Shift Work and Metabolic Disease: How Working Against Your Biological Clock Drives Insulin Resistance

Why Shift Work Damages Metabolism Even When Diet and Exercise Are Reasonable

Shift work is one of the most underappreciated drivers of metabolic disease in modern clinical practice. Patients who eat reasonably well, train regularly, and avoid the obvious dietary pitfalls can still arrive with a deteriorating metabolic profile that does not match their effort. When the work history surfaces in the consultation — five years on rotating nights as an ICU nurse, fifteen years driving long-haul, a decade in hospitality service finishing shifts at 2am — the metabolic picture suddenly makes sense. The diet is not the dominant variable. The schedule is.

Shift work and metabolic disease are linked through a mechanism that is structural rather than behavioral. The body is not designed to eat, work, and metabolize fuel at 3am. The peripheral clocks that govern liver glucose output, pancreatic insulin secretion, adipose tissue lipid handling, and skeletal muscle insulin sensitivity all expect to be active during daylight and quiescent during darkness. When the work schedule forces metabolic activity into the biological night, every one of those clocks loses its alignment with the central clock in the suprachiasmatic nucleus — and the metabolic system begins to fail in ways that no amount of dietary optimization fully compensates for.

This post addresses what shift work specifically does to metabolic health, why it produces a distinct lab and CGM signature that overlaps with but is not identical to the standard insulin resistance presentation, what the recovery picture looks like in former shift workers, and what the realistic harm-reduction protocol looks like for patients who cannot leave the work itself.

What you will learn:
Why shift work produces metabolic damage independent of diet and total sleep hours | The specific lab and CGM signature seen in current and former shift workers | Why GGT is often the most disproportionately elevated marker in this population | The “former shift worker phenotype” and what residual damage looks like years after the schedule changes | The practical harm-reduction protocol for patients who cannot stop shift work

Circadian Misalignment Is the Mechanism, Not Sleep Deprivation Alone

The most common framing of shift-work harm is that it produces sleep deprivation, and sleep deprivation drives metabolic dysfunction. This is partially true, but it misses the more fundamental mechanism. As covered in detail in the post on how poor sleep causes insulin resistance, insufficient sleep impairs insulin sensitivity through cortisol elevation, sympathetic activation, GLUT4 suppression, and incretin blunting. These mechanisms are real and they apply to shift workers — but they describe acute and intermittent sleep loss, not the chronic structural problem that shift work imposes.

Shift work is not just sleep deprivation. It is scheduled circadian misalignment. The patient is not simply tired — they are eating during the biological night, working during the biological night, exercising during the biological night, and attempting to sleep during the biological day. Every one of those activities produces a metabolic response that depends on circadian context, and every one of them produces a worse metabolic response when performed at the wrong biological time.

The same meal eaten at 2pm and at 2am does not produce the same glucose curve. The 2am meal lands in a metabolic window where pancreatic beta cells have reduced insulin secretion capacity, where peripheral insulin sensitivity is suppressed, where melatonin is actively interfering with insulin signaling at the MTNR1B receptor on beta cells, where mitochondrial oxidative capacity is at its circadian low, and where hepatic glucose output is rising in anticipation of the morning awakening that will not actually occur. The food has not changed. The metabolic context has been inverted.

This distinction matters clinically because it explains why interventions that focus solely on improving sleep duration in shift workers — earplugs, blackout curtains, longer rest periods — produce only partial improvement. The sleep is one variable. The misalignment between meal timing, work timing, light exposure, and biological time is the larger variable, and it persists even when sleep duration is restored.

The circadian rhythm and metabolism post details how peripheral clocks govern glucose disposal, lipid handling, and hormonal rhythms across the 24-hour day. Shift work breaks the synchrony between those peripheral clocks and the central clock — and breaks the synchrony between the peripheral clocks and the actual behavior of the patient. The liver expects food at 8am. The food arrives at 3am. The mismatch is what does the damage.

What Shift Work Specifically Does to Metabolic Physiology

The metabolic consequences of shift work operate through several pathways that compound each other across days, weeks, and years of exposure.

Pancreatic beta cell function follows a circadian rhythm, with peak insulin secretion capacity occurring during daylight hours and a measurable trough during the biological night. When food arrives during the biological night, the insulin response to a given glucose load is reduced — the same meal produces a higher peak glucose and a more prolonged elevation than the same meal eaten during daylight. Over years of repeated exposure, this chronic mismatch contributes to beta cell stress, glucotoxicity, and accelerated progression toward beta cell exhaustion.

