This study used genetic proxies to examine how childhood adiposity and adulthood adiposity influence sleep health later in life. Outcomes assessed were insomnia, morning chronotype, sleep duration, daytime sleepiness and daytime napping. The analysis distinguished total effects and direct effects of adiposity at different life stages by applying multivariable approaches that adjust childhood adiposity estimates for adulthood adiposity, and vice versa.
Genetic instruments for childhood and adulthood body size were obtained from genome-wide association study (GWAS) summary statistics based on UK Biobank participants (n = 453,169). Summary data for obstructive sleep apnea (OSA) were drawn from the Million Veteran Program (n = 410,268). GWAS summary statistics for insomnia and chronotype were requested from 23andMe and large-scale consortia; the source reports large combined sample sizes for those outcomes. The authors state that GWAS summary statistics are available by application to the respective data holders (UK Biobank, MVP, 23andMe).
The study applied two-sample Mendelian randomization (MR). Univariable MR estimated the total effect of genetically predicted adiposity at each life stage on sleep traits. Multivariable MR estimated direct effects of childhood adiposity while adjusting for adulthood adiposity (and additional models adjusted for OSA where specified). The intention was to disentangle whether exposures at specific life stages have independent causal impacts on later-life sleep outcomes.
In univariable MR, there was no evidence that genetically predicted childhood adiposity increased later-life insomnia risk (Odds ratio [OR] = 0.94, 95% Confidence interval [CI] = 0.87–1.03). However, in multivariable MR adjusting for adulthood adiposity, the authors report a direct protective association of higher childhood body size with later-life insomnia (OR = 0.70, 95% CI = 0.64–0.77).
The study also reports that increased adulthood body size increased the risk of insomnia in both univariable and multivariable analyses. The source indicates that both childhood and adulthood adiposity were associated with OSA, and that OSA was associated with insomnia; additional adjustment for OSA did not materially change the childhood adiposity estimates reported for insomnia.
Both univariable and multivariable MR showed a positive association between increased childhood body size and being a morning chronotype. Reported estimates were:
These estimates reportedly persisted after additional adjustment for OSA. The source does not provide detailed numeric results for sleep duration, daytime sleepiness or daytime napping in the abstract; those outcome-specific effect estimates were not reported in the source abstract.
The authors examined OSA in the causal pathway and report that both childhood and adulthood adiposity were associated with OSA, and that OSA was associated with insomnia. The source states that further adjustment for OSA did not substantially change the estimates relating childhood adiposity to insomnia or chronotype in the analyses described.
The authors conclude that higher body size in childhood does not appear to be a risk factor for later-life insomnia when adulthood adiposity is taken into account; by contrast, higher body size in adulthood was associated with increased insomnia risk. They further conclude that if a healthy body size is maintained in adulthood, higher childhood adiposity may be associated with a lower risk of insomnia and a higher likelihood of being a morning person in later life. These conclusions are derived from MR analyses using genetic proxies and reported GWAS summary statistics.
The source reports that GWAS summary statistics for childhood and adulthood body size are available on request from UK Biobank. OSA summary statistics are available via application through the Million Veteran Program accession listed in the source. Insomnia and chronotype summary data are available from 23andMe on request. The authors state that ethical approvals and participant consent were obtained as appropriate and that all reporting guidelines relevant to the work were followed. One author is employed by GlaxoSmithKline; the remaining authors report no competing interests.
Note: the source abstract provides the principal methods and key estimates for insomnia and chronotype but does not report full numerical results for all sleep traits (sleep duration, daytime sleepiness, daytime napping) within the abstract text. Readers should consult the full preprint or supplementary materials for complete outcome-specific effect estimates and sensitivity analyses.