Skip to main content
Tessa NewsLink
Paste a health news link, or browse

Source study found

Story checked

More REM sleep linked to lower risk of 83 diseases | ScienceDaily (opens in a new tab)

sciencedaily.com · 2026-09-23

Short answerEvidenceSource

Short answer

Mixed

Mixed.

One claim goes further than the study. 3 other points were not covered by the paper.

  • 2 supported
  • 1 overstated
  • 3 not covered

Checked against the study summary. The full text wasn't available, so some details couldn't be settled either way.

Share this check

Follow the evidence trail
1
2

NewsLink checks it

Mixed

One claim overstates the study. Two of six check out. Three claims the study doesn't address.

  • 2 supported
  • 1 overstated
  • 3 not covered
Open claim evidence
3
Then inspect each claim

Evidence layer

Claim by claim

Each claim gets a verdict. Expand it to see the evidence directly below.

6 claims in this story

Showing all 6 claimsChoose a verdict to focus the list.

Claim 1 of 6OverstatedLess than 5 hours of total sleep was associated with the highest disease risk, while the lowest risk for many conditions was concentrated in a 6-to-8-hour sleep window.View evidenceHide evidence

As statedincreased risk of 37 conditions; 6-to-8-hour window

Why this verdict

The 6-to-8-hour minimum-risk window is supported by the spline analyses, and the <5-hour group is supported as having the broadest pattern of adverse associations, accounting for 37 of 41 significant adverse associations versus the 6–8h reference. However, saying <5 hours had the 'highest disease risk' overstates the abstract evidence, which describes the most widespread adverse association pattern rather than uniformly highest or largest risk.

1

Study evidence

RCS analyses identified significant non-linear relationships between sleep duration and incident disease for 86 phenotypes (P for nonlinear < 0.05).

“Phenome-wide association analysis and restricted cubic spline (RCS) analyses were performed using Cox proportional hazard regression...”
2

Study evidence

Extreme short sleep (<5 hours) exhibited the most widespread adverse disease-association pattern versus the 6–8 hour reference.37 of 41 significant adverse associations (count) attributed to <5h group vs 6–8h reference

“In the category-specific analysis, individuals with extreme short sleep (<5 hours) exhibited the most widespread clinical vulnerabilities, accounting for 37 of the 41 identified significant adverse associations compared to the 6-8 hours reference group.”
Claim 2 of 6Not coveredGreater amounts of REM sleep were associated with lower risks of 83 diseases, including dementia, heart failure, and Parkinson’s disease.View evidenceHide evidence

As stated83 diseases; HR 0.74 for heart failure, 0.54 for dementia, 0.20 for Parkinson’s disease

Why this verdict

The abstract profile supports that greater REM sleep was associated with lower risks for 83 diseases. However, at abstract depth the specific named outcomes and effect estimates cited by the story—dementia, heart failure, Parkinson’s disease, and the stated HRs—are not enumerated or verifiable.

1

Study evidence

Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)

“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
Claim 3 of 6Not coveredGreater deep sleep was associated with lower risks of 7 conditions, including type 2 diabetes and major depressive disorder.View evidenceHide evidence

As stated7 conditions

Why this verdict

The abstract profile supports that higher deep sleep/N3 was associated with lower risk for 7 diseases. But the abstract-level evidence does not identify the specific diseases, so the story’s examples of type 2 diabetes and major depressive disorder are not verifiable at this depth.

1

Study evidence

Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)

“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
Claim 4 of 6Not coveredGreater sleep irregularity and wakefulness after sleep onset were each linked to higher risk of several conditions, including anxiety and substance use disorders.View evidenceHide evidence

As statedseveral conditions

Why this verdict

The abstract profile supports elevated-risk associations for greater sleep irregularity and increased WASO, with counts of 3 and 6 diseases respectively. But the specific examples named by the story, including anxiety and substance use disorders, are not provided in the abstract-level profile.

1

Study evidence

Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)

“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
Claim 5 of 6SupportedTracking sleep in more than 95,000 people revealed striking links between sleep patterns and long-term health.View evidenceHide evidence

As statedmore than 95,000 people

Why this verdict

The abstract-level profile supports a UK Biobank prospective cohort analysis of 95,559 participants using wrist accelerometer-derived sleep metrics, with median 8.9 years of follow-up, and reports broad associations between sleep patterns and incident disease. The headline wording is promotional ('striking') but remains framed as links rather than causation.

1

Study evidence

Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)

“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
Claim 6 of 6SupportedThe authors say the study cannot establish that sleep patterns directly cause disease risk.View evidenceHide evidence

Why this verdict

The paper profile explicitly identifies the study as observational and states that residual confounding and lack of causal inference are key limitations. The story’s caveat that the study cannot establish direct causation is supported.

1

Study evidence

Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)

“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
2

Study evidence

RCS analyses identified significant non-linear relationships between sleep duration and incident disease for 86 phenotypes (P for nonlinear < 0.05).

“Phenome-wide association analysis and restricted cubic spline (RCS) analyses were performed using Cox proportional hazard regression...”
Then look for missing context

Context layer

What the story left out

Important study details the story did not include.

  • Phenome-wide association analysis tested sleep metrics against 1,049 incident health outcomes using adjusted Cox proportional hazards models.

    The story conveys broad disease-association findings but does not reflect the full phenome-wide scope of 1,049 outcomes or the adjusted Cox modeling framework.

