Associations between multidimensional sleep health and inflammatory bowel disease onset and activity: A systematic review of the RU-SATED framework

Abstract

BACKGROUND

Inflammatory bowel disease (IBD) is a relapsing-remitting disorder affecting approximately 3.8 million people worldwide. Sleep disturbances are highly prevalent in patients with IBD, with a prevalence of up to 60%. Previous systematic reviews have shown that sleep quality is significantly poorer in patients with IBD than in healthy individuals and worsens during active disease. However, the associations of other sleep dimensions with IBD onset and activity status remain unclear.

AIM

To investigate the associations of sleep duration, efficiency, timing, regularity, and alertness with IBD onset and activity.

METHODS

The protocol was registered with PROSPERO (CRD420261301771). PubMed, Cochrane Library, EMBASE, and PsycINFO were searched from inception to January 16, 2026. Studies were grouped by sleep dimensions, assessment method (subjective or objective) and comparison type (IBD vs controls; active vs inactive). Risk of bias was assessed using the Newcastle-Ottawa Scale. A narrative synthesis was conducted following the Synthesis Without Meta-analysis guideline, and evidence certainty was graded as strong, moderate, limited, conflicting, or insufficient using the best evidence synthesis approach.

RESULTS

Of 644 initially identified records, 23 studies were ultimately included, including 17 cross-sectional, 4 cohort, and 2 case-control studies. These studies assessed sleep duration (n = 18), efficiency (n = 13), timing (n = 4), regularity (n = 2), and alertness (n = 4). Moderate-level evidence showed that sleep duration under 6 hours per day was associated with a higher risk of incident IBD (hazard ratios 1.17 to 1.36, 95%CI 1.05 to 1.59). Objectively measured sleep duration did not differ significantly between active and inactive status. Limited evidence indicated that objectively measured sleep duration and efficiency did not differ significantly between patients with IBD and healthy controls. Subjectively assessed sleep duration, sleep efficiency, and chronotype distribution did not differ significantly between active and inactive IBD, whereas patients with active disease exhibited more severe daytime sleepiness. For most sleep dimensions, the evidence remained limited or conflicting.

CONCLUSION

The association between sleep health and IBD is multidimensional and complex. Evidence for most sleep dimensions remains limited or conflicting, precluding definitive conclusions about the overall impact of sleep health on IBD. Large-scale, prospective studies are needed to clarify these associations and inform clinical management.

Key Words: Inflammatory bowel disease; Sleep health; Sleep duration; Sleep efficiency; Sleep timing; Sleep regularity; Alertness; Chronotype; Social jetlag; Sleep midpoint

Core Tip: Sleep disturbance is a highly prevalent comorbidity in inflammatory bowel disease that affects disease course. This study systematically evaluated multidimensional sleep indicators in relation to disease onset and activity of inflammatory bowel disease. It reveals that short sleep increases disease risk, yet objectively measured sleep duration does not differ between active and inactive stages. However, substantial methodological heterogeneity precludes definitive conclusions for most sleep dimensions. High-quality, standardized prospective studies are needed to clarify these associations and inform clinical management.

INTRODUCTION

Inflammatory bowel disease (IBD), encompassing Crohn’s disease (CD) and ulcerative colitis (UC), is a chronic relapsing–remitting disorder characterized by disruption of the intestinal mucosal barrier and persistent inflammation[1-3]. Its clinical presentation is complex and heterogeneous, with common symptoms including abdominal pain, hematochezia, weight loss, and fatigue[4]. The lifelong course of alternating relapse and remission profoundly affects patients' psychological well-being and quality of life[5-7]. The pathogenesis involves a complex interplay among gut microbiota, genetic susceptibility, environmental factors, and host immunity[8].

Sleep is a fundamental physiological process that maintains homeostasis and finely regulates immune and inflammatory balance through multiple pathways, including the gut microbiota and epigenetic modifications[9-12]. Insufficient or disrupted sleep activates the hypothalamic-pituitary-adrenal axis, alters gut microbiota composition, and induces systemic inflammation. Conversely, inflammation can reciprocally disturb sleep via neuroendocrine pathways, creating a vicious cycle[9,10]. In the context of IBD, a disease underpinned by chronic inflammation and immune dysregulation, sleep problems may play a critical role in its pathogenesis. The prevalence of sleep disturbance in patients with IBD reaches 60%, and their sleep quality is significantly poorer than that of healthy individuals[13,14]. Sleep quality further deteriorates during active status and represents an independent risk factor for relapse[15-17]. Therefore, sleep disturbance is not merely a prevalent comorbidity of IBD, but a key factor reflecting disease activity and influencing prognosis. Elucidating the relationship between sleep and IBD in greater depth holds significant clinical value for optimizing long-term patient management.

Achieving this goal requires a multidimensional and refined assessment of sleep. As sleep medicine has advanced towards precision in recent years, sleep research has expanded from global measures to a multidimensional sleep health construct. The RU-SATED framework proposed by Buysse[18] deconstructs sleep health into six core dimensions, namely regularity, satisfaction, alertness, timing, efficiency, and duration. However, previous systematic reviews in IBD have largely concentrated on sleep quality, which primarily corresponds to the satisfaction dimension[15-17]. The relationships between the remaining five sleep dimensions and IBD have yet to be systematically integrated and remain poorly understood. This evidence gap not only hinders clinicians from gaining a comprehensive understanding of the sleep characteristics of patients with IBD, but also precludes the development of targeted sleep management strategies.

