Table 2.

Proportional hazards analysis from first CHI-identified prescription to methadone-specific DRDs: with adjustment for baseline covariates excluding or including quintiles for prescribed quantity of methadone.

Covariates	361 Methadone-specific DRDs; 33,124 at-risk clients Regression chi-square test on 3df for interaction (gender*age-group) when region also fitted = 3.30, p ∼0.35
	Excluding QUINTILES for prescribed quantity			Including QUINTILES for prescribed quantity Regression chi-square (9 df) = 98.88
	HR	95% CI	p-value	HR	95% CI	p-value
No interaction needed for gender * age-group
Regression chi-square (4df) for QUINTILES				22.98, p ∼0.0001
Other-source	1.42	1.15–1.77	0.001	1.44	1.16–1.79	0.001
Female	1.12	0.90–−1.39	0.322	1.14	0.91–1.42	0.246
AGE-GROUP with 25–34 years at first CHI-identified prescription as baseline
Age < 25	0.51	0.26–1.00	0.050	0.53	0.27–1.04	0.064
Age 35–44	1.95	1.53–2.48	<0.001	1.91	1.50–2.44	<0.001
Aged 45+	2.95	2.18–4.00	<0.001	2.90	2.14–3.93	<0.001
QUINTILES for prescribed quantity of methadone with lowest quintile as baseline
Quintile 2: 281–700				1.10	0.76–1.59	0.604
Quintile 3: 701–1200				0.90	0.61–1.32	0.577
Quintile 4: 1201–1960				1.22	0.86–1.74	0.261
Quintile 5: > 1960 mg (For example, 1960 mg =  14 days@140 mg daily or 28 days @ 70 mg)				1.79	1.29–2.50	<0.001
When covariates for REGION [Greater Glasgow and Clyde (as baseline) versus Lothian, Tayside, rest of Scotland] were added to the above models, neither regression chi-square (on 3df) for REGION was statistically significant, see below.
Regression chi-square (3df) for REGION	5.80, p ∼0.12			2.42, p ∼0.49