Table 3.

Effect of PAMG (prospective vs. retrospective) on acute asthma management in ED (N = 309).

Asthma Management in ED	Unadjusted			Adjusteda
Logistic Regression	ORb	95% CI	p-value	OR	95% CI	p-value
Systematic corticosteroids given in ED	3.11	1.82–5.30	<0.0001	2.26	1.63–3.14	< 0.0001
SaO2 reassessed before discharge	2.14	1.34–3.42	0.0016	2.02	1.45–2.82	< 0.0001
Linear Regression	βc	95% CI	p-value	β	95% CI	p-value
Number of salbutamol doses given in first hour	0.22	0.07–0.37	0.0410	0.23	0.03–0.44	0.0283

CI: confidence intervals.

^aAll models were adjusted for the propensity score (or conditional probability) of a patient being selected for each treatment, given the pre-treatment characteristics including age, gender, history of ED visit/hospitalization for asthma, presence of wheeze, use of accessory muscles, SaO2 at initial presentation, length of stay, season of ED visit, day of ED visit, and hour of ED visit.

^bOR: Odds ratio (prospective versus retrospective group) was estimated using logistic regression.

^cRegression coefficient β (prospective versus retrospective group) was estimated using linear regression. β equals the average increase in the number of salbutamol doses in the prospective group.