The present systematic review addresses the remaining five sleep health dimensions within the RU-SATED framework and evaluates their associations with IBD onset and activity status, providing a broader evidence base for the sleep assessment and clinical management.

MATERIALS AND METHODS

This systematic review was conducted in accordance with the PRISMA guidelines and was registered with PROSPERO (CRD420261301771).

Data sources and search strategy

Relevant studies were identified through systematic searches of PubMed, Cochrane Library, EMBASE, and PsycINFO from database inception to January 16, 2026. The search strategy combined Medical Subject Headings and free-text terms related to sleep health and IBD, with no restrictions on language or publication date. The complete search string is provided in the Supplementary material.

Inclusion and exclusion criteria

Eligibility criteria were defined according to the PICOS framework: (1) Population: Patients with IBD (CD or UC), aged ≥ 18 years, irrespective of sex; (2) Intervention: Not applicable, as this review focused on observational studies rather than interventional designs; (3) Comparison: Two distinct comparisons were conducted. Comparison 1 involved IBD patients vs controls, with healthy individuals or patients with gastrointestinal diseases other than IBD serving as controls. Comparison 2 involved patients with active IBD vs those with inactive IBD; (4) Outcomes: At least one sleep health indicator measured by subjective questionnaire or objective device, including sleep duration, sleep efficiency, chronotype, sleep midpoint, social jetlag, sleep regularity, or alertness; and (5) Study design: Observational studies, including cross-sectional, case-control, and cohort studies. Animal experiments, reviews, abstracts without full text, and conference proceedings were excluded.

Study selection and data extraction

All retrieved records were imported into EndNote for deduplication. Two reviewers independently screened titles and abstracts, then reviewed the full texts of records that passed the initial screening. Reasons for exclusion were documented for all articles excluded at the full-text stage. Disagreements between the two reviewers were resolved through discussion, with a third reviewer adjudicating when consensus could not be reached.

Data extraction was performed independently by the two reviewers using a standardized, pilot-tested data extraction form. Extracted information included first author, year of publication, country, study design, IBD subtype, comparison, sample size, participant age, sleep measurement method, sleep dimensions, and key findings. The reviewers cross-checked the extracted results afterwards. When key statistics were missing, the corresponding authors were contacted by email. If no response was received, the study was retained in the narrative synthesis but excluded from quantitative pooling.

Risk of bias assessment

Study quality was appraised using tools adapted to each study design. Cohort and case-control studies were assessed with the corresponding versions of the Newcastle-Ottawa Scale (NOS)[19]. Cross-sectional studies were evaluated using the NOS adaptation validated by Herzog et al[20]. This version retains the three-domain structure but rephrases the items for single-time-point designs, namely selection (representativeness, sample-size justification, description of nonrespondents, ascertainment of exposure), comparability (confounding control), and outcome (validity of the outcome measure, appropriateness of the statistical test, response rate). Two reviewers independently performed the quality appraisal, and disagreements were resolved through discussion or adjudication by a third reviewer. Scores ≥ 7 indicated low risk of bias, scores 5-6 indicated moderate risk, and scores ≤ 4 indicated high risk.

Certainty of evidence assessment

Given the substantial variability in study designs and outcome measurement methods, the five-level best-evidence synthesis approach proposed by van Tulder et al[21] was adopted to narratively summarize the certainty of evidence. Evidence was graded as strong, moderate, limited, conflicting, or insufficient based on the number of studies, their methodological quality as determined by NOS categories, and the consistency of findings. Strong evidence was defined as at least two low-bias studies with consistent findings in ≥ 75% of studies. Moderate evidence requires one low-bias study combined with at least one moderate-bias study with consistent findings, or at least two moderate-bias studies with consistent findings. Limited evidence was defined as only one eligible study. Conflicting evidence was defined as at least two studies with consistent findings in < 75% of studies. Insufficient evidence was defined as no study meeting the inclusion criteria, or only one study with major methodological limitations. Because all included studies were observational rather than randomized controlled trials, a conservative one-level downgrade was applied to the original framework. For evidence bodies dominated by cross-sectional studies, a more cautious interpretation strategy was adopted. High heterogeneity across studies precluded a formal meta-analysis, and therefore a narrative synthesis was conducted in accordance with the Synthesis Without Meta-analysis (SWiM) reporting guideline[22]. Owing to the small number of studies per comparison group and the inability to pool data, publication bias was not assessed via funnel plots or Egger’s test.

RESULTS

Search results and study characteristics

The systematic database search yielded a total of 644 records, comprising 93 from PubMed, 167 from Cochrane, 211 from EMBASE, and 173 from PsycINFO. After title and abstract screening, 55 full-text articles were reviewed, and 23 studies ultimately met the eligibility criteria. The study selection process is illustrated in Figure 1.

Figure 1  PRISMA flow diagram.

These studies covered multiple sleep health dimensions, including sleep duration, sleep efficiency, sleep timing (chronotype, sleep midpoint, social jetlag), sleep regularity, and alertness. Sleep was assessed using both subjective and objective measures. Detailed study characteristics by dimension are presented in Table 1.

Table 1 Study characteristics.

