Skip to main content

Table 3 Difference-in-difference analysis for the relationship between IPVA victimisation at ages 18–21 and logged depressive symptom scores at ages 16 and 23 (models fitted in imputed datasets, N = 1764 women and 1028 men)

From: Exploring the causal role of intimate partner violence and abuse on depressive symptoms in young adults: a population-based cohort study

 

Women

Men

Term

% changeÏ®

Ratio of geometric means (95% CI)

p value

% changeÏ®

Ratio of geometric means (95% CI)

p value

IPVAÂ¥

29

1.29 (1.17, 1.42)

< 0.001

34

1.34 (1.16, 1.55)

< 0.001

Time¥¥

5

1.05 (0.98, 1.13)

0.18

34

1.34 (1.22, 1.47)

< 0.001

IPVA*Time¥¥¥

2

1.02 (0.89, 1.18)

0.76

−5

0.95 (0.77, 1.18)

0.665

  1. CI confidence interval
  2. *An interaction between two variables
  3. ϮIn geometric mean. % change calculated from the estimated coefficient for victimisation in each model, as % change = [exp(coefficient) − 1] × 100
  4. ¥The difference in depressive symptoms at baseline (age 16 years) comparing people exposed to IPVA with people not exposed to IPVA
  5. ¥¥The change in depressive symptoms over time, in people not exposed to IPVA
  6. ¥¥¥The difference in difference, i.e. the difference in the change in depressive symptoms over time, comparing people exposed to IPVA with people not exposed to IPVA. This is estimated as an interaction between IPVA and time
Back to article page

BMC Medicine

ISSN: 1741-7015

Contact us