"When Your Stomach Keeps You from Sleeping": NHANES Analysis Shows Link Between Gastrointestinal Diseases and Sleep Disorders

Sleep problems and gastrointestinal (GI) disorders are two massive “invisible” burdens: they impair quality of life, increase chronic inflammation, and push people toward comorbidities. In a new study in BMC Gastroenterology, researchers used nationally representative data from the US (NHANES) and asked: is there a consistent statistical association between GI problems and sleep problems – and to what extent is this association mediated by depression? The answer was yes: people with a recent GI disorder were more likely to report “sleep problems,” physician-diagnosed “sleep disorders,” and slightly shorter sleep duration, and some of these associations did indeed pass through depressive symptoms.

Background of the study

Sleep disturbances and gastrointestinal complaints are two extremely common “invisible” burdens that reduce quality of life and are associated with an increased risk of chronic diseases. Growing evidence suggests that there is a bidirectional relationship between them: inflammation, visceral hypersensitivity, circadian disruptions, and the microbiota-gut-brain axis can simultaneously impact both the gastrointestinal tract and sleep. Recent review literature highlights the contribution of dysbiosis to affective symptoms and sleep regulation via cytokines, neurotransmitter systems, and microbiota metabolites, making the gut ↔ sleep link biologically plausible.

A separate piece of the puzzle is depression. It often coexists with both functional gastrointestinal disorders and insomnia, and observational studies increasingly show that depressive symptoms can be an intermediate link in associations between somatic complaints and sleep (up to “chain” mediations through somatic symptoms). Therefore, checking what part of the “gastrointestinal problems ↔ sleep disorders” connection goes through depression is not an academic exercise, but a step towards more precise clinical tactics.

Reliable assessment of such associations requires a large, representative sample with standardized sleep questions. The US National Health and Nutrition Examination Survey (NHANES) is valuable for this: starting with the 2005–2006 cycle, it introduced the SLQ module with standardized questions about sleep duration and whether the participant has been told by a physician that he or she has “sleep problems” or “sleep disorders.” These formulations are widely used in epidemiological studies as valid proxies for sleep outcomes, although they remain self-reported measures without polysomnography. On the GI side, NHANES includes a simple but reproducible question about a recent episode of “stomach or intestinal illness (vomiting/diarrhea),” a broad but useful indicator of recent GI load at the population level.

Finally, diet is a common modifiable factor for the GI tract, microbiota, and sleep, so proper accounting of diet is essential. For this purpose, NHANES analyses increasingly use DI-GM, a new “dietary index for microbiota” that reflects the degree of “friendliness” of the diet to the microbial community (links with microbiota diversity and metabolites have been confirmed). Even after adjusting for DI-GM and the traditional HEI-2015, associations between GI episodes and sleep problems can persist, which emphasizes that in addition to diet, other mechanisms are also at play - inflammation, mental health, and behavioral factors.

Who, how and what was measured

The authors analyzed NHANES 2005-2014: out of 50,965 participants, after standard exclusions (missing key data, oncopathology, etc.), the final sample included 10,626 adults. The presence of GI disease was determined by a simple questionnaire question: "In the past 30 days, have you had a stomach or intestinal disease with vomiting or diarrhea?" - the answer "yes" classified the person as GI. Sleep was described by three indicators: self-assessment of the average duration of sleep on weekdays; the answers "Has your doctor told you that you have problems with sleep?" and "Has your doctor told you that you have a sleep disorder?" Depression was assessed by the validated PHQ-9 scale; the threshold of ≥10 points was interpreted as clinically significant. The models consistently took into account dozens of covariates (age, gender, education and income, BMI, smoking/alcohol, hypertension, diabetes, physical activity, diet quality HEI-2015, the “diet usefulness index for microbiota” DI-GM, cardiac comorbidities, etc.).

Key Results

After full adjustment for confounding factors, people with a GI episode had a 70% higher odds of “sleep trouble” (adjusted OR = 1.70; 95% CI: 1.41-2.05) and an 80% higher odds of having a diagnosed sleep disorder (aOR = 1.80; 95% CI: 1.34-2.41). Their average sleep duration was shorter by about 0.15 h per night (β = −0.15; 95% CI: −0.29…−0.01). These associations persisted across subgroups: non-smokers, those without hypertension and diabetes, as well as those with coronary heart disease and even those with a more “friendly microbiota” according to the DI-GM index.