Ref.	Country	Design	Comparison	Sample size (n)	Mean age (years, mean)	Sleep assessment	Sleep dimension	Disease activity criteria	Main findings
Barnes et al[33], 2024	Australia	Cross-sectional	IBD active vs IBD inactive	CD = 17, UC = 3	Total sample = 41 (median)	PSG	Duration, efficiency	CD: HBI (active > 4, remission ≤ 4); UC: SCCAI (remission ≤ 2, active ≥ 3)	Active IBD: Shorter TST (466.8 minutes vs 489.0 minutes, P < 0.05), lower SE (79.0% vs 87.9%, P < 0.05)
Ananthakrishnan et al[23], 2014	United States	Cohort	IBD vs controls	Controls = 151450, CD = 191, UC = 230	NR	Questionnaire	Duration	-	Short sleep (< 6 hours/day): UC, HR = 1.51 (1.10-2.09); > 9 hours/day, HR = 2.05 (1.44-2.92) (all P < 0.05); no CD association (all P > 0.05)
Bazin et al[24], 2020	France	Cross-sectional	CD active vs CD inactive	CD active = 14, CD inactive = 20	CD active = 41, CD inactive = 35 (median)	Wrist actigraphy, PSQI, ESS	Duration, efficiency, alertness	HBI (remission ≤ 4, active ≥ 5), CDAI (remission < 150)	Active CD: Lower SE (85.0% vs 90.5%, P < 0.01), shorter subjective sleep (6.43 hours vs 7.97 hours, P < 0.01), higher ESS (11.1 vs 6.8, P < 0.05)
Chakradeo et al[44], 2018	United States	Cross-sectional	IBD vs controls	Controls = 76, IBD = 115	Controls = 34.13, IBD = 41.40	MEQ, MCTQ, PSQI	Timing	-	IBD: Greater social jetlag (1.32 hours vs 1.05 hours, P < 0.05); CD: Higher rate of social jetlag > 2 hours (40% vs 16%, P < 0.05)
Chrobak et al[43], 2018	Poland	Cross-sectional	CD vs controls UC vs controls	Controls = 57, CD = 34, UC = 38	Controls = 41.4, CD = 35.8, UC = 42.6	CSM	Timing	-	CD: Lower CSM (36.00 vs 39.54, P < 0.05)
Conley et al[34], 2021	United States	Cross-sectional	IBD active vs IBD inactive	CD = 18, UC = 19	Total sample = 38	Wrist actigraphy	Duration, efficiency, regularity	Physician global assessment	No significant differences (all P > 0.05)
Hirten et al[35], 2026	United States	Cohort	IBD active vs IBD inactive	CD = 63, UC = 38	Total sample = 41.45	Wearable devices	Duration, efficiency	PRO-2; FC > 150/CRP > 5/ESR > 30	No significant differences (all P > 0.05)
Iskandar et al[36], 2020	United States	Cross-sectional	CD vs controls CD active vs CD inactive	Controls = 60, CD = 61	Controls = 32, CD = 31 (median)	Wrist actigraphy, ESS	Duration, efficiency, timing	HBI (active > 4, remission ≤ 4)	No significant differences in TST or SE; active CD: Higher daytime sleepiness (45.5% vs 19.2%, P < 0.05)
Salwen-Deremer et al[37], 2023	United States	Cross-sectional	IBD vs controls	Controls = 8, CD = 9, UC = 5	Total sample = 43.65	PSG	Duration, efficiency, alertness	-	No significant differences (all P > 0.05)
Kani et al[50], 2020	Turkey	Cross-sectional	IBD vs controls	Controls = 168, CD = 72, UC = 64	Controls = 36.54, CD = 37.13, UC = 41.70	ESS	Alertness	-	IBD: Lower ESS than controls (3.14 vs 6.99, P < 0.001)
Keefer et al[25], 2006	United States	Cross-sectional	IBD vs controls	Controls = 7, CD = 8, UC = 8	Controls = 34.29, IBD = 41.44	PSG, wrist actigraphy, PSQI, ESS	Duration, efficiency, alertness	-	No significant differences in SE, TST, or ESS (all P > 0.05)
Kamp et al[26], 2023	United States	Cross-sectional	IBD active vs IBD inactive	CD = 21, UC = 7	Total sample = 33.4	Wrist actigraphy, PSQI	Duration, efficiency	CD: SES-CD (0-2 remission, 3-6 mild, 7-15 moderate, > 15 severe), HBI (remission ≤ 4, active > 4), UC: Mayo endoscopic score (0-1 remission, 2-3 active), SCCAI (remission ≤ 2, active ≥ 3)	No significant differences in TST or SE (all P > 0.05)
Şahbaz and Keskin[27], 2020	Turkey	Cross-sectional	IBD vs controls IBD active vs IBD inactive	Controls = 44, CD = 41, UC = 49	Controls = 40.3, CD = 33.7, UC = 40 (median)	MEQ, PSQI	Duration, efficiency, timing	NR	No significant differences in TST or SE (all P > 0.05); IBD: Higher eveningness prevalence (CD 12.2%, UC 18.4% vs controls 0%, P < 0.001)
Mokros et al[42], 2021	Poland	Cross-sectional	CD active vs CD inactive UC active vs UC inactive	CD = 47, UC = 37	CD = 37.3, UC = 46.0	CQ	Timing	CD: CDAI (< 150 remission, ≥ 150 active); UC: Mayo Score (≤ 2 remission, ≥ 3 active)	No significant correlation between chronotype and disease activity (all P > 0.05)
Paixão et al[38], 2019	Brazil	Cross-sectional	IBD active vs IBD inactive	CD = 9, UC = 11	CD = 30.89, UC = 46.36	PSG	Duration, efficiency	CDAI (≤ 150 remission, > 150 active)	No significant differences in TST or SE (all P > 0.05)
Qualqili et al[28], 2021	Jordan	Case control	IBD vs controls	Controls = 150, CD = 85, UC = 100	Controls = 41.4, CD = 41.2, UC = 39.8	Questionnaire	Duration	-	IBD patients had higher proportion of moderate sleep (6-8 hours) (P < 0.001)
Qazi et al[39], 2019	United States	Cross-sectional	CD active vs CD inactive	CD active = 44, CD inactive = 28	CD active = 34.5, CD inactive = 38 (median)	Wrist actigraphy, ESS	Duration, efficiency, alertness	HBI (remission ≤ 4)	Remission CD: Higher SE (89.9% vs 86.6%, P < 0.05)
Swanson et al[48], 2021	United States	Cross-sectional	IBD active vs IBD inactive vs controls	Controls = 10, CD active = 12, CD inactive = 8, UC active = 10, UC inactive = 12	Controls = 34.7, CD active = 33.5, CD inactive = 40.7, UC active = 38.3, UC inactive = 41.0	Wrist actigraphy	Regularity	HBI (remission ≤ 4)	Aggressive IBD: Lower IS (0.39 ± 0.15 vs 0.51 ± 0.10 non-aggressive and 0.55 ± 0.09 controls, P < 0.05); higher IV (vs non-aggressive, P < 0.05)
Chaemsupaphan et al[40], 2025	Thailand	Cohort	IBD active vs IBD inactive	CD = 48, UC = 50	Total sample = 46.2	Wrist actigraphy	Duration, efficiency	CD: HBI (remission ≤ 4, active > 4), SES-CD (0-2 remission); UC: Mayo score (≤ 2 remission, ≥ 3 active), MES (0-1 remission)	No significant differences in TST or SE (all P > 0.05)
Wu et al[29], 2020	China	Case control	CD vs controls UC vs controls	Controls = 671, CD = 380, UC = 188	Controls = 32, CD = 32, UC = 37 (median)	Questionnaire	Duration	-	CD: No significant differences in TST (P > 0.05), UC: Shorter weekend sleep (7 hours vs 8 hours, P = 0.05)
Yuan et al[30], 2023	China	Cohort	IBD vs controls	Controls = 485722, CD = 806, UC = 1798	Controls = 56.5, CD = 57.1, UC = 57.4	Questionnaire	Duration	-	Short sleep (≤ 5 hours/day): IBD, HR = 1.36 (1.17-1.59); CD, HR = 1.53 (1.17-2.00); UC, HR = 1.29 (1.07-1.56). 6 hours/day: IBD, HR = 1.17 (1.05-1.30); CD, HR = 1.23 (1.02-1.48); UC, HR = 1.14 (1.01-1.30) (all P < 0.05)
Zhang et al[31], 2019	China	Cross-sectional	IBD vs controls IBD active vs IBD inactive	Controls = 120, CD = 39, UC = 81	Controls = 36.28, IBD = 36.01	PSG, PSQI	Duration, efficiency	CD: HBI (≤ 4 remission, > 4 active); UC: Mayo endoscopic score (0-1 remission, 2-3 active)	IBD: Lower TST and SE (378 minutes vs 431 minutes; 76.45% vs 85.45%, P < 0.001); active IBD: Lower SE (64.88% vs 73.20%, P < 0.001)
Zhang et al[32], 2022	China	Cross-sectional	IBD vs controls	Controls = 4300, IBD = 56	Controls = 44.0, IBD = 51.8	Questionnaire	Duration	-	IBD: Shorter sleep (6.3 hours vs 6.8 hours, P < 0.01)

