What earlier evidence established about fasting, shift work, and metabolic risk
Night-shift work is consistently linked to circadian disruption, and reviews of healthcare-worker cohorts describe a dose-response relationship between night-shift exposure and cardiometabolic risk, with circadian misalignment proposed as the core pathological driver rather than sleep loss alone [3]. That framing matters because it predicts that when people eat may matter independently of how much they eat. Controlled feeding and simulated shift-work studies have shown impaired glucose tolerance, insulin resistance, and reduced postprandial energy expenditure when eating occurs at night [5]. In parallel, trials in non-shift populations have repeatedly found that intermittent energy restriction and continuous energy restriction produce comparable weight loss. A 12-month noninferiority trial in women with prior gestational diabetes reported a between-group difference of −1.6 kg (95% CI −4.2 to 1.0) and no differences in HbA1c, fasting glucose, insulin, or HOMA-IR [4]. A 24-month follow-up of a type 2 diabetes trial found weight loss maintained at about −3.9 kg in both groups, with a between-group difference of 0.07 kg (90% CI −2.5 to 2.6) [7]. Reviews of randomized trials of at least 8 weeks similarly concluded that intermittent paradigms produce equivalent weight loss to continuous restriction, with 9 of 11 studies showing no difference [8]. Short-term mechanistic work has hinted at subtle differences — one 4-week trial found fasting glucose fell after continuous but not intermittent restriction, while non-esterified fatty acids were lower after intermittent restriction [6] — but these did not translate into clear clinical separation.
The gap the SWIFt team set out to fill was therefore specific: whether a modified fasting schedule aligned to shift timing could improve on continuous restriction in a population whose circadian system is chronically disrupted, and whether any advantage would persist after structured support ended [1,7].
The SWIFt 18-month results: three strategies, one trajectory
SWIFt randomized 250 night-shift workers with overweight or obesity 1:1:1 to 24 weeks of moderate 20% continuous energy restriction (CER), modified intermittent fasting on two days per week on days off or day shifts (IF:2D), or the same two-day fasting pattern timed to night shifts (IF:2N) [1,7]. The primary 24-week report found no statistically significant or clinically important between-group differences in weight or HOMA-IR, with maximum between-group mean differences of 0.6 kg for weight and 0.3 units for HOMA-IR, although the IF:2D group showed greater reductions in total and LDL cholesterol [5]. The new 18-month secondary analysis assessed outcomes 12 months after the supported intervention ended, with only three tele-health dietetic reviews and withdrawal of study foods during that follow-up year [1,2]. At 18 months, 141 participants provided data. Within-group weight remained below baseline in all arms: CER −4.8 kg (95% CI −6.1 to −3.1), IF:2D −4.5 kg (−6.3 to −2.8), and IF:2N −3.2 kg (−5.1 to −1.04) [1]. Between-group differences were not apparent: IF:2D versus CER was 2.7 kg (95% CI −3.6 to 9.2) and IF:2N versus CER was 0.9 kg (−5.6 to 7.3). HOMA-IR differences were similarly null, at 0.7 (95% CI −0.3 to 1.7) for both comparisons, and no between-group differences appeared in other cardiometabolic measures [1].
The authors are explicit that the study may have insufficient power to detect small but clinically meaningful effects, and the wide confidence intervals support that caution [1]. The 18-month data therefore indicate no apparent difference between strategies rather than proving equivalence.
How SWIFt fits the comparator landscape
The 18-month SWIFt findings align closely with the two prior extended intermittent-fasting comparisons the authors identify. In 112 participants randomized to modified fasting or continuous restriction for six months followed by a six-month maintenance period, there was no difference in weight loss at one year and no between-group differences in cardiometabolic risk factors [2]. In 137 participants with type 2 diabetes who followed either a 12-month modified fasting or continuous restriction intervention, weight loss maintained after a further 12-month follow-up did not differ significantly between groups, although participants regained about 33% of lost weight and glycaemic control regressed [2,9]. The gestational diabetes trial similarly found comparable 12-month weight loss and no differences in diabetes risk markers, with a between-group difference of −1.6 kg (95% CI −4.2 to 1.0) [4]. Reviews of shorter randomized trials reached the same overall conclusion [8]. What SWIFt adds is the shift-work context: the largest study to date examining weight-loss strategies specifically in night-shift workers, with a 24-week intervention and an 18-month endpoint, and with fasting days deliberately scheduled around shift timing [1,7]. The absence of a fasting-timing advantage is notable because it tests the circadian hypothesis directly in the population most likely to benefit from it.
