To estimate the impact of the UK nationwide campaign to End loneliness on loneliness and mental health outcomes among older people in England.
Quasi-experimental design, namely, a difference-in-differences approach.
Local authorities across England.
Older adults aged 65 and over participating in waves 4-8 (2008-2017) of the English Longitudinal Study of Aging (ELSA) and waves 1-9 (2009-2019) of the UK Household Longitudinal Study (UKHLS).
Loneliness was measured through the UCLA Loneliness scale. A social isolation scale with components of household composition, social contact and participation was constructed. Mental health was measured by The Centre for Epidemiological Studies of Depression (CES-D) score, the General Health Questionnaire (GHQ-12) score, and the Short-Form-12 Mental Component Summary (SF-12 MCS) score.
There was no evidence of change in loneliness scores over the study period. Difference-in-differences estimates suggest that explicitly developed and implemented antiloneliness strategies led to no change in loneliness scores (estimate = 0.044, SE = 0.085), social isolation caseness (estimate = 0.038, SE = 0.020) or levels of depressive symptoms (estimate = 0.130, SE = 0.165). Heterogeneity analyses indicate that antiloneliness strategies produced little impact on loneliness or mental health overall, despite small reductions in loneliness and increases in social engagement among well-educated and higher-income older adults. The results were robust to various sensitivity and robustness analyses.
Antiloneliness strategies implemented by local authorities have not generated a significant change in loneliness or mental health in older adults in England. Generating changes in loneliness in the older population might require longer periods of exposure, larger scope of intervention or more targeted strategies.