IBD: Inflammatory bowel diseases; CD: Crohn’s disease; UC: Ulcerative colitis; PSG: Polysomnography; HBI: Harvey-Bradshaw Index; SCCAI: Simple Clinical Colitis Activity Index; TST: Total sleep time; SE: Sleep efficiency; NR: Not reported; HR: Hazard ratio; PSQI: Pittsburgh Sleep Quality Index; ESS: Epworth Sleepiness Scale; CDAI: Crohn’s Disease Activity Index; MEQ: Morningness-eveningness Questionnaire; MCTQ: Munich ChronoType Questionnaire; CSM: Composite Scale of Morningness; PRO-2: Patient-Reported Outcome-2; FC: Fecal calprotectin; CRP: C-reactive protein; ESR: Erythrocyte sedimentation rate; SES-CD: Simple Endoscopic Score for Crohn’s Disease; IS: Interdaily stability; IV: Intradaily variability; MES: Mayo Endoscopic Score; CQ: Chronotype Questionnaire.

Heterogeneity characteristics and synthesis feasibility analysis

This review used narrative synthesis rather than quantitative pooling for the following reasons. First, the sleep outcome indicators exhibited high methodological heterogeneity, and effect-size reporting was incomplete. Studies differed substantially in measurement instruments and variable data types. For sleep duration, some studies reported continuous variables from objective devices, others used self-reported categorical variables, and still others reported continuous variables based on the sleep duration component of the Pittsburgh Sleep Quality Index. These differences made it impossible to directly unify or convert across data types. Furthermore, most studies did not simultaneously provide the means and standard deviations required for meta-analysis, and some reported only means or medians. Although the total number of included studies was acceptable for certain sleep indicators, such as sleep duration and sleep efficiency, subgroup analyses by measurement instrument or variable type would involve too few studies per subgroup to provide reliable statistical power.

