Table 4.

Generalized linear model regression results for the cost of exacerbations (any type) during the follow-up year (n = 17,382)^*

Covariates	RR	SE	95% CI	P-value
Intercept	9576.28	1474.23	7082.02–12,949.00	<0.0001
US geographic region
Midwest	0.8969	0.0522	0.8002–1.0053	0.0617
Northeast	0.8584	0.0603	0.7480–0.9851	0.0298
West	0.9121	0.0747	0.7769–1.0708	0.2610
South (reference)
Health plan type
HMO	0.9857	0.0588	0.8769–1.1081	0.8096
Other	1.3216	0.1060	1.1293–1.5465	0.0005
PPO (reference)
Gender
Female	0.9703	0.0458	0.8845–1.0643	0.5223
Male (reference)
Age	1.0041	0.0023	0.9996–1.0087	0.0756
DCI	1.1555	0.0168	1.1231–1.1889	<0.0001
Fills for COPD maintenance medications during the follow-up year
LAMA	0.9934	0.0067	0.9805–1.0066	0.3275
Fixed-dose ICS + LABA inhaler	0.9827	0.0075	0.9681–0.9975	0.0221
LABA	0.9773	0.0137	0.9507–1.0046	0.1023
ICS	0.9790	0.0133	0.9533–1.0054	0.1176
Fills for COPD rescue medications during the follow-up year
SABA	0.9948	0.0058	0.9836–1.0062	0.3715
SAMA	1.0214	0.0071	1.0076–1.0354	0.0023
Number of exacerbations during baseline year	1.0908	0.0252	1.0425–1.1415	0.0002

Note:

^*Gamma distribution with log–link function.

Abbreviations: COPD, chronic obstructive pulmonary disease; DCI, Deyo–Charlson Comorbidity Index; HMO, health maintenance organization; ICS, inhaled corticosteroid; LABA, long-acting β_2-agonist; LAMA, long-acting muscarinic antagonist; PPO, preferred provider organization; SABA, short-acting β₂-agonist; SAMA, short-acting muscarinic antagonist; CI, confidence interval; RR, rate ratio; SE, standard error.