancy between social jet lag/chronotype and shiftwork studies is the age of the participants. The mean age of individuals included in social jet lag/chronotype papers was more than a decade younger than the mean age of individuals included in shift work papers. This may be important because at the individual and population levels, alcohol consumption tends to decline with increasing age among adults (Moore et al. 2005). In addition, advancing age may affect earlier wake-up time (morningness) and has been associated with reduced D1R in the caudate, putamen, and nucleus accumbens and reduced subjective alcohol effects (Zhang et al. 2021). Therefore, age-related changes in alcohol consumption, dopamine signaling, and its influence on sensitivity to reward may contribute to the discrepant outcomes among the different types of studies included here.

This systematic review highlights a significant relationship between the disruption of circadian rhythms via social jet lag or eveningness and an increase in alcohol consumption. These forms of circadian disruption predict increases in future alcohol use (Urbán et al. 2011; Pasch et al. 2012; Warren et al. 2017), suggesting a potential causal relationship in which circadian rhythm disruption promotes alcohol use. Despite the consistent and strong relationship between eveningness and increased alcohol use, there are currently no published intervention trials. However, bright-light therapy offers the possibility of a low-cost treatment. It is used to treat circadian rhythm disorders and improves depressive symptoms in people with late chronotype (Chan et al. 2022), but whether it also normalizes alcohol consumption remains an open question.

There are several limitations of this review that warrant mention. First, studies using different alcohol measures were grouped together for analysis. The assessments varied among quantitative measures of alcohol consumed over various time periods, frequency of alcohol consumption, AUDIT scores, discrete measures (i.e. yes/no responses to past alcohol use), and abstinence rates. It is possible that some of the measures employed to assess alcohol use were not sufficiently sensitive to allow detection of smaller effects. Second, studies were categorized as including both sexes regardless of how skewed the proportions were toward one sex; this may have resulted in over-reporting of female participation, particularly in the shift work studies. In addition, a large proportion of the female-only night shift studies collected data from nurses and midwives; the heavy reliance on women from a highly educated, skilled profession may limit the generalizability of these data.

Taken together, the systematic review revealed that not all forms of circadian rhythm disruption have comparable relationships with alcohol use. The most at-risk individuals for increased alcohol consumption appear to be those with social jet lag or an eveningness chronotype because the results across a large number of studies consistently reported increased alcohol use. Likewise, DSP was consistently associated with increased alcohol use, although the number of papers addressing this topic was relatively small. Improved understanding of the relationship between circadian disruption, alcohol intake, age, and gender differences may provide insight into temporal risk factors that could be valuable to clinicians who treat individuals at high risk of developing AUD. Given the highly reproducible effect of chronotype and social jet lag on alcohol use, these data may also be of interest to researchers studying other risk factors for AUD that may interact with or be potentiated by the disruption of circadian rhythms.

Author contributions

Morgan J. Nelson (Conceptualization [lead], Data curation [equal], Formal analysis [lead], Investigation [lead], Methodology [lead], Project administration [lead], Validation [equal], Visualization [lead], Writing—original draft [lead], Writing—review & editing [lead]), Paul S. Soliman (Data curation [equal], Writing—review & editing [supporting]), Ryan Rhew (Data curation [supporting], Writing—review & editing [supporting]), Rachel N. Cassidy (Visualization [supporting], Writing—review & editing [supporting]), and Carolina L. Haass-Koffler (Conceptualization [equal], Formal analysis [supporting], Funding acquisition [lead], Investigation [equal], Methodology [equal], Project administration [equal], Supervision [equal], Validation [equal], Visualization [equal], Writing—review & editing [equal])

 

Conflict of interest: None declared.

Funding

The study was conducted in the Clinical Neuroscience Laboratory (PI: C.L.H-K.) funded by the National Institute on Alcohol Abuse and Alcoholism (R01 AA026589, R01 AA027760, R21 AA027614) and by the National Institute of General Medical Sciences (NIGMS), Center of Biomedical Research Excellence (COBRE, P10 GM130414).