The Role of Depression as a "Bridge"

The authors then tested the mediating role of depression. It turned out that it explained ~21% of the overall GI ↔ “sleep troubles” association; ~19% for “sleep disorder”; and ~27% for sleep shortening. That is, depression is an important, but not the only, mediator of the “gut ↔ sleep” axis. The results were robust to bootstrap tests and sensitivity analyses.

Why the Gut "Interferes" with Sleep (and Vice Versa)

The authors discuss several biological and behavioral mechanisms. First, inflammatory cytokines (TNF-α, IL-1, IL-6), elevated in many GI conditions, themselves disrupt sleep architecture. Second, the microbiota-gut-brain axis: dysbiosis and microbiota metabolites affect circadian rhythms, serotonergic transmission, and stress responses, affecting both sleep and mood. Third, pain and visceral hypersensitivity maintain a vicious cycle: pain → anxiety and depressive symptoms → sleep fragmentation → increased pain/discomfort. Finally, behavioral factors (irregular meals, caffeine, low physical activity) add “noise,” which the authors tried to account for statistically.

What practical lessons can be learned right now?

The study is cross-sectional and does not prove causality, but it does encourage integrated patient management.

• For clinicians: If a patient with gastrointestinal complaints has poor sleep, check for depressive symptoms (PHQ-9/analog) and consider parallel interventions: psychoeducation, CBT-I (cognitive behavioral therapy for insomnia), stress management, nutritional interventions, and, if indicated, pharmacotherapy.
• For patients: signs of "GI ailments" in recent weeks + "poor sleep" - a reason to discuss both issues at the appointment, rather than treating only one. Sensible sleep hygiene, regular diet/exercise, and mood management are sensible first steps.
• For health policy: sleep and mental health programs should be linked to gastro-routes - this may be more effective than separate approaches.

Important details of the methodology

- In NHANES, "GI illness" was defined as self-report of GI illness in the past 30 days with vomiting/diarrhea - essentially a "broad net" that includes both acute infectious episodes and exacerbations of functional disorders. This is not a clinical diagnosis of IBS/GERD/IBD, and the authors explicitly attribute this approach to limitations.
- "Sleep disorders" were also defined by self-report "doctor told", without validation by polysomnography; sleep apnea could not be assessed separately due to data limitations. This could either under- or overestimate the precise estimates.
- The study is cross-sectional, so the direction of the arrow (GI → sleep or sleep → GI) cannot be determined; the authors emphasize the possibility of a two-way loop.

What is HEI-2015 and DI-GM - and what does microbiota have to do with it?

To more accurately account for dietary style, the models included the HEI-2015, an index of compliance with the US Dietary Guidelines, and DI-GM, a new “dietary index for gut microbiota” that summarizes the consumption of food groups associated in the literature with a favorable/unfavorable microbiota profile. DI-GM was validated in NHANES and correlates with markers of microbial diversity; it is now being widely tested in epidemiology. Importantly, even with a higher DI-GM, the GI ↔ sleep disturbance association remained, suggesting that a “good” diet alone may not be enough to protect against sleep problems in GI conditions.

Limitations and what's next

In addition to the points already mentioned (self-reporting, impossibility of causal inference, under-reported factors such as chronic pain or sleeping pills), the authors note the risk of misclassification and residual confounding. A logical next step would be longitudinal cohorts and intervention studies: for example, to test whether combined correction of gastrointestinal symptoms and depression reduces the risk of chronic insomnia; and whether “chrono-nutrition” strategies and a microbiome-oriented diet work as an adjuvant.

The main thing in three points

• In American adults, GI episodes are associated with a higher frequency of sleep problems and disorders and slightly shorter sleep; part of the association (~20-27%) is mediated by depression.
• Effects are consistent across subgroups and sensitive analyses, but the design is cross-sectional and GI conditions and sleep disorders are defined by self-report/clinician report.
• Nutrition (HEI-2015, DI-GM) is important but does not cancel out the GI ↔ sleep association; the optimal approach is an integrated one (GI + mental health + sleep behavioral factors).

Study source: Ye S., Sui L., Zeng X., et al. Association between gastrointestinal disorders and sleep-related problems: the mediating effect of depression. BMC Gastroenterology, August 19, 2025. DOI: https://doi.org/10.1186/s12876-025-04180-8