No improvement in AUDIT-C screening and brief intervention rates among wait-list controls following support of Aboriginal Community Controlled Health Services: evidence from a cluster randomised trial

Table 3 Model predicting the odds of clients being screened with AUDIT-C by condition and study phase

Predictors	OR [95% CI]	lnOR	SE	\(z\)	p
Fixed effects
Intercept	0.15 [0.11, 0.20]	-1.88	0.15	-12.38	< 0.001
Active Support	0.83 [0.52, 1.32]	-0.18	0.24	-0.78	0.43
Time	0.94 [0.76, 1.17]	-0.06	0.11	-0.53	0.60
Active Support & Time	0.94 [0.67, 1.32]	-0.06	0.17	-0.34	0.74
Random effect summaries
\({\tau }_{00}\)id	0.07
\({\tau }_{00}\)service	0.50
\({\tau }_{11}\)service.Time	0.23
\({\rho }_{01}\)service	-0.69
ICC	11.34%

Note lnOR = Natural logarithm of the OR (logits); SE = Standard error of the estimate (lnOR); \(z\) = the ratio of the lnOR to its associated standard error; p values were estimated using the two-tailed Wald z-test. Active Support = the relative effect of attending an Active Support service, Time = Observation occurred following the start of active support. The interaction between Active Support and Time tests the effect of the intervention. Variable type of random effect summaries listed in rows. \(\tau\) = random effect variance (lnOR). \(\rho\) = correlation between random intercepts and slopes. ICC = conditional intraclass correlation coefficient. Analysis by original assigned groups

ISSN: 1472-6963