Table 3.

Multiple linear regression of factors associated with perceiving risk in disclosure of prenatal mental health problems.

Variablea	Beta (95% CI)	Standard error	Beta	P value	Partial eta squaredb
Income (less than Can $40,000)	1.11 (0.25-1.98)	0.44	.11	.01	0.010
Marital status (unpartnered)	.69 (−0.77 to 2.16)	0.75	.04	.35	0.001
Treated previously for depression anxiety (treated)c	.84 (0.19-1.49)	0.33	.10	.01	0.010
Born in Canada (No)c	−.76 (−1.55 to 0.03)	0.40	−.08	.06	0.006
Parity (first child)	.85 (0.23-1.46)	0.31	.11	.007	0.012

^aIndependent variables with P<.20 were entered simultaneously into the final model, including income, marital status, previously treated for depression or anxiety, born in Canada, and parity. The supplementary table of the univariate analysis is available from the corresponding author.

^bOn the basis of guidelines [32], a partial eta squared of >0.01 is a small effect size, >0.06 is medium, and >0.14 is large.

^c”Diagnosed and treated previously for depression or anxiety” were highly correlated and could not be entered into the same model (Pearson r=.85). Similarly, “born in Canada” and “ethnicity” were highly correlated (Pearson r=.60) and not entered together.