. 2022 Nov 8;10:20503121221135990. doi: 10.1177/20503121221135990

Table 3.

Follow-up assessment for mental health patients in a randomized controlled trial.

	Dependent variable: Follow-up status
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Main independent variable
Treatment	0.092^*	0.091^*	0.086^*	0.082^*	0.089^*	0.072	0.045
Treatment	[0.050] (0.01, 0.17)	[0.052] (0.01, 0.17)	[0.066] (0.007, 0.16)	[0.078] (0.003, 0.16)	[0.060] (0.008, 0.17)	[0.119] (–0.006, 0.15)	[0.327] (–0.03, 0.12)
Other associated factors
PHQ-9 pre score		0.008^*	0.0078^*	0.0079^*	0.0077^*	0.008^*	0.005
PHQ-9 pre score		(0.004)	(0.004)	(0.005)	(0.004)	(0.004)	(0.004)
Age			−0.005^***	−0.004^**	−0.003^**	−0.0027	−0.003^*
Age			(0.0016)	(0.0017)	(0.0017)	(0.0017)	(0.0017)
Lebanese				−0.110^**	−0.113^**	−0.148^***	−0.158^***
Lebanese				(0.054)	(0.054)	(0.054)	(0.055)
Basic education					−0.05	−0.07	−0.07
Basic education					(0.05)	(0.05)	(0.05)
New patient						−0.23^***	−0.17^***
New patient						(0.046)	(0.049)
Month of visit dummies							Yes
Observations	405	405	401	401	400	400	400

PHQ: Patient Health Questionnaire; CI: confidence interval.

Coefficients from the linear probability model (LPM) were reported. Similar results were obtained from maximum likelihood estimation. The table shows seven different specifications with the gradual addition of control variables (referred to as other associated factors). The last specification includes month of visit dummies representing dummies for 11 out of 12 months of the year with the remaining month left as a reference category. Significance represented through stars corresponds to unadjusted p-values. For the main independent variable, we report both the unadjusted p-values (in brackets) and the design effects adjusted confidence intervals (90% CI) (in parentheses). For the other associated factors, we report the estimates along with the standard errors (in parentheses).

p < 0.1. **p < 0.05. ***p < 0.01.