Hepatic clock genes — including BMAL1, CLOCK, PER, and CRY — orchestrate the rhythmic expression of enzymes governing gluconeogenesis, glycolysis, lipogenesis, and bile acid synthesis. In shift workers, these hepatic rhythms become dampened and phase-shifted. The liver continues to produce glucose during periods when it should be storing it, and continues to synthesize fat during periods when it should be oxidizing it. The result over time is the hepatic insulin resistance and de novo lipogenesis pattern that predisposes shift workers disproportionately to fatty liver — even at body weights that would not predict significant hepatic dysfunction in a daytime worker.

Adipose tissue insulin sensitivity is also circadian. The capacity of adipocytes to suppress lipolysis in response to insulin is greatest during the daytime and weakest at night. When shift workers eat overnight, the postprandial insulin signal does not effectively suppress fatty acid release from adipose tissue, producing a paradoxical state in which both glucose and free fatty acids are elevated in the circulation simultaneously. This combination is highly inflammatory at the vascular endothelium and contributes to the elevated cardiovascular risk that epidemiological studies of shift workers consistently document.

Cortisol rhythm becomes distorted in a specific way that is clinically distinct from the rhythm seen in daytime workers with poor sleep. As covered in nighttime cortisol dysregulation, normal cortisol biology is a sharp morning peak and a low evening trough. In shift workers, the peak flattens and the trough elevates — the curve becomes shallower and chronically displaced. The result is a pattern that I describe to patients as a “false dawn phenomenon all day long” — hepatic glucose output is elevated almost continuously, and fasting glucose readings taken at any point in the day tend to run higher than the diet would predict.

Melatonin, the body’s master signal for biological night, peaks during the early morning hours regardless of what the patient is doing. As detailed in melatonin and metabolism, melatonin reduces insulin secretion through MTNR1B receptors on pancreatic beta cells. When shift workers eat overnight, they are eating during peak melatonin — and the carbohydrate load they consume produces a higher glucose excursion because insulin secretion is being actively suppressed by the very hormone that the schedule is artificially overriding. The MTNR1B variant that confers higher type 2 diabetes risk is particularly relevant in shift-working populations, where the gene-environment interaction between circadian timing of food and genetic insulin secretion capacity is repeatedly stressed.

Sleep architecture itself is altered in shift workers in ways that are not captured by total sleep time. Daytime sleep tends to be shorter, more fragmented, lighter, and contains less slow-wave sleep and less REM than equivalent-duration nighttime sleep. The hormonal repair functions that depend on slow-wave sleep — growth hormone release, testosterone production, prolactin rhythms — are correspondingly diminished. Sleep apnea risk is also elevated in shift workers, particularly long-haul drivers and other sedentary occupational groups, and the apnea-driven overnight glucose pulses described in sleep apnea and metabolic disease compound the circadian misalignment.

The combined effect of these mechanisms is a metabolic system that is operating against its own design specifications for hours every day, week after week, year after year. The damage is cumulative, mechanistic, and largely invisible until it shows up in the lab work.

The Lab Signature of the Shift Worker

Shift workers arrive in clinical practice with a distinctive lab and CGM signature that overlaps with the standard insulin resistance picture but contains specific markers that are disproportionately elevated. Recognizing this signature in advance shifts the clinical approach.

Fasting insulin is typically elevated, often disproportionately so relative to HbA1c. The hyperinsulinemia is compensatory — the beta cells are producing more insulin to maintain glucose at acceptable levels in the face of chronic insulin resistance. HbA1c may still appear acceptable on paper because the compensation is working. This is the precise pattern that makes the standard German GP screening miss the diagnosis entirely: glucose looks fine, HbA1c looks fine, the patient is sent home with a clean panel, and the underlying hyperinsulinemia continues to drive damage for another decade.

The TG/HDL ratio is consistently poor in shift workers. Triglycerides are often mildly to moderately elevated, HDL is suppressed, and the ratio reliably exceeds 2.0 in patients who have been doing shift work for years. This pattern reflects the hepatic and adipose insulin resistance described above — the liver is producing more triglyceride-rich VLDL particles, and the impaired insulin signaling at adipose tissue is keeping circulating lipid traffic elevated.