Second, the definition of IBD disease activity status was inconsistent across studies. Researchers defined disease status using inflammatory biomarkers (C-reactive protein, fecal calprotectin, erythrocyte sedimentation rate), clinical activity indices (Harvey-Bradshaw Index, CD Activity Index, Mayo Score), endoscopic scores, or comprehensive clinical assessments. Even for the same disease, different studies used different clinical scales (Harvey-Bradshaw Index and CD Activity Index). Because patients classified as having active disease under different criteria may differ in inflammatory burden and clinical symptoms, combining these non-equivalent groups would introduce clinical heterogeneity and compromise comparability.

Therefore, a structured narrative synthesis was the most appropriate approach for presenting the current evidence, interpreting differences between studies, and identifying knowledge gaps. This review adopted a structured narrative integration method in accordance with the SWiM guideline to summarize findings across the individual sleep health dimensions. The evidence summary is presented in Table 2.

Table 2 Evidence summary.

Indicator	Comparison	Measure type	Number of studies			Direction		Risk of bias		Level of evidence	Summary
			Cohort	Case-control	Cross-sectional	Positive	Negative	Low	Moderate
Sleep duration	Risk of developing IBD	Subjective	2	-	-	2	-	2	-	Moderate	Short sleep duration (≤ 5-6 hours/day) is associated with a significantly increased risk of developing IBD
Sleep duration	IBD vs controls	Subjective	-	2	4	3	3	3	3	Conflicting
Sleep duration	IBD vs controls	Objective	-	-	5	1	4	2	3	Limited	No significant difference in sleep duration between IBD patients and healthy controls
Sleep duration	IBD active vs IBD inactive	Subjective	-	-	4	1	3	1	3	Limited	No significant difference in sleep duration between active disease and remission
Sleep duration	IBD active vs IBD inactive	Objective	2	-	7	1	8	3	6	Moderate	No significant difference in sleep duration between active disease and remission
Sleep efficiency	IBD vs controls	Subjective	-	-	3	1	2	1	2	Conflicting
Sleep efficiency	IBD vs controls	Objective	-	-	4	1	3	2	2	Limited	No significant difference in sleep efficiency between IBD patients and healthy controls
Sleep efficiency	IBD active vs IBD inactive	Subjective	-	-	4	1	3	1	3	Limited	No significant difference in sleep efficiency between active disease and remission
Sleep efficiency	IBD active vs IBD inactive	Objective	2	-	8	4	6	4	6	Conflicting
Chronotype	IBD vs controls	Subjective	-	-	3	2	1	2	1	Conflicting
Chronotype	IBD active vs IBD inactive	Subjective	-	-	2	-	2	-	2	Limited	No significant association between chronotype and disease activity
Midpoint of sleep	IBD vs controls	Subjective	-	-	1	1	-	1	-	Insufficient
Social jetlag	IBD vs controls	Subjective	-	-	-	1	-	1	-	Insufficient
Social jetlag	IBD active vs IBD inactive	Subjective	-	-	-	1	-	1	-	Insufficient
Sleep regularity	IBD active vs IBD inactive	Objective	-	-	2	1	1	1	1	Conflicting
Alertness	IBD vs controls	Subjective	-	-	3	1	2	2	1	Conflicting
Alertness	IBD active vs IBD inactive	Subjective	-	-	2	2	-	1	1	Limited	Patients with active IBD exhibit significantly higher daytime sleepiness than those in remission

IBD: Inflammatory bowel diseases.

Sleep duration

A total of 18 studies examined the association between sleep duration and IBD. Of these, 10 employed subjective measures, including self-report and the Pittsburgh Sleep Quality Index questionnaire[23-32], and 12 utilized objective measures, including polysomnography (PSG), wrist actigraphy, and wearable devices[24-26,31,33-40].

Regarding disease onset risk, two large-scale prospective cohort studies using questionnaire-based assessments consistently showed that short sleep was associated with an increased risk of developing IBD[23,30]. Both studies were rated as having a low risk of bias, and their findings were directionally consistent. The evidence was therefore graded as moderate. Specifically, compared with sleeping 7 hours to 8 hours per day, individuals sleeping fewer than 6 hours per day had a significantly elevated risk of developing IBD, with hazard ratios ranging from 1.17 to 1.36 (95%CI 1.05 to 1.59)[30]. This association was particularly pronounced among women with UC, for whom short sleep was associated with a significantly increased risk of UC (hazard ratios 1.51, 95%CI 1.10 to 2.09)[23]. However, because only two studies were available, the robustness of this association warrants confirmation in future independent cohort studies.

In comparison with healthy controls, six studies used subjective self-report measures. Three studies found that patients with IBD had significantly shorter sleep duration than healthy individuals[28,31,32], whereas the remaining three observed no significant difference[25,27,29]. The evidence was therefore graded as conflicting. Moreover, among five cross-sectional studies employing objective measurement, only one reported significantly shorter total sleep time in patients with IBD than in healthy controls[31], and the other four all reported no significant difference[25,36-38]. Although the direction of findings was consistent, three of those four studies had fewer than 30 participants[25,36,38], limiting their representativeness and resulting in a moderate risk of bias rating. Therefore, the evidence was graded as limited.

When comparing active IBD with inactive IBD, the four studies based on subjective self-reports yielded mixed results. One study reported that subjectively perceived sleep duration was significantly shorter in patients with active CD compared to those in remission[24], whereas the remaining three observed no significant difference[26,27,31]. The evidence was graded as limited, given the small number of studies and the fact that two included fewer than 40 participants[24,26], resulting in a moderate risk of bias. In contrast, objective measurements provided a clearer pattern. Among the nine studies employing objective assessment, eight reported no significant difference in sleep duration between active and inactive IBD[24,26,31,33-36,39,40]. Owing to the relatively large number of studies and the highly convergent direction of findings, the evidence was graded as moderate that sleep duration does not differ significantly between active and inactive IBD.

