# Sleep and circadian health in the UK Biobank: report on the 2023 sleep questionnaire enhancement.

- **DOI:** [10.1093/sleep/zsag068](https://doi.org/10.1093/sleep/zsag068)
- **Rating:** Strong (3.8 out of 5)
- **Verdict:** A useful large-scale sleep phenotyping resource for UK Biobank that enables downstream research, constrained by unvalidated self-reported disorder definitions.
- **Framework version:** v5.9
- **Pipeline version:** v0.4
- **Evaluation date:** 2026-10-08
- **Evaluation page:** https://www.nabu.science/paper/10.1093%2Fsleep%2Fzsag068

## Key claims

1. Approximately one-quarter (25.2%) met the criteria for at least one operationally defined sleep disorder, with insomnia being the most common (14.4%) followed by obstructive sleep apnea (8.0%), restless legs syndrome (4.1%), and frequent nightmares (3.7%).
   - Support: Supported
   - Type: Primary finding
   - Basis: large descriptive sample, but self-reported operational definitions lack clinical confirmation
   - Location: Abstract, Results; Results, Operationally Defined Sleep Disorders; Table 2
2. Sleep disorders were associated with higher levels of anxiety, depression, fatigue, and cognitive complaints.
   - Support: Supported
   - Type: Secondary
   - Basis: descriptive group differences without covariate adjustment or independent symptom validation
   - Location: Abstract, Results; Results, Operationally Defined Sleep Disorders; Supplementary eTable S6
3. Around 40 per cent of respondents reported sleep duration less than 7 h, and 49 per cent reported poor sleep quality (Pittsburgh Sleep Quality Index >5).
   - Support: Supported
   - Type: Secondary
   - Basis: direct respondent-level frequencies from a selected older cohort
   - Location: Abstract, Results; Results, Sleep Health Dimensions; Supplementary eTable S2

## Strengths

- **Fills a documented measurement gap** The 138-item questionnaire expands UK Biobank beyond the single-item sleep measures criticised in the Introduction and supports multi-domain sleep and circadian phenotyping.
  - Dimension: contribution
  - Location: Introduction, paragraphs 2–4; Methods, Development of the Sleep Questionnaire
- **Transparent descriptive phenotype construction** Case definitions draw on diagnostic criteria, scale scoring, and working-group review, while two analysts independently applied and cross-checked the operational phenotypes.
  - Dimension: methodological rigour
  - Location: Methods, Defining Outcomes/Phenotypes and Analysis
- **Selection limits are directly examined** Respondents are compared with invited non-respondents and the broader cohort, and the Discussion explicitly rejects interpreting the reported frequencies as general-population prevalence.
  - Dimension: positioning
  - Location: Results, Respondent Characteristics, Table 1; Discussion, paragraph 2

## Limitations

- **Case definitions lack clinical validation** Operational classifications were not confirmed by clinical interviews, polysomnography, or prospective sleep–wake monitoring, and some definitions use single items or criteria with uncertain predictive validity.
  - Dimension: methodological rigour
  - Location: Discussion, paragraph 3
- **Transferability is substantially constrained** Respondents were predominantly White, older, healthier, and less deprived, limiting application of the observed distributions to other populations and settings.
  - Dimension: application pathway
  - Location: Abstract, Results; Results, Respondent Characteristics; Discussion, paragraph 2
- **Minor reporting and availability inconsistencies** Employment percentages differ across the Abstract and Results, and the public GitHub address contains a literal “[Journal]” placeholder, reducing confidence in those reporting details.
  - Dimension: reporting
  - Location: Abstract, Results; Results, Sleep Health Dimensions; Methods, Analysis

## Reliability checks

Caveats (4 of 4 checks).

- **Results check:** Noted. One reported value does not reconcile, but the key claims do not rest on it.
  - Minor reporting notes (1):
    - Descriptive: "and 25.5 per cent were in employment." (Abstract, Results) vs "The majority (73.6%) of respondents did not work, but those engaged in employment (and worked typical hours of 9 am–5 pm, n = 35 632, 19.3%) woke 53 min later on free days compared to work days in the past month (Supplementary eTable S3). For shift-workers (n = 10 908, 5.9%), sleep timing on work days in the past month followed a predictable pattern (e.g. morning shift workers had advanced sleep onset and offset times, while night-shift workers had markedly delayed sleep onset and offset times)." (Results, Sleep Health Dimensions). The complement of 73.6% not working is 26.4%, whereas the listed work subgroups sum to 19.3% + 5.9% = 25.2%, neither matching the Abstract's 25.5%; these employment comparisons do not change the reported sleep results. → Conclusion holds
- **Reporting and transparency:** Noted. Reference [19] is assigned to both the UK Biobank cohort description and the Berlin Questionnaire; the correct citation cannot be established from the supplied text.
  - Conflict of interest: not stated · Ethics approval: declared · Preregistration: not stated · Data availability: declared · Code availability: declared
- **Reference integrity:** Clean
  - 43 references in manuscript
  - 42 of 42 checkable references found in an index
  - 1 have no canonical index record — counted, but not index-checkable
  - 1 reference confirmed by manual review
  - References not verified automatically are resolved manually within 24 hours.

## Post-publication record

- **Status:** Clean
- **Summary:** No retraction or expression of concern found in Crossref (includes Retraction Watch).
- **Last checked:** 2026-10-17
