CBT-I vs sleep education and epigenetic ageing in older adults with insomnia
In a secondary analysis of a randomised trial in adults aged 60 or older with insomnia, CBT-I was linked to a slower DunedinPACE pace of biological ageing than sleep education therapy; GrimAge and PCPhenoAge showed no significant differences. [1]
Automatic summary · not editorially reviewed
The key findings
This was a secondary analysis within a larger randomised controlled trial of adults aged 60 or older with insomnia disorder, comparing CBT-I with sleep education therapy. [1]
CBT-I was associated with a significantly slower DunedinPACE pace of biological ageing than SET, while GrimAge and PCPhenoAge showed no significant differences. [1]
Participants receiving CBT-I were more likely to have full remission after treatment than those receiving SET (34% vs 13%). [1]
The source at a glance
- Study type / format
- Secondary analysis of a randomised controlled trial (abstract)
- Population / context
- Adults aged 60 or older meeting DSM-IV criteria for insomnia disorder, recruited in Los Angeles, USA.
- Summary based on
- Study abstract
- Original publication
- July 1, 2026
Insomnia treatment and biological ageing in adults 60+
The study asked whether treating insomnia affects epigenetic indicators of biological ageing in older adults. It was a secondary analysis within a larger single-site randomised controlled trial of adults aged 60 or older meeting DSM-IV criteria for insomnia disorder, recruited in Los Angeles, USA. Participants were randomly assigned to cognitive behavioural therapy for insomnia (CBT-I) or sleep education therapy (SET), an active comparator. Analyses included only participants with complete paired epigenetic data.
Supporting source [1]
Blood samples before and after treatment
Peripheral blood mononuclear cells were collected before treatment started (baseline) and at a follow-up visit 20 to 39 months after baseline; follow-up times over 30 months were winsorised to 30 months. Biological age was estimated with DNA methylation-derived epigenetic clocks: DunedinPACE, GrimAge and the principal component version of PhenoAge (PCPhenoAge). The analysed groups were CBT-I (n=47; mean age 69.5 years) and SET (n=45; mean age 69.4 years), randomly selected based on sample availability.
Supporting source [1]
Slower DunedinPACE pace, other clocks not significant
Participants receiving CBT-I were more likely to have full remission after treatment than those receiving SET (34% [16 of 47] vs 13% [six of 45]). Compared with SET, CBT-I was associated with a significantly slower pace of biological ageing estimated by DunedinPACE (-0.02, 95% CI -0.04 to -0.01, false discovery rate-adjusted p=0.03). No statistically significant differences were seen for GrimAge (-0.33, -0.68 to 0.03, p=0.11) or PCPhenoAge (-0.49, -1.34 to 0.36, p=0.26). These are associations from a secondary analysis of an abstract; the full text was not reviewed.
Supporting source [1]
How to interpret this
This is an abstract of a secondary analysis; the full text was not reviewed. Only participants with complete paired epigenetic data were analysed, groups were randomly selected based on sample availability, and follow-up times over 30 months were winsorised. The findings are associations, not proof that CBT-I causes slower biological ageing.
Based on the study abstract; the full paper, tables and supplementary data were not reviewed.
Source & supporting passages
Carroll JE, Kusters CD, Taha HB, Olmstead R, Breen EC, Irwin MR. · europepmc.org · July 1, 2026