Sleep efficiency

Sleep efficiency is defined as the ratio of total sleep time to time spent in bed[41]. Thirteen studies examined the association between sleep efficiency and IBD[24-27,31,33-40]. Both subjective and objective assessments yielded conflicting findings.

In comparison with healthy controls, three studies used subjective assessment. Only one reported significantly lower sleep efficiency in patients with IBD than in healthy controls[31], and the other two observed no significant difference[25,27]. The evidence was graded as conflicting. Among the four studies employing objective measurement, only one reported significantly lower sleep efficiency in patients with IBD compared with healthy controls[31]. Although the direction of findings was consistent, two studies had fewer than 30 participants and were at moderate risk of bias[25,37]. Therefore, the evidence was graded as limited.

In comparisons of active vs inactive IBD, four studies subjectively assessed sleep efficiency. Only one observed significantly lower efficiency in patients with active disease compared with those in remission[24], indicating a generally consistent direction of findings. However, this study had fewer than 40 participants and limited representativeness, so the evidence was graded as limited. Among the ten studies employing objective measurement, six reported no significant difference in sleep efficiency between active and inactive IBD[26,34-36,38,40], whereas four observed reduced efficiency in the active status[24,31,33,39]. The direction of findings was inconsistent, and the evidence was therefore graded as conflicting.

Sleep timing

Chronotype: Chronotype reflects an individual’s personal circadian preference for activity-rest and sleep-wake patterns, commonly categorized as morning, evening, or intermediate types[27,42]. Four studies examined the association between chronotype and IBD, all using questionnaires such as the Morningness-Eveningness Questionnaire, the Composite Scale of Morningness, and the Chronotype Questionnaire[27,42-44].

When comparing IBD patients with healthy controls, two studies reported a significantly higher proportion of evening types among patients with IBD[27,43], whereas one study found no significant difference[44]. The evidence was therefore graded as conflicting. Regarding disease activity, both studies that examined this association reported no significant relationship between chronotype preference and disease activity[27,42]. Although the direction of findings was highly consistent, both studies were at moderate risk of bias and employed different questionnaires. The evidence was therefore graded as limited.

Sleep midpoint and social jetlag: Sleep midpoint is the midpoint between sleep onset and sleep offset, and serves as a core indicator of circadian phase[45]. Social jetlag refers to the discrepancy between sleep midpoints on workdays and free days, reflecting the misalignment between an individual’s endogenous biological clock and social clock[46]. Only one cross-sectional study with a low risk of bias investigated these parameters using the Munich ChronoType Questionnaire. It showed that patients with IBD had a significantly earlier sleep midpoint and greater social jetlag than healthy controls. It also found that the proportion of patients with social jetlag exceeding 2 hours was significantly larger among those with aggressive CD compared to those with non-aggressive CD[44]. Because this is an isolated single study, the evidence for both indicators was graded as insufficient.

Sleep regularity

Sleep regularity refers to the day-to-day stability and consistency of an individual’s sleep-wake patterns over multiple days, reflecting the robustness of the circadian system[47]. The evidence in this field was preliminary. Only two studies employed wrist actigraphy to examine the association between sleep regularity and IBD disease activity, using interdaily stability (IS) and intradaily variability (IV) as core indicators[34,48]. Specifically, one study reported that patients with aggressive IBD exhibited significantly lower IS and higher IV than those with non-aggressive IBD, suggesting poorer sleep regularity in aggressive disease[48]. Conversely, the other study found no significant differences in either IS or IV between patients with active and inactive IBD[34]. Given the inconsistency between the two studies, the evidence was graded as conflicting.

Alertness

Alertness refers to the capacity to sustain attention, reaction speed, and cognitive readiness during wakefulness, reflecting an individual’s daytime functioning. Low alertness manifests as sleepiness and sluggish responsiveness, and is commonly assessed indirectly through the measurement of daytime sleepiness[49]. Four studies examined the association between alertness and IBD, all using the Epworth Sleepiness Scale (ESS) to measure daytime sleepiness[24,25,36,50]. An ESS score above 10 indicated excessive daytime sleepiness.

Among the three cross-sectional studies comparing patients with IBD with healthy controls, only one reported that patients with CD had significantly lower daytime sleepiness[50], whereas the remaining two found no significant difference[25,36]. Regarding disease activity, both cross-sectional studies found significantly higher ESS scores or a greater proportion of daytime sleepiness in patients with active CD than in those in remission[24,36]. The findings were consistent in direction. However, given that the studies are cross-sectional in design and carry low and moderate risk of bias, respectively, the evidence was graded as limited.

Risk of bias summary

A summary of the risk of bias assessment for the included studies is presented in Table 3, with detailed results in Supplementary Tables 1 and 2. Four cohort studies and two case-control studies were assessed using the corresponding NOS versions, each with a maximum score of 9. The risk of bias in these studies was concentrated in the moderate-to-low range (7.50 ± 1.22). Inadequate sample representativeness, exposure recall bias, and attrition bias were the most common weaknesses and the main reasons for downgrading to moderate risk.

Table 3 Newcastle-Ottawa Scale ratings summary, n (%)/mean ± SD.

