Table 2.

Results from the Anderson and Gill multivariate model estimating the likelihood of experiencing exacerbations during the 2-year follow-up (n = 22,102)

	Hazard ratio	95% confidence interval		P value
		Lower limit	Upper limit
Characteristics
Number of SABA canisters/year
≤ 2 SABA canisters/year	1.0 (ref.)	–	–
> 2 SABA canisters/year	1.27	1.21	1.33	< 0.0001
ICS mean daily dosage (BDP equivalent micrograms)
No ICS	1.0 (ref.)	–	–
0 < ICS < 200 mcg	1.02	0.98	1.06	0.4173
200 ≤ ICS ≤ 500 mcg	0.97	0.93	1.02	0.2534
500 < ICS ≤ 1000 mcg	1.10	1.05	1.16	0.0002
ICS > 1000 mcg	1.30	1.23	1.38	< 0.0001
Sex
Male	1.0 (ref.)	–	–
Female	1.35	1.31	1.39	< 0.0001
Age class
Age < 18 years	1.0 (ref.)	–	–
18 ≤ Age < 40 years	0.84	0.77	0.91	< 0.0001
40 ≤ Age < 65 years	1.03	0.95	1.11	0.4731
Age ≥ 65 years	1.17	1.08	1.28	0.0001
Number of baseline comorbidities
No comorbidities	1.0 (ref.)	–	–
One comorbidity	1.20	1.16	1.24	< 0.0001
Two comorbidities	1.22	1.16	1.28	< 0.0001
At least three comorbidities	1.43	1.35	1.53	< 0.0001
Treatment step
Step 1	1.0 (ref.)	–	–
Step 2	0.84	0.78	0.91	< 0.0001
Step 3	0.78	0.74	0.83	< 0.0001
Step 4	0.83	0.80	0.87	< 0.0001
Step 5	1.42	1.34	1.49	< 0.0001

SABA short-acting β₂-agonists, ICS inhaled corticosteroids, BDP beclomethasone dipropionate