The marker that I find disproportionately elevated in shift workers, more than any other routine lab, is GGT. The mechanism is multifactorial. Chronic circadian disruption impairs hepatic detoxification rhythms, reduces glutathione turnover capacity, and elevates oxidative stress at the liver. Many shift workers also use more alcohol than average — particularly hospitality workers and long-haul drivers — and the combination of circadian-driven hepatic stress with alcohol exposure produces a GGT elevation that is often clinically striking.

GGT is one of the most underutilized markers in metabolic medicine. In conventional practice it is typically interpreted only as an alcohol marker or as a generic liver enzyme. In functional metabolic interpretation, as detailed in ALT and GGT as early liver markers, GGT is a sensitive proxy for oxidative stress at the liver, glutathione depletion, and metabolic strain — and in shift workers it is frequently the most abnormal marker on the panel even when HbA1c and ALT remain unremarkable.

ALT runs mildly elevated in many shift workers, consistent with the early fatty liver picture that develops from chronic circadian misalignment of hepatic metabolism. hsCRP is variable but frequently elevated — sleep deprivation amplifies inflammatory tone, and the chronic low-grade inflammation pattern detailed in chronic low-grade inflammation is common in this population.

Fasting glucose tends to run higher than the diet would predict, reflecting the distorted cortisol curve and the persistent hepatic glucose output that shift workers display almost continuously rather than only at the dawn window. CGM data, where available, shows the pattern: elevated glucose during the biological night when the patient is working and eating, exaggerated glucose responses to overnight meals, blunted insulin sensitivity to the morning meal taken before sleep, and a flattened diurnal glucose rhythm overall — the glucose curve loses its day-night structure and becomes a chronically elevated, more variable trace.

The composite lab signature that I recognize in current shift workers is therefore: fasting insulin disproportionately elevated, TG/HDL ratio above 2.0, GGT disproportionately elevated, ALT mildly elevated, hsCRP variable but often elevated, fasting glucose higher than diet predicts, and HbA1c that may still be deceptively acceptable. Patients with this constellation, in the setting of a current or recent shift work history, are not “early insulin resistance” patients in the conventional sense. They are circadian-misaligned metabolic patients, and the intervention has to address the misalignment, not just the dietary pattern.

Clinical Perspective: What I See in Practice

Shift workers and former shift workers represent roughly 10 to 25 percent of the metabolic patients I see, and the proportion runs higher among the type 2 diabetes and severe insulin resistance cases — circadian disruption is a major driver of glucose dysregulation, and the patients who progress most quickly often have a shift work history embedded somewhere in their occupational past.

The clinical pattern I see most often in current shift workers is the “stressed insulin-resistant phenotype” — fasting insulin disproportionately high, TG/HDL ratio poor, GGT elevated more than the rest of the panel would predict, hsCRP variable, and an HbA1c that is only mildly elevated despite significant underlying dysfunction because the patient is still compensating with hyperinsulinemia. The first time I see GGT running high in a patient whose ALT is barely above normal and whose alcohol intake is unremarkable, I check the work history. The shift work explanation appears more often than chance would predict.

The pattern that surprises patients most is the disconnect between their HbA1c and their fasting insulin. Their GP has measured HbA1c, found it acceptable, and reassured them. They feel exhausted, gain weight, and recognize that something is wrong, but the lab work they have been shown does not corroborate the experience. When fasting insulin is added to the panel and reveals a value of 18 or 22 µIU/mL — well above my functional optimal of below 5 — the picture suddenly aligns with the symptoms. The glucose looks acceptable because the pancreas is working overtime to keep it there. The work is what is making the pancreas work overtime.

The second clinical observation worth naming is that current shift workers often look metabolically older than their chronological age. A 38-year-old nurse on rotating nights for ten years frequently has the lab profile of a 55-year-old daytime worker with the same diet and activity level. The skin, the visceral adiposity, the morning energy, and the labs all converge on a more advanced metabolic age than the calendar would suggest. This is not exaggeration — it is a coherent clinical observation that lines up with the molecular damage that chronic circadian misalignment produces.

The “former shift worker phenotype” is one of the most clinically interesting patterns I encounter. Someone stops shift work, their subjective energy improves within months, and they expect their metabolic markers to normalize on the same timeline. They do not. Sleep and cortisol rhythm partially improve in weeks to months. Fasting insulin and triglycerides typically take six to twenty-four months to recover, and only with aggressive lifestyle correction. GGT can take longer. Visceral adiposity is often stubborn. Autonomic dysfunction — the exaggerated stress response, the early waking, the reduced ability to handle a poor night of sleep — can persist surprisingly long.

