Skip to main content
. 2021 Jan 12;52(13):2805–2813. doi: 10.1017/S0033291720005000

Table 4.

Unstandardized regression coefficients, and 95% confidence intervals, for the change in sMFQ scores (continuous outcome) per 20-point increase in risk-taking (continuous exposure) at age 14, in the complete-case sample (n = 8628)

Sample overall Coefficient (95% CI) p
Model 1a: Univariable (n = 8628) −0.52 (−0.71 to −0.33) <0.0001
Model 1b: Model 1a adjusteda −0.54 (−0.73 to −0.35) <0.0001
Model 1c: Model 1b adjusted for sex 0.05 (−0.15 to 0.25) 0.637
Sub-group analyses by sex Coefficient (95% CI) P
Model 2a: Univariable, females (n = 4354) 0.08 (−0.29 to 0.44) 0.680
Model 2b: Model 2a adjustedb 0.02 (−0.34 to 0.38) 0.893
Model 3a: Univariable, males (n = 4274) 0.05 (−0.16 to 0.26) 0.612
Model 3b: Model 3a adjustedb 0.08 (−0.14 to 0.29) 0.472
a

Adjusted for confounders measured at or as close as possible to the time of the exposure (age 14): family income, maternal education, child age, child ethnicity, child vocabulary score, main carer depressive symptoms.

b

Adjusted for the above confounders and, in addition, stage of breast development in females or facial hair development in males at age 14.