Background An association between parental separation occurring during childhood and psychological distress has been shown many times. UK divorce rates have increased rapidly since the mid-twentieth century. The “reduced effect hypothesis” suggests that the effect of parental separation may have reduced over time as separation has become more common and consequently less stigmatising. Previous studies have looked at this using outcomes of educational attainment, psychological distress in mid-adulthood and receipt of welfare benefits, finding that the effect has not reduced over time; however the effect upon psychological distress in early adulthood has not yet been investigated.

Objective To examine whether the effect of parental separation occurring during childhood on psychological distress in early adulthood has reduced over time, and whether this differs for men and women.

Data, participants and variables This study uses data on 9064 and 6906 participants of the 1958 National Child Development Study (NCDS) and 1970 British Cohort Study (BCS70), respectively. Five sweeps of each study were used – birth, age 5/7, age 10/11, age 16 and age 23/26 years. Parental separation was measured from 0 to 16 years and psychological distress was measured using Rutter's Malaise Inventory at age 23 (NCDS) or age 26 (BCS70), treated as a binary variable (0–7=no distress, 8–24=distress). Mental illness in the family was treated as a confounder.

Statistical methods Logistic regression was used to test the association between parental separation and psychological distress in either cohort, both unadjusted and adjusted for confounders, using a pooled data set of both cohorts. Period changes were assessed by testing a cohort-separation interaction. Analyses were conducted separately for men and women.

Results 9.4% of NCDS participants experienced parental separation in comparison to 21.0% of BCS70 participants. After adjusting for confounders parental separation increased the odds of reporting psychological distress in NCDS men (OR 2.18, 95% CI 1.34 to 3.54) and women (OR 1.48, 95% CI 1.05 to 2.08) and this did not differ by gender (p=0.11). Parental separation was not associated with psychological distress in BCS70 adjusted analyses for men or women. However cohort-separation interactions were not statistically significant (men: p=0.11, women: p=0.43).

Conclusions Despite finding an effect of separation in the NCDS and not in the BCS70, the cohort-separation interactions tested were statistically insignificant. The findings of this study therefore imply that the impact of parental separation has not changed over time (“reduced effect hypothesis” is not supported) and that men and women are affected equally.