Background In England, nearly one child in ten lives in overcrowded housing. Crowding is likely to worsen with increasing population size, urbanisation, and the housing crisis. Children with behavioural difficulties are at increased risk of mental and physical health problems and poorer employment prospects later in life. We examined the relationship between crowding and behavioural problems in children aged 3 years within the Southampton Women’s Survey.

Methods The SWS is a population-based cohort study of women aged 20–34 years assessed when not pregnant; those who became pregnant and their children (n=3,158) were followed up. Mothers were interviewed when children were aged 2 years; they reported the number of people living in their home, the number of rooms and their perception of space within their home. A measure of crowding was derived as the number of people per room. At age 3 years, children’s behaviour was assessed with the Strengths and Difficulties Questionnaire (SDQ). We used the total SDQ score as a continuous measure and univariate and multiple linear regression analyses were performed adjusting for confounding factors identified in a Directed Acyclic Graph: sex, single-parent family, maternal educational attainment, receipt of benefits, and occupation.

Results 2602 children (52% boys) followed to age 3 years had complete data on crowding at age 2 years. The characteristics of the sample were broadly representative of the population in England. Median (IQR) SDQ score was 9 (6–12) and people-per-room was 0.75 (0.6–1), with 327 (13%) of mothers and children living in socially rented accommodation and 13.5% living in homes with ≥1 people-per-room. Mothers’ perception of space within their home tended to agree with the objective people-per-room measure of crowding. An increase of one household person-per-room was associated with a 1.30 increase SDQ points (95% CI 0.70 to 1.90) adjusting for confounding factors. This relationship was partially mediated by greater maternal stress and depression, increased use of childcare, less sleep and strained parent-child interactions. Behavioural problems were greater in children who lived in socially-rented houses (by 2.69 SDQ points, 95% CI 2.17 to 3.20) than in children living in owner-occupied homes.

Conclusion Living in a more crowded home was associated with a greater risk of behavioural problems at age 3 years, independent of the effects of confounding factors. These findings support the need for improved housing for families, and suggest that families living in crowded circumstances might benefit from greater support.