What I see in patients who did shift work for ten or twenty years and stopped five years ago is something that behaves like residual circadian damage. They have an exaggerated glucose response to a single poor night of sleep that a non-shift-work patient would handle without disruption. Their metabolic flexibility is reduced. Their fasting glucose runs higher than their diet predicts. Their stress tolerance is diminished. The body behaves as if the circadian system was reprogrammed during the exposure period and never fully returned to its baseline configuration.

I do not tell patients that this damage is permanent. Some former shift workers — particularly those who were younger when they stopped, who had shorter exposure periods under seven years, who corrected diet aggressively, who trained, and who restored sleep consistency — appear to recover fully. But after long exposure of fifteen or twenty years of nights, I am clinically suspicious that some residual circadian, autonomic, mitochondrial, and hepatic dysfunction persists indefinitely. Long-term night-shift exposure behaves, in my clinical observation, almost like a chronic environmental toxin exposure — the recovery is not symmetrical with the exposure, and the damage is not fully reversible at the level the patient would hope.

This framing matters for patients who are still doing shift work and are weighing whether to continue. I do not lecture. I show them their labs, explain the mechanism, and let the data make the case. The decision about whether to leave the work belongs to the patient.

The Most Common Dietary Mistake in Shift Workers

The single dietary mistake I correct first in shift workers is eating high-carbohydrate meals during the biological night. At 2 or 3am, insulin sensitivity is impaired, glucose tolerance is reduced, melatonin is interfering with insulin secretion, cortisol rhythm is abnormal, and mitochondrial efficiency is at its circadian low. The exact same meal that produces a tolerable glucose curve at 2pm produces a markedly worse curve at 2am. As detailed in poor sleep and glucose control, the metabolic context shapes the response to food as much as the food itself does — and during the biological night, the context is hostile.

The actual food choices during overnight shifts in real-world practice are often poor. Ultra-processed convenience foods, sandwiches and wraps from cafeteria vending, pastries, chips, sugary coffee drinks, energy drinks, cereal bars, and pizza dominate the hospital break room and the truck stop. Even the more health-conscious shift workers tend to fall back on oats, bananas, protein bars, granola, or frequent grazing — all of which deliver carbohydrate loads that arrive at the worst possible biological time for insulin handling.

I do not impose a rigid “never eat at night” rule. In real-world shift workers, that approach fails on adherence within weeks. The clinical compromise is staged. The first interventions are to stop continuous grazing across the shift, eliminate sugary caffeine drinks, remove ultra-processed snack foods, and remove the carbohydrate-heavy meals that are producing the worst overnight glucose excursions. The food during the shift becomes protein-focused — eggs, meat, fish, hard cheese, or animal-based foods that produce minimal glucose response and maintain steady energy without spiking insulin. Carbohydrate density is kept low overnight. The largest meal is anchored to the patient’s waking biological window, after sleep, in daylight hours where possible.

The frequency of eating is reduced. Most shift workers come in eating five or six times across a shift to combat fatigue. Compressing intake to one or two meals during the shift, both of them protein-and-fat focused, dramatically reduces the metabolic insult of the night. The fatigue does not disappear, but the glucose curve becomes much flatter, and over weeks the energy stability actually improves rather than worsens — because the reactive cravings driven by overnight glucose volatility are no longer fueling the cycle.

For patients with severe insulin resistance, established type 2 diabetes, or significant obesity, I move them progressively toward a tighter feeding window aligned as closely as possible to their daylight biological window — even if that window is shifted relative to the rest of the population. The principle is alignment of food with light, not adherence to a generic clock time.

A patient who sleeps from 9am to 4pm has a “biological day” from approximately 4pm to 9am from the perspective of light exposure, but their peripheral clocks are still trying to rhythm to the actual solar day. The compromise is to eat in the patient’s subjective active period, while keeping the largest meals as close as possible to their daylight exposure window. Perfection is not the goal. Reducing the misalignment is the goal.

For patients managing active insulin resistance, the carbohydrate target remains around 50 grams per day, with the bulk of those carbohydrates eaten during the daylight window and minimal carbohydrate density overnight. The framework laid out in The Animal-Based Protocol for Insulin Resistance → translates well to shift-working populations because the protein-and-fat foundation is metabolically forgiving across the 24-hour cycle in a way that carbohydrate-dense meals are not.

