Household occupancy and general health: Evidence from England and Wales census-based study

There is growing evidence that living conditions affect health, and that self-rated health (SRH) is strongly linked to morbidity and mortality. However, less is known about how specific household characteristics relate to SRH. This study examined how SRH varies by household occupancy and how poor SRH is associated with the duration of living alone (LA), using Office for National Statistics Longitudinal Study (ONS LS) data. The ONS LS includes linked Census data for 1% of the population of England and Wales since 1971, along with additional administrative records. SRH was classified as ‘good’ or ‘poor’ using an upper quartile method for better comparability. The sample included around 270,000 individuals aged 16+ who were present in all three censuses 2001, 2011, and 2021. At baseline (2001), 46% were male, mean age was 40.6, 28% reported poor SRH, and 10% lived alone (LA). In crude analyses, compare to not-LA, LA was strongly associated with poor SRH across all waves (ORs: 1.56, 1.95, 2.14). After adjustment, the association increased in 2001 (OR: 1.65), weakened in 2011 (OR: 1.05), and was explained completely in 2021 (OR: 0.97). Examining frequency of LA, compare to those not LA, those who LA once, twice, or three times had higher odds of poor SRH in unadjusted models (ORs: 1.41, 2.03, 2.56), but these associations were largely explained after adjustment (ORs: 1.07, 1.07, 1.08). Overall, household composition appears to be an important predictor of poor SRH, with potential implications for policies targeting single-person households.

Key messages

• Household occupancy, especially living alone, links to poorer self-rated health, but is mainly explained by demographic, socioeconomic factors and frequency of living alone not the arrangement itself.

• Household composition still matters as a predictor of health, highlighting the need for policies that better support individuals in single-person households.