TABLE 3.

Regression analyses for IPTS factors, gender, and their interactions predicting suicidal ideation at follow-up

	Suicidal ideation at follow-up
	Model 1		Model 2	Model 3		Model 4
	PB	TB	PB & TB	PB	TB	PB	TB
Baseline suicidal ideation	.46 (.05)***	.50 (.04)***	.46 (.05)***	.47 (.05)***	.50 (.05)***	.47 (.05)***	.47 (.05)***
Perceived burdensomeness	.20 (.05)***	-	.16 (.06)**	−0.09 (.11)	-	−.11 (.11)	.14 (.06)*
Thwarted belongingness	-	.15 (.04)***	.07 (.05)	-	−.10 (.09)	.06 (.05)	−.14 (.09)
Cisgender young man (reference)	-	-	-	-	-	-	-
Cisgender young woman	-	-	-	.12 (.12)	.07 (.11)	.12 (.12)	.06 (.12)
Transgender/genderqueer	-	-	-	.22 (.11)*	.21 (.11)*	.22 (.11)*	.16 (.11)
Cisgender young woman × PB	-	-	-	.34 (.13)**		.32 (.13)*
Transgender/genderqueer × PB	-	-	-	.33 (.12)**		.33 (.13)*
Cisgender young woman × TB	-	-	-		.23 (.11)*		.21 (.11)
Transgender/genderqueer × TB	-	-	-		.34 (.10)**		.32 (.11)**

Note: Standardized regression coefficients are shown with SE in parentheses. In Model 1, PB and TB were tested as separate predictors of SI at follow-up, controlling for SI at baseline; in Model 2, PB and TB were tested as simultaneous predictors of SI at follow-up, controlling for SI at baseline; in Model 3, gender was tested as a moderator of the associations between PB/TB (in separate models) and SI at follow-up, controlling for SI at baseline; in Model 4, gender was tested as a moderator of the associations between PB/TB and SI at follow-up, controlling for SI at baseline and the other IPTS factor; gender was dummy-coded and the IPTS factors were mean-centered to compute the interaction terms.

Abbreviations: PB, perceived burdensomeness; SI, suicidal ideation; TB, thwarted belongingness.