The Harm-Reduction Protocol for Patients Who Cannot Stop Shift Work

Most shift workers cannot simply leave the work. The income, the seniority, the family circumstances, and the labor market structure mean that “stop doing shift work” is not actionable advice for the majority of patients in front of me. The clinical task is therefore harm reduction — identifying the highest-leverage interventions that produce the largest metabolic improvement when the schedule itself cannot change.

In practice, two or three interventions produce the majority of the gain.

Strict light management is the most underestimated and probably the most powerful intervention available. The principle is to maximize light exposure during the work shift and aggressively minimize light exposure after the shift ends. During the night shift, the work environment should be brightly lit — bright artificial light during the active period helps the central clock and improves alertness. After the shift ends, the patient wears dark wraparound sunglasses during the commute home to prevent morning sunlight from setting the clock to a daytime phase the patient is about to fail to honor.

The sleep environment is fully blacked out — heavy curtains, no light leakage, eye mask if needed. The room is cold and silent. Morning sunlight before sleep is avoided more rigorously than almost any other variable. On off-days, the patient gets aggressive bright daylight exposure within the first hour of waking to reset what alignment is recoverable. This intervention alone, applied consistently, often produces measurable improvements in fasting glucose, cortisol rhythm, sleep depth, and appetite regulation within four to six weeks.

Removal of nighttime carbohydrate load is the second highest-leverage intervention. As described above, the overnight shift becomes protein-focused, with minimal sugar and refined carbohydrate, and reduced eating frequency. The metabolic payoff is usually visible in TG/HDL, fasting insulin, and GGT within twelve weeks, often without major calorie restriction or weight loss as the primary mechanism.

Resistance training to protect muscle and metabolic signaling is the third intervention. Skeletal muscle is the largest insulin-sensitive tissue and behaves as a metabolic sink that improves glucose disposal even in the face of circadian disruption. Twenty to thirty minutes of resistance training three times per week produces meaningful improvements in insulin sensitivity, HbA1c trajectory, and body composition. The timing should be after waking or before the shift begins — never immediately before intended sleep, when the cortisol and adrenaline elevation from training will fragment whatever sleep follows. Walking after meals, where the schedule permits, adds further glucose disposal capacity through the AMPK pathway that operates independently of insulin.

Supportive interventions that I add when the foundational three are in place include magnesium glycinate to support sleep depth, creatine to support muscle and cognitive function under fatigue, adequate sodium and mineral intake to compensate for the stress-driven losses that accompany shift work, sleep consistency on weekends and off-days rather than swinging the schedule back to “normal,” and limiting alcohol after shifts — particularly because the GGT elevation that is already present from the work itself is compounded by even modest alcohol intake.

The realistic clinical goal is not optimal ancestral biology. It is reducing the gap between the patient’s biology and their occupational reality enough to prevent metabolic collapse. Patients who apply the light, food, and training interventions consistently can hold their metabolic markers stable across years of continued shift work — not optimal, but stable. That is the achievable goal, and for most patients it is the appropriate one.

Shift work is the occupational extreme of a broader phenomenon — chronic mismatch between human biology and modern schedules — and fits within the unified framework laid out in the cornerstone on the sleep and circadian system as the foundation of metabolic health.

A Note on Uncertainty

The epidemiological evidence linking shift work to type 2 diabetes, cardiovascular disease, metabolic syndrome, and certain cancers is strong and consistent across multiple large cohort studies. The mechanistic evidence for circadian misalignment as a driver of insulin resistance, hepatic dysfunction, and impaired glucose tolerance is also well-established in controlled human and animal studies.

What remains less certain is the precise dose-response relationship between years of shift work and irreversible metabolic damage, the individual variability in resilience to circadian disruption, and the degree to which various harm-reduction interventions actually mitigate the long-term risk in real-world adherence conditions. The “former shift worker phenotype” I describe in clinical practice is consistent with what the molecular research on circadian damage would predict, but the specific biomarker thresholds for residual damage and the timeline for recovery are not yet standardized.

The recommendations in this post reflect functional medicine clinical practice and emerging research. They are educational and should not replace individualized medical care. Patients with established type 2 diabetes, cardiovascular disease, or sleep disorders should coordinate any dietary, exercise, or sleep schedule changes with their treating physicians.

People Also Ask

About the Author

Morteza Ariana is a State-Certified Functional Nutritionist based in Germany, specializing in insulin resistance, type 2 diabetes, and root-cause metabolic restoration. He holds advanced training in systems-based physiology and has worked with patients across the U.S. and Europe for over 10 years.

