Factors affecting the sleep health of older New Zealanders : the role of alcohol and work status : a thesis presented in partial fulfilment of the requirements for the degree of Master of Arts in Psychology at Massey University, Wellington, New Zealand

Abstract

Increased life expectancy is contributing to proportionally longer retirements, necessitating improved insight on factors that impact healthy ageing. How work/retirement status influences sleep and alcohol consumption is under-researched in New Zealand (NZ). To bridge literature deficits, this study explored how alcohol influences sleep in older workers and retirees. Retrospective analysis of the 2022 New Zealand Health Work and Retirement study used hierarchical regression modelling to assess whether the alcohol consumption (frequency/quantity) of retirees (no paid work, n=1774) and workers (n=3049) was independently associated with indicators of sleep disturbance (sleep duration (total hours per 24-hour day) (retiree Mdn=8 (IQR: 7-8), workers=7 (IQR: 6.5 8))), and substantial sleep problems (SSP) (retirees=35.16%, workers= 31.04%). Regression analysis controlled for demographics, physical and mental health status, job satisfaction, and weekly work hours. Workers’ drinking ≤monthly reported sleeping 17 minutes less than abstainers (b=-0.29, 95% CI=-0.55– -0.04, ϐ=0.09)). Drinking alcohol with any frequency above ‘never’ while employed correlated to increased SSP likelihood (drinking ≤monthly=122% increase (95% CI=1.34–3.68), drinking 2-4 times per month=99% (95% CI=1.19–3.33), drinking 2-3 times per week=151% (95% CI=1.51–4.17), and drinking ≥4 times weekly=175% (95% CI=1.63–4.64)). Retirees’ drinking 3-4 standard drinks reported 33 minutes less sleep than abstainers (b=-0.55, 95% CI=-1.09–0.00, ϐ=-0.12) and 157% increased likelihood of SSP (95% CI=1.01–6.49). Shorter duration correlated with being female, better self-rated physical health, and not smoking for retirees (fit: R²=.03, ΔR²=0, p= .516), and with more diagnoses, worse self-rated mental health, and more weekly work hours for workers (fit: R²= .04, ΔR²= .01, p= <.001). SSP correlated with poorer living standards, poorer self-rated physical and mental health, and not smoking for retirees (fit: χ²=240.9 (19), p= <.001, Nagelkerke’s R2= 0.25), and with being female, Māori, poorer living standards, poorer self-rated physical and mental health, and more diagnoses for workers (fit: χ²=406.475 (21), p= <.001, Nagelkerke’s R²=0.26). Alcohol was independently associated with sleep duration and SPP. Proposed mechanisms differed by work/retirement status, suggesting NZ policy needs clarity on how alcohol harms health during ageing. Future research exploring time use/scheduling, typical drinking routines, retirement causes, or longitudinal designs could further understanding and indicate health promotion opportunities.