Table 4.

Health-related quality of life, healthcare resource use and productivity outcome measures as a function of ‘COPD Risk (undiagnosed)’ status

Outcome	Group	Coefficient (β/ IRR)	P value	Adjusted mean	SE
MCS (SF-12v2)	Controls (ref.)			47.44	0.59
(n = 19,481)	COPD Risk	β: − 3.75	< 0.000	43.70	0.59
PCS (SF-12v2)	Controls (ref.)			47.01	0.49
(n = 19,481)	COPD Risk	β: −4.28	< 0.000	42.73	0.50
Health utility (EQ-5D)	Controls (ref.)			0.86	0.01
(n = 19,481)	COPD Risk	β: −0.08	< 0.000	0.61	0.01
ER visits	Controls (ref.)			0.41	0.03
(n = 18,616)	COPD Risk	IRR: 1.71	< 0.000	0.70	0.06
Hospitalisations	Controls (ref.)			0.11	0.01
(n = 18,616)	COPD Risk	IRR: 2.92	< 0.000	0.31	0.04
HCP visits	Controls (ref.)			2.16	0.11
(n = 18,616)	COPD Risk	IRR: 1.31	< 0.000	2.83	0.15
Respiratory specialist visits	Controls (ref.)			0.17	0.03
(n = 18,616)	COPD Risk	IRR: 1.38	0.008	0.23	0.04
Absenteeism	Controls (ref.)			4.26	0.58
(n = 13,658)	COPD Risk	IRR: 2.46	< 0.000	10.48	1.55
Presenteeism	Controls (ref.)			21.57	1.52
(n = 13,658)	COPD Risk	IRR: 1.74	< 0.000	37.51	2.85
Overall work impairment	Controls (ref.)			24.10	1.66
(n = 13,658)	COPD Risk	IRR: 1.74	< 0.000	42.01	3.13
Activity impairment	Controls (ref.)			20.85	1.12
(n = 18,616)	COPD Risk	IRR: 1.71	< 0.000	35.69	2.02

Note: All models adjusted for gender, age, marital status, education level, household income, employment status, smoking status, alcohol consumption level, body mass index, exercise level, Charlson Comorbidity Index and self-reported asthma diagnosis. Betas (β) were from normal regression models, whereas Incident Rate Ratios (IRR) were from negative binomial models. Models predicting Absenteeism, Presenteeism and Overall Work Impairment are limited to the employed population, while employment status has been dropped as a control variable in these models accordingly. Abbreviations: SF-12v2 Short Form-12 version 2; MCS mental component summary; PCS physical component summary; ER emergency room; HCP healthcare provider; SE standard error