Design	n	Low	Moderate	High	Median Newcastle-Ottawa Scale
Cohort/case control	6	4 (67)	2 (33)	0 (0)	7.50 ± 1.22
Cross-sectional	17	7 (41)	10 (59)	0 (0)	6.29 ± 1.05

Low ≥ 7; moderate = 5-6; high ≤ 4. Scores reflect reconciled, post-agreement ratings.

Seventeen cross-sectional studies were appraised using the adapted NOS validated by Herzog et al[20], with a maximum total score of 10. The risk of bias was concentrated in the low-to-moderate range (6.29 ± 1.05). Nearly all studies lacked formal sample-size justification and a description of non-respondents. Studies rated as low risk typically used validated measurement instruments and adequately controlled for key confounders. Those rated as moderate risk were constrained by small sample sizes, insufficient confounding control, or reliance on self-reported outcomes. The uneven distribution of bias risk restricted the strength of evidence ratings assigned to certain dimensions, including sleep duration, chronotype, and alertness.

DISCUSSION

This systematic review is the first to evaluate the relationship between multidimensional sleep health and IBD. The findings indicate that the relationship between sleep health and IBD is complex and multidimensional. Across most sleep dimensions, the existing evidence remains conflicting or insufficient, which currently precludes definitive conclusions.

Among the various dimensions, sleep duration was the most extensively studied, likely because it is readily quantifiable and a primary focus for both clinicians and patients. This review found that sleeping fewer than 6 hours per day was significantly associated with an increased risk of developing IBD, particularly among women with UC. However, a Mendelian randomization study based on genetic instruments reported no causal association between sleep duration and IBD[51]. This discrepancy suggests that sleep duration may not be an independent causal factor but instead influences IBD risk indirectly through interactions with other environmental or behavioral factors. Specifically, short sleepers often concurrently have unhealthy dietary habits[52,53], insufficient physical activity[54,55], chronic psychological stress[56,57], and other adverse lifestyle factors. The metabolic disturbances and chronic low-grade inflammation arising from these behaviors may represent the true underlying causal factors. Further high-quality prospective cohort and interventional studies are needed to clarify the causal direction and the clinical value of sleep-targeted intervention.

Notably, this review identified a dissociation between subjectively reported and objectively measured sleep outcomes in patients with IBD. Two studies using concurrent subjective and objective measures found that patients with IBD subjectively reported significantly shorter sleep duration than healthy controls, or shorter sleep during the active stage, yet objective measures detected no significant abnormalities[24,25]. A previous systematic review focusing on sleep quality in IBD similarly documented this phenomenon[16]. Such inconsistencies may originate from differences in measurement instruments and the underlying pathophysiological mechanisms. Subjective questionnaires reflect an individual’s perceived sleep experience, whereas objective devices precisely record physiological parameters. Actigraphy infers sleep-wake states based on wrist movement and is prone to misclassifying motionless wakeful periods in bed as sleep[58]. Some patients may prolong their time in bed due to somatic discomfort, which could lead to actigraphy-recorded total sleep time showing no difference from healthy controls, thereby masking true sleep insufficiency. Patients with IBD commonly experience chronic somatic discomfort, such as abdominal pain and bloating. Although often intermittent or mild to moderate and insufficient, these symptoms interfere with the subjective sleep experience and lead patients to perceive sleep as severely impaired, even when objective parameters are unaffected[59,60]. Concurrently, patients with IBD frequently present with psychological comorbidities, particularly anxiety and depression, which can amplify negative sleep perceptions[61,62]. Furthermore, pro-inflammatory cytokines such as interleukin-1β and tumor necrosis factor-α, released during intestinal inflammation, can act on the central nervous system via vagal nerve signaling and humoral circulation across the blood-brain barrier[9]. This disrupts sleep architecture, resulting in loss of deep sleep and sleep fragmentation. This disruption of sleep architecture preserves the total objective sleep duration while leaving patients with a poor subjective sleep experience. These findings suggest that, in the clinical management of IBD, a sole focus on objective sleep parameters may underestimate the true burden experienced by patients, and that subjective sleep complaints carry equally important clinical value.

The highly contradictory objective findings regarding sleep efficiency may originate from systematic differences between measurement instruments. As the gold standard for sleep assessment, PSG provides multidimensional objective data, including electroencephalography, electrooculography, respiration, limb movements, and heart rate, thereby supporting sleep staging and the diagnosis of sleep disorders[63]. Actigraphy infers sleep-wake states through accelerometer-based monitoring of limb movement and differs from PSG in sensitivity and algorithmic settings, which may lead to measurement bias. The mean discrepancy in sleep efficiency between actigraphy and PSG reaches 7.46%, with actigraphy tending to overestimate sleep efficiency[64]. After one night of simultaneous PSG calibration, this discrepancy can be reduced to 5.18%[64]. Wearable devices equipped with both accelerometers and photoplethysmography sensors can collect limb movement and heart rate data simultaneously, partially mitigating the inherent bias of actigraphy. However, the proprietary algorithms embedded in these devices still systematically overestimate sleep efficiency, and the closed nature of raw data prevents researchers from applying additional corrections[65]. Therefore, future studies should report PSG-calibrated sleep efficiency indices or adopt a multi-instrument joint assessment to improve both accuracy and comparability.