His clinical framework is built around a core principle that mainstream medicine consistently overlooks: chronically elevated insulin — not blood glucose — is the earliest and most actionable driver of metabolic disease. That conviction was shaped in part by his own experience with hyperinsulinemia in 2016, and deepened through a decade of clinical practice and the study of leading researchers in metabolic medicine including Benjamin Bikman, Joseph Kraft, Gerald Reaven, Jason Fung, and Stephen Phinney.

His work focuses on identifying and correcting the upstream metabolic signals — insulin load, liver-gut axis dysfunction, circadian misalignment, and micronutrient gaps — that standard screening misses entirely. Patient outcomes are documented, anonymized, and published on this site.

Read the full bio →

If this resonates, the next step is clarity

The Metabolic Restoration Blueprint is a structured 12-week framework designed to correct upstream metabolic drivers — not just manage symptoms.

Explore the Metabolic Restoration Blueprint →

The Animal-Based Protocol for Insulin Resistance →

Key References

1. Scheer FA, Hilton MF, Mantzoros CS, Shea SA. Adverse metabolic and cardiovascular consequences of circadian misalignment. PNAS. 2009;106(11):4453–4458. 🔗 https://pubmed.ncbi.nlm.nih.gov/19255424/
2. Morris CJ, Yang JN, Garcia JI, et al. Endogenous circadian system and circadian misalignment impact glucose tolerance via separate mechanisms in humans. PNAS. 2015;112(17):E2225–E2234. 🔗 https://pubmed.ncbi.nlm.nih.gov/25870289/
3. Vetter C, Dashti HS, Lane JM, et al. Night shift work, genetic risk, and type 2 diabetes in the UK Biobank. Diabetes Care. 2018;41(4):762–769. 🔗 https://pubmed.ncbi.nlm.nih.gov/29440150/
4. Pan A, Schernhammer ES, Sun Q, Hu FB. Rotating night shift work and risk of type 2 diabetes: two prospective cohort studies in women. PLoS Medicine. 2011;8(12):e1001141. 🔗 https://pubmed.ncbi.nlm.nih.gov/22162955/
5. Vyas MV, Garg AX, Iansavichus AV, et al. Shift work and vascular events: systematic review and meta-analysis. BMJ. 2012;345:e4800. 🔗 https://pubmed.ncbi.nlm.nih.gov/22835925/
6. Knutsson A. Health disorders of shift workers. Occupational Medicine. 2003;53(2):103–108. 🔗 https://pubmed.ncbi.nlm.nih.gov/12637594/
7. Wang F, Zhang L, Zhang Y, et al. Meta-analysis on night shift work and risk of metabolic syndrome. Obesity Reviews. 2014;15(9):709–720. 🔗 https://pubmed.ncbi.nlm.nih.gov/24888416/
8. Lyssenko V, Nagorny CL, Erdos MR, et al. Common variant in MTNR1B associated with increased risk of type 2 diabetes and impaired early insulin secretion. Nature Genetics. 2009;41(1):82–88. 🔗 https://pubmed.ncbi.nlm.nih.gov/19060908/
9. Bass J, Takahashi JS. Circadian integration of metabolism and energetics. Science. 2010;330(6009):1349–1354. 🔗 https://pubmed.ncbi.nlm.nih.gov/21127246/
10. Sharma A, Laurenti MC, Dalla Man C, et al. Glucose metabolism during rotational shift-work in healthcare workers. Diabetologia. 2017;60(8):1483–1490. 🔗 https://pubmed.ncbi.nlm.nih.gov/28528402/
11. Gan Y, Yang C, Tong X, et al. Shift work and diabetes mellitus: a meta-analysis of observational studies. Occupational and Environmental Medicine. 2015;72(1):72–78. 🔗 https://pubmed.ncbi.nlm.nih.gov/25030030/
12. Ruggiero JS, Redeker NS. Effects of napping on sleepiness and sleep-related performance deficits in night-shift workers: a systematic review. Biological Research for Nursing. 2014;16(2):134–142. 🔗 https://pubmed.ncbi.nlm.nih.gov/23411360/
13. Wang XS, Armstrong ME, Cairns BJ, Key TJ, Travis RC. Shift work and chronic disease: the epidemiological evidence. Occupational Medicine. 2011;61(2):78–89. 🔗 https://pubmed.ncbi.nlm.nih.gov/21355031/