Insomnia as a long-term predictor of anxiety symptoms in the English Longitudinal Study of Ageing (ELSA), and the role of inflammation

introduction: insomnia may increase the risk of developing anxiety symptoms, although the long-term association between insomnia and anxiety has been rarely explored. Interestingly, disturbed sleep may also upregulate inflammatory immune system activity in late life, which may putatively be involved in the aetiology of anxiety. In this study, we examined whether insomnia may predict anxiety symptoms in a 9-year follow-up, and whether inflammation may play a mediating role. method: data from 1355 participants (mean age 63.44, SD = 7.47 years, 55.1% females) of the english longitudinal study of ageing (ELSA) were analysed. Insomnia symptoms were assessed in 2012/13 using the jenkins sleep problems scale. high-sensitivity serum C-reactive protein (hs-CRP) a marker of systemic inflammation, was measured in 2016/17. anxiety symptoms were assessed in 2020/21 using the 7-item generalized anxiety disorder scale. covariates included BMI, alcohol intake, physical activity, cardiovascular diseases, and cholesterol. factorial validity and empirical distinctiveness of the latent dimensions of insomnia and anxiety were analysed using a confirmatory factor analysis and structural equation models were performed. results: after adjusting for covariates and baseline levels, structural equation modelling revealed that insomnia was a significant predictor of anxiety symptoms at 9-year follow-up (β = 0.366, p < 0.001). insomnia was associated with hs-CRP cross-sectionally (r = 0.102, p < 0.05) but not longitudinally (β = −0.017, p = 0.606). similarly, baseline diagnoses of anxiety were predictive of later higher hs-CRP (B = 0.075, p = 0.032), but hs-CRP did not longitudinally predict anxiety. accordingly, the mediation hypothesis was rejected (β = 0.0004; 95% BCI −0.001 to 0.006). moreover, multi-group structural equation modelling highlighted that gender did not moderate longitudinal paths. conclusion: the current study highlight insomnia as a long-term predictor of anxiety symptoms in older adults from the general population, without a mediation of hs-CRP. alternative psychophysiological processes linking insomnia and anxiety should be therefore considered in future mechanistic studies, including autonomic and cortical pre-sleep arousal, cortisol reactivity, and the role of pro-inflammatory cytokines.