Table 3. Differences (95% CI) between treatment arms (CATCH-ED—Usual care) at 12 months, intent-to-treat analysis.

Count outcomes1	Rate ratio (95% CI)	p-value
Primary outcome
Emergency department (ED) visits	0.86 (0.64 to 1.15)	0.31
Secondary outcomes
Days in hospital	1.16 (0.59 to 2.29)	0.66
Exploratory outcomes
Hospital admissions	0.78 (0.46 to 1.30)	0.34
Number of primary care provider visits	0.83 (0.63 to 1.09)	0.18
Normally distributed outcomes2	Mean difference (95% CI)	p-value
Secondary outcomes
Physical Component Score (SF-12)	1.50 (-2.15 to 5.15)	0.42
Mental Component Score (SF-12)	-3.97 (-8.13 to 0.19)	0.06
Severity of mental health problems (CSI total score)	-0.41 (-2.30 to 1.49)	0.68
Alcohol composite score (ASI)	0.053 (-0.017 to 0.12)	0.14
Drug composite score (ASI)	-0.0027 (-0.028 to 0.023)	0.84
Exploratory outcomes
Disease-specific quality of life (QoLI-20 total score)	-9.12 (-17.57 to -0.67)	0.04
Global item (QoLI-20)	0.19 (-0.36 to 0.74)	0.51
Overall health visual analogue scale (VAS) of the EQ-5D	-4.17 (-12.4 to 4.03)	0.32

¹ Count outcomes were modelled using an analysis of covariance framework, using a negative binomial distribution that compared the number of outcome events in the CATCH-ED group compared to the TAU group during the 12-month post-randomization period, adjusting for baseline number of events accrued in the 12-month pre-randomization period, as well as participant age and sex. Resulting rate ratios (RR) and their 95% CI calculate the ratio of the estimated frequency of CATCH-ED events divided by the frequency of TAU events for the 12-month post-randomization period, adjusting for baseline frequencies, age and sex.

² Self-reported health outcomes which approximated the normal distribution were analysed using an analysis of covariance framework to model difference in means (95% CI) between CATCH-ED and TAU groups at 12-month post-randomization, adjusting for baseline values.