Table 3.

Generalised linear model results with follow‐up HoNOS as dependent variable

	(1) Study sample			(2) Long EIP			(3) Short EIP
	Coeff.	SE	dy/dx	Coeff.	SE	dy/dx	Coeff.	SE	dy/dx
Log waiting time (continuous)	0.032***	(0.009)	0.20	0.043***	(0.010)	0.27	0.022	(0.015)	0.15
Waiting time quintiles (ref.cat.: shortest quintile)
2nd shortest quintile	−0.050	(0.036)	−0.32	−0.015	(0.049)	−0.09	−0.109	(0.064)	−0.73
3rd shortest quintile	−0.057	(0.041)	−0.36	0.004	(0.051)	0.02	−0.186**	(0.060)	−1.24
4th shortest quintile	0.031	(0.033)	0.20	0.050	(0.039)	0.31	0.036	(0.081)	0.25
Longest quintile	0.119**	(0.045)	0.78	0.199***	(0.053)	1.27	0.021	(0.072)	0.14
Waiting time intervals (ref.cat.: less than 0.5 month)
Waiting time 0.5 to 3 months	0.040	(0.023)	0.25	0.054*	(0.027)	0.34	0.034	(0.045)	0.23
Waiting time 3 to 6 months	0.120***	(0.034)	0.78	0.181***	(0.048)	1.15	0.076	(0.052)	0.52
Waiting time 6 to 12 months	0.215***	(0.059)	1.41	0.250***	(0.065)	1.61	0.189	(0.098)	1.30
Observations	8,949			6,135			2,814
Provider and year fixed effects	yes			yes			yes
Covariates	yes			yes			yes

Note. Model (1) includes the complete study sample. Model (2) includes only patients with an EIP episode longer than the follow‐up. Model (3) includes only patients with an EIP episode shorter than the follow‐up. “dy/dx” represents average marginal effects in days. For factor levels, they present the discrete change from reference category. All models use cluster robust standard errors for 48 provider clusters. EIP: early intervention in psychosis.

^* p < 0.05.

^** p < 0.01.

^*** p < 0.001.