Background Previous studies have shown that making multiple residential moves in childhood leads to an increased risk of emotional and behavioural problems in early adulthood and to poorer self-reported health in midlife. Such studies tend to focus on one or two health variables, measured at one time point. This study examines health status in early adulthood and midlife across a wider range of measures.

Aim To compare subjects who were residentially stable in childhood with those who had moved more often in terms of a wide range of health measurements at 18 and 36.

Methods Analysis of the 1970s cohort of the West of Scotland Twenty-07 Study. In total, 850 respondents who participated in waves 1 (1987/88), 2 (1990/92) and 5 (2007/8) of the study, and whose childhood residential history was available, were included in regression analyses. Residential stability was derived from the number of addresses at which the respondent had lived between birth and age 15 and 18. We considered directly measured health variables (BMI, waist-hip ratio and lung function), self-reported health, psychological wellbeing (GHQ12) and self-reported health behaviours (smoking, drinking and trying drugs).

Results Twenty percent of respondents remained residentially stable during childhood, 59% had moved 1-2 times and 21% had moved at least 3 times. Directly measured health variables were not associated with number of residential moves made at 18 or 36. Odds of scoring at least 3 on the GHQ12 questionnaire were significantly increased at age 18 for those moving 1-2 times (OR 2.01 (1.36–2.96)) and those moving 3 times or more (OR 2.04 (1.3, 3.22)) compared to those who remained stable. Similarly, odds of reporting a long-standing illness at 18 were increased for 1-2 moves (OR 1.88 (1.11, 3.18)) and at least 3 moves (OR=2.03 (1.11, 3.69)). Odds were elevated, but not significant, at 36 for these health variables. Odds of trying drugs and smoking at 18 were significantly increased but only for those moving at least 3 times. Although elevated, odds for these health behaviours were not significant at 36.

Conclusions Increased mobility during childhood is independently associated with adverse health status. At 18, the relationships between residential mobility and self-reported health outcomes, psychological wellbeing and some health behaviours were significant; however, by 36 findings were no longer significant. Directly measured health variables, at 18 and 36, do not appear to be associated with childhood mobility.