One signal complicates a purely null reading. At 18 months, participants allocated to IF:2D self-reported consuming on average 938 kJ per day more than those allocated to CER after adjusting for baseline energy intake [2]. The authors flag this as a possible sustainability issue and call for future work on mechanisms [2]. It is a self-reported, adjusted finding and should be treated as hypothesis-generating rather than as an explanation for the null between-group result.
Where the conclusion stops
The most important boundary is statistical. With 141 of 250 participants providing 18-month data, the trial was not powered to detect small differences, and the confidence intervals around the between-group estimates span clinically relevant magnitudes in both directions [1]. A null result here cannot be read as equivalence. The population is also specific: adults aged 25–65 with overweight or obesity working at least two nights per week, recruited in Melbourne and Adelaide, excluding people with diabetes, cardiovascular disease, inflammatory bowel disease, or medications affecting body composition or metabolism [2,7]. Findings should not be generalized to night-shift workers with established type 2 diabetes or cardiovascular disease. The follow-up period itself was deliberately low-support, which is a strength for assessing maintenance but means the trial does not test whether continued structured support would have preserved a between-group difference. Attrition was similar across groups, which reduces differential-dropout concerns, but the overall retention limits precision [1]. Finally, the trial was stopped before meeting its pre-specified sample size because of COVID-19-related recruitment constraints, a limitation carried forward from the 24-week report [5].
For clinicians, the defensible reading is that either approach can produce clinically relevant weight loss in night-shift workers who stay engaged, and that the choice can reasonably be guided by patient preference and schedule fit rather than by an expectation of metabolic superiority for fasting [1,2]. The unresolved question is whether a larger trial with adequate power would detect a small advantage for one strategy, and whether the higher self-reported energy intake in the IF:2D arm at 18 months reflects a real sustainability penalty [2].
About These Sources
This research page is built on 8 peer-reviewed studies — published from 2019 to 2026, 5 from 2024 or later — selected as the most relevant from 12 studies that passed quality screening, drawn from 68 papers retrieved from a database of over 500 million.
Sources used in this answer
Weight loss maintenance one year after modified intermittent fasting in circadian-disrupted night shift workers: longer-term secondary outcomes from the SWIFt randomised clinical trial
The anchor SWIFt secondary-outcome paper reports that at 18 months, modified intermittent fasting and continuous energy restriction did not differ on body weight or HOMA-IR in 250 night-shift workers, with wide confidence intervals and explicit acknowledgment of possible insufficient power [1].
… one year after modified intermittent fasting in circadian-disrupted night shift workers: longer-term secondary outcomes from the SWIFt randomised clinical trial
The full-text SWIFt report supplies the trial design, eligibility criteria, follow-up structure, and the observation that the IF:2D group self-reported about 938 kJ/day more energy intake than CER at 18 months after baseline adjustment [2].
Long-term night shift work and cardiometabolic health in healthcare workers: a review of mechanisms and precision intervention strategies
This review of night-shift cardiometabolic mechanisms establishes circadian misalignment as the core driver of shift-work metabolic harm and documents a dose-response relationship between night-shift exposure and cardiometabolic risk [3].
The effect of intermittent energy restriction on weight loss and diabetes risk markers in women with a history of gestational diabetes: a 12-month randomized control trial.
This 12-month randomized trial in women with prior gestational diabetes found intermittent and continuous energy restriction produced comparable weight loss and no differences in diabetes risk markers, with a between-group difference of −1.6 kg (95% CI −4.2 to 1.0) [4].
Intermittent fasting for weight loss in night shift workers: a three-arm, superiority randomised clinical trial.
The primary 24-week SWIFt report found no significant or clinically important between-group differences in weight or HOMA-IR, with maximum mean differences of 0.6 kg and 0.3 units, and noted greater total and LDL cholesterol reductions in the IF:2D group [7].
Intermittent energy restriction is comparable to continuous energy restriction for cardiometabolic health in adults with central obesity: A randomized controlled trial; the Met-IER study.
This 4-week randomized trial in adults with central obesity found intermittent and continuous restriction comparable on most cardiometabolic markers, with subtle differences in fasting glucose and non-esterified fatty acids [8].
The effect of intermittent compared with continuous energy restriction on glycaemic control in patients with type 2 diabetes: 24-month follow-up of a randomised noninferiority trial.
This 24-month follow-up of a type 2 diabetes trial found weight loss maintained at about −3.9 kg in both intermittent and continuous restriction groups, with a between-group difference of 0.07 kg (90% CI −2.5 to 2.6), but HbA1c rose above baseline in both groups [9].
Effectiveness of Intermittent Fasting and Time-Restricted Feeding Compared to Continuous Energy Restriction for Weight Loss.
This review of randomized trials of at least 8 weeks concluded that intermittent energy restriction paradigms produce equivalent weight loss to continuous restriction, with 9 of 11 studies showing no difference [11].