輪班工作與免疫介導疾病之相關性：系統性文獻回顧與統合分析

Abstract

輪班工作包含各種非正常工作時間的班表，包含夜班、傳統三班制以及不定期的班表等。輪班的工作型態會導致勞工日夜顛倒，造成人體生理時鐘紊亂以及晝夜節律失調。晝夜節律失調引起的荷爾蒙異常與睡眠障礙，可能是誘發免疫介導疾病的潛在途徑。本研究以系統性文獻回顧，探討班工作與免疫介導疾病之間的相關性。 本研究依照PRISMA指引與已註冊於PROSPERO的系統性回顧計劃書進行（註冊編號：CRD420251029228）。本回顧所納入的文獻中需符合以下條件：研究對象須為從事輪班工作，且之前未罹患免疫疾病的成年勞工；暴露組為從事任何類型的輪班工作的勞工，而對照組則為未曾從事輪班工作的一般工作者；研究結果則為新發生之免疫疾病。為了確保研究時序性，研究僅納入世代研究與病例對照研究。截至2025年8月31日為止，我們總共檢索了五個主要的數據庫與三個灰色文獻數據庫。我們採用 ROBINS-E 工具評估世代研究的偏誤風險，另根據Cochrane原則評估病例對照研究。本研究採用統合分析對具備相似設計與結果的文獻進行數據整合；針對無法進行統合分析的文獻，我們根據研究結果的效應方向，採用投票分類法統整每個疾病的效應方向。證據的準確程度則採用GRADE評分方法進行評估。 本研究最終納入了九篇研究，包含四篇世代研究與五篇病例對照研究，共涵蓋五種免疫疾病。多發性硬化症（n=4）中，我們針對其中三篇病例對照研究進行統合分析，發現輪班工作會增加疾病的發生風險（pooled OR= 1.17, 95% CI: 1.04–1.33）；另一篇世代研究亦指出從事輪班工作達20年或以上也會增加罹患疾病的風險（HR：2.62, 95% CI：1.06–6.46）。有一篇研究觀察到輪班工作會增加乾癬的發生風險（HR：1.19, 95% CI：1.07–1.32）。在探討類風濕性關節炎的研究（n=2），一篇芬蘭的研究指出，從事輪班工作的女性其致病風險增加（HR：1.36, 95% CI：1.02–1.82），而另外一篇研究則指出傳統三班制會增加類風濕性關節炎的發生風險。至於發炎性腸道疾病（n=1）與全身性紅斑狼瘡（n=1）的研究中觀察到的效應方向不明確，因此目前輪班工作和疾病風險之間的關係仍無定論。多數納入研究的偏誤風險為中度風險，多發性硬化症類別的偏誤風險則介於中度至高度之間。在證據評級的部分，多發性硬化症的準確度為「低」，而類風濕性關節炎則為「極低」。本文獻回顧結果顯示輪班工作可能與部分免疫疾病相關，但目前研究的結果並不一致，並且僅存在於特定的暴露族群；整體的證據等級偏低，目前尚不足以支持兩者之間存在明確的關聯。為了提高後續研究的外推性與準確程度，未來研究設計應著重於世代研究，並採用標準化的暴露定義，以利準確釐清暴露與疾病之間的時序性。

Shift work involves various nonstandard schedules, including night, rotating, split, and irregular shifts, and may disrupt circadian rhythms. This disruption can alter hormonal regulation and induce sleep disturbance, representing a potential pathway linking occupational exposure to immune-mediated diseases. This review aims to examine the association between shift work and immune-mediated diseases. Following PRISMA guidelines and the registered review protocol (PROSPERO registration number: CRD420251029228), we included cohort and case-control studies conducted among adult workers without prevalent immune-mediated diseases. Exposure consisted of shift work, including rotating night shift and permanent night shift. The comparison group comprised regular daytime workers. The outcomes were incident immune-mediated diseases. We searched five major databases and three grey literature databases up to August 31, 2025. Risk of bias was assessed via ROBINS-E for cohort studies and a Cochrane-based tool for case-control studies. We used a meta-analysis to quantitatively synthesize the results if feasible. For diseases without sufficient data, a vote-counting method was used to summarize the direction of effects. The certainty of evidence was evaluated using the GRADE approach. This review synthesized nine studies involving four cohorts and five case-control studies across five immune-mediated disease categories. For multiple sclerosis (n=4), our meta-analysis of three case-control studies showed an increased risk of multiple sclerosis among shift workers (pooled OR= 1.17, 95% CI: 1.04–1.33), while the remaining cohort also reported an increased risk among individuals with ≥20 years of rotating night shift work (HR: 2.62, 95% CI: 1.06–6.46). For psoriasis (n=1), shift work was associated with a positive direction of effect. For rheumatoid arthritis (n=2), a positive association was observed among females in a cohort study, and for ACPA-positive subtypes under specific shifts in a case-control study. Evidence for inflammatory bowel disease (n=1) and systemic lupus erythematosus (n=1) remained inconclusive with unclear directions of effect. Most included studies carried a moderate risk of bias, except for those on multiple sclerosis, which ranged from moderate to high. Consequently, the certainty of evidence was evaluated as low for multiple sclerosis and very low for rheumatoid arthritis due to risk of bias and inconsistent findings. This review suggests that shift work might be associated with certain immune-mediated diseases. However, the findings were inconsistent across studies and often limited to specific exposure groups, and the level of evidence was low. Consequently, current evidence remains insufficient to support an association between shift work and immune-mediated diseases. Future research should focus on cohort designs and using standardized exposure definitions, to ensure temporal relationships between exposure and disease onset to improve the interpretation of the findings.