    From Prospective cohort (UK Biobank) observational analysis

7 things the story did carry across
  • Prospective UK Biobank cohort design using wrist-worn accelerometry in 95,559 participants with median 8.9 years of follow-up.
  • Sleep stages and pattern metrics were algorithm-derived from wrist accelerometer data using SleepNet, including REM, N1/N2/N3, total sleep duration, sleep irregularity, and WASO.
  • Greater REM sleep was associated with lower risk for 83 diseases; higher deep sleep/N3 was associated with lower risk for 7 diseases.
  • Greater sleep irregularity and increased wakefulness after sleep onset were associated with elevated risk for several diseases, with abstract-level counts of 3 and 6 diseases respectively.
  • Restricted cubic spline analyses found nonlinear sleep-duration associations for 86 phenotypes, with minimum-risk sleep duration for 69 phenotypes predominantly concentrated in the 6–8-hour range.
  • Category-specific analysis found extreme short sleep below 5 hours had the broadest adverse association pattern, accounting for 37 of 41 significant adverse associations versus the 6–8-hour reference.
  • Observational design limits causal inference and leaves potential for residual confounding.
Then read the study layer

Study layer

Study at a glance

Scan the study first. Expand only the parts you want to inspect.

Pieces of work

3

Evidence read

study summary

Lead result

secondary data

1Lead resultsecondary dataCreate an atlas (phenome-wide map) of associations between accelerometer-derived real-world sleep stages/pattern metrics and risk of incident diseases in UK Biobank.Prospective cohort (UK Biobank) observational analysisExpand

In plain English

Prospective phenome-wide cohort analysis in UK Biobank (n=95,559) using wrist accelerometer data processed with the SleepNet algorithm to derive sleep stages (REM, N1, N2, N3) and pattern metrics (total sleep duration, sleep irregularity, WASO), relating these exposures to incidence of 1,049 disease outcomes using adjusted Cox proportional hazards models and spline/category-specific analyses over a median 8.9-year follow-up.

Key findings

  • Overall variation in accelerometer-derived sleep patterns was associated with incidence of 156 diseases across the phenome.156 disease associations identified (direction and magnitude varied by exposure and outcome)
  • Greater REM sleep amount was associated with lower risk for multiple diseases.Lower risks for 83 diseases associated with higher REM sleep (counts reported in abstract)
“we analyzed wrist-worn accelerometer data from 95,559 Biobank participants and derived key metrics of real-world sleep patterns: rapid eye movement [REM], N1, N2, and N3; total sleep duration; sleep irregularity; and wakefulness after sleep onset using the SleepNet algorithm.”
What this piece can’t prove
  • Observational cohort design; susceptible to residual confounding and does not permit causal inference (stated as main limitation in abstract).
2secondary dataCharacterize (and test for) non-linear dose–response relationships between sleep duration (and related sleep metrics) and incident disease risk, including identification of minimum-risk sleep-duration ranges (e.g., 6–8 hours).restricted cubic spline dose–response within Cox proportional hazards modelsExpand

In plain English

Within a UK Biobank cohort (n=95,559, median follow-up 8.9 years), the authors used restricted cubic spline (RCS) models implemented in Cox proportional hazards regression to test for non-linear dose–response relationships between accelerometer-derived sleep duration (hours) and incidence of 1,049 disease phenotypes. RCS analyses identified significant non-linear associations for 86 phenotypes (P for nonlinear < 0.05) and estimated minimum-risk sleep-duration windows for 69 phenotypes that were predominantly concentrated in the 6–8 hours range.

Key findings

  • RCS analyses identified significant non-linear relationships between sleep duration and incident disease for 86 phenotypes (P for nonlinear < 0.05).
  • Estimated minimum-risk sleep-duration windows were concentrated in the 6–8 hours range for 69 phenotypes with nonlinear associations.
“Phenome-wide association analysis and restricted cubic spline (RCS) analyses were performed using Cox proportional hazard regression...”
What this piece can’t prove
  • Observational cohort design limits causal inference and remains susceptible to residual confounding (noted by authors).
  • Abstract lacks detail on spline specification (knot number/placement), covariate sets for each spline model, and multiple-testing correction strategy.
  • Measurement limitations: accelerometer-derived sleep-duration estimates and derived sleep metrics may have classification error relative to polysomnography.

1 further detail could not be confirmed from the summary.

3secondary dataCompare disease-risk associations across sleep-duration categories (e.g., extreme short sleep <5h versus 6–8h reference) to identify which categories show the broadest clinical vulnerability.secondary data cohort categorical exposure phenome-wide analysisExpand

In plain English

In a category-specific analysis of accelerometer-derived sleep duration in UK Biobank participants, individuals with extreme short sleep (<5 hours) showed the broadest pattern of adverse incident-disease associations: 37 of 41 identified significant adverse associations were attributed to the <5h group versus the 6–8h reference.

Key findings

  • Extreme short sleep (<5 hours) exhibited the most widespread adverse disease-association pattern versus the 6–8 hour reference.37 of 41 significant adverse associations (count) attributed to <5h group vs 6–8h reference
“In the category-specific analysis, individuals with extreme short sleep (<5 hours) exhibited the most widespread clinical vulnerabilities, accounting for 37 of the 41 identified significant adverse associations compared to the 6-8 hours reference group.”
What this piece can’t prove
  • Observational design susceptible to residual confounding and does not permit causal inference (as stated by the authors).
Finally, the search trail

Method layer

NewsLink found the paper. Tessa takes you deeper.

NewsLink checks the story. Tessa is where you inspect the paper, authors, evidence, and research context.

Papers considered

The selected paper, plus nearby candidates.

PubMed, Europe PMC, Crossref · 37 candidate papers

Candidate

Author response for "Behavioral medicine in cholangiocarcinoma: bridging the gap between biology and behavior"

2026 · Crossref

Source (opens in a new tab)

And 31 more candidates considered.