Research on sleep timing and regularity in the field of IBD remains in its infancy, yet these dimensions have shown notable potential. One study that simultaneously assessed chronotype, sleep midpoint, and social jetlag reported that although the chronotype distribution did not differ between patients with IBD and healthy individuals, sleep midpoint was significantly phase-advanced and social jetlag was significantly increased[44]. This suggests that shifts in actual sleep timing may carry greater biological relevance than sleep-wake preference alone, and that disruption of the circadian system may underlie these shifts. In recent years, clock genes have been found to be deeply involved in the development and progression of intestinal inflammation through the regulation of the intestinal barrier, immune responses, and other pathways. For instance, the relative expression of the peripheral clock gene CRY1 was shown to be significantly elevated in patients with UC[66]. The expression of the clock gene Bmal1 is negatively correlated with C-reactive protein and fecal calprotectin levels, and its downregulation can disrupt the intestinal barrier, induce epithelial cell apoptosis, disturb the rhythmic recruitment of immune cells, and exacerbate intestinal inflammation[67-70]. These findings position the circadian system as a potentially modifiable target, offering new avenues for the management of IBD.

Evidence regarding alertness is similarly limited at present. The two studies consistently showed that patients with active CD experienced significantly more severe daytime sleepiness than those in remission[24,36]. Notably, although patients with active CD perceived themselves as subjectively sleepy during the day, their objectively measured total sleep duration did not differ significantly from that of patients in remission. This suggests that the diminished daytime alertness in patients with IBD may not primarily originate from insufficient sleep duration, but rather from multiple factors converging at the level of subjective experience. Specifically, daytime sleepiness reflects an increased propensity to fall asleep or difficulty maintaining wakefulness, whereas fatigue denotes a sense of physical or mental exhaustion that is not necessarily accompanied by a tendency to fall asleep. The prevalence of fatigue among patients with active CD has been reported to reach 80%, significantly exceeding the rate of 44% in patients in remission[36]. This systemic fatigue is readily perceived by patients as daytime sleepiness[71]. Furthermore, the low mood, reduced energy, and psychomotor retardation characteristic of depression can exacerbate fatigue[72]. Therefore, clinical assessment should differentiate daytime sleepiness from fatigue and depression, and avoid simplistically attributing daytime sleepiness to insomnia. Intervention efforts should prioritize controlling intestinal inflammation and ameliorating fatigue, rather than relying solely on sedative-hypnotic medications. Future research should involve large-scale, multicenter studies to further disentangle the intrinsic relationships among disease activity, daytime sleepiness, and fatigue.

The strengths of this review include its novel application of the multidimensional sleep health framework to IBD, and its explicit distinction between subjective and objective assessment tools. A rigorous narrative synthesis was conducted in accordance with the SWiM guideline, avoiding inappropriate quantitative pooling of highly heterogeneous data. These findings carry implications for the clinical management of IBD. Sleep assessment and sleep hygiene education could be integrated into routine IBD follow-up. For patients presenting with poor sleep complaints, clinicians should attend to their subjective sleep experience while concurrently evaluating psychological comorbidities such as anxiety, depression, and fatigue. When necessary, multidisciplinary consultations involving gastroenterology, psychology, and sleep specialties should be initiated.

The evidence on multidimensional sleep health and IBD should be interpreted with caution, given the methodological differences and risk of bias across the included studies. These factors limit the reliability of the conclusions to some extent. Regarding disease activity, the criteria for classifying active disease and remission varied across studies, with a mix of clinical activity indices and endoscopic scores. Because patients in clinical remission may still show endoscopic activity, the actual inflammatory burden among patients classified as in remission differed across studies. Regarding confounding factors, the included studies generally provided insufficient control for key confounders, with comorbidities and medications as the primary sources of bias. Anxiety and depression can amplify negative sleep perceptions, whereas fatigue is easily confused with daytime sleepiness. For example, glucocorticoids are associated with side effects such as sleep disruption[73,74]. The use of anxiolytics or antidepressants, methotrexate, and infliximab has also been linked to an increased risk of poor sleep. Because these confounders were not adequately controlled in some of the original studies, the accuracy of the findings may have been affected.

Additionally, several limitations should be acknowledged. First, the evidence base remains relatively sparse. The included studies were predominantly cross-sectional in design, and for most dimensions, fewer than five studies were available, which limited the ability to draw causal inferences. Second, many studies published only as abstracts were excluded during fulltext screening. Although this ensured data completeness, it may have omitted some emerging research. Third, of the 23 included studies, five had fewer than 30 participants. This increased the risk of type II error and restricted the reliability of their findings. These limitations point to several directions for future research. A standardized sleep assessment framework for IBD should be established, incorporating PSG-calibrated wearable devices and actigraphy. Key confounding variables, including medication use and psychological comorbidities, should be systematically reported. In addition, large-scale, multicenter prospective cohort studies are needed to clarify the causal direction between multidimensional sleep health and the onset, relapse, and long-term prognosis of IBD. Finally, randomized controlled trials should be carried out to evaluate the effects of sleep interventions, including circadian-based therapies, on sleep and disease activity in patients with IBD.

CONCLUSION

The relationship between sleep health and IBD is multidimensional and complex, with the strength of evidence varying across dimensions. Moderate evidence indicates that short sleep is associated with an increased risk of developing IBD, whereas objectively measured sleep duration does not differ significantly between active and inactive stages. Evidence on sleep efficiency, timing, regularity, and alertness remains limited or conflicting, precluding definitive conclusions at present. Future research should prioritize standardizing measurement methods across sleep dimensions and conducting prospective cohort studies to establish causal directions between each sleep health dimension and IBD onset and activity, thereby providing a stronger evidence base for clinical management.