Association Between Hypnotic Use and All-Cause Mortality in Patients with Chronic Obstructive Pulmonary Disease and Insomnia

Figures & data

Table 1 Baseline Characteristics of the Study Population Before and After Propensity Score Matching

	Full Unweighted Cohort			P value	Propensity Score-Weighted Cohort			P value
	Hypnotic Users (n=3585)	Hypnotic Nonusers (n=2174)	SMD		Hypnotic Users (n=1840)	Hypnotic Nonusers (n=1840)	SMD
Age, years	71 (13)	73 (16)	0.17	<0.001	72 (15)	72 (15)	0.02	0.56
Age, n (%)
<50	109 (3)	50 (2)	0.05	0.1	48 (3)	41 (2)	0.02	0.45
50–59	316 (9)	187 (9)	0.007	0.78	176 (10)	160 (9)	0.03	0.36
60–69	1179 (33)	600 (27)	0.12	<0.001	547 (30)	553 (30)	0.007	0.83
70–79	1268 (35)	723 (33)	0.04	0.1	620 (34)	617 (34)	0.003	0.92
≥80	713 (20)	614 (29)	0.19	<0.001	456 (23)	462 (25)	0.007	0.82
Sex, n (%)			0.21	<0.001			0.03	0.37
Male	3351 (93)	1967 (90)			1734 (94)	1721 (94)
Female	234 (7)	207 (10)			106 (6)	119 (6)
Race, n (%)
Caucasians	2925 (82)	1806 (83)	0.04	0.15	1492 (81)	1510 (82)	0.009	0.44
Black	452 (13)	266 (12)	0.01	0.68	249 (13)	230 (13)	0.02	0.35
Hispanic	85 (2)	41 (2)	0.03	0.22	44 (2)	36 (2)	0.03	0.37
BMI, kg/m^2	27.6 (8.7)	27.1 (8.9)	0.1	0.002	27.1 (8.8)	27.3 (8.8)	0.04	0.57
BMI categories, n (%)
Underweight	185 (5)	123 (6)	0.02	0.42	110 (6)	105 (6)	0.01	0.73
Normal	1018 (28)	672 (31)	0.06	0.04	545 (29)	543 (29)	0.002	0.94
Overweight	1042 (29)	640 (29)	0.008	0.76	537 (29)	553 (30)	0.02	0.56
Obese	1134 (32)	607 (28)	0.08	0.003	520 (28)	533 (30)	0.02	0.64
Morbidly Obese	206 (6)	132 (6)	0.01	0.61	128 (7)	106 (6)	0.05	0.14
Tobacco use, n (%)			0.1	0.02			0.006	0.81
Active smoker	1537 (43)	865 (40)			797 (43)	781 (42)
Former smoker	1815 (51)	1134 (52)			954 (52)	973 (53)
Never smoker	233 (6)	175 (8)			89 (5)	86 (5)
Comorbidities, n (%)
Depression	476 (13)	312 (14)	0.03	0.25	254 (14)	261 (14)	0.01	0.74
Hypertension	3156 (88)	1786 (82)	0.17	<0.001	1566 (85)	1560 (85)	0.009	0.78
Cerebrovascular disease	110 (3)	46 (2)	0.06	0.03	47 (3)	42 (2)	0.02	0.59
Diabetes mellitus	1699 (47)	834 (38)	0.18	<0.001	773 (42)	747 (41)	0.03	0.38
MI	478 (13)	261 (12)	0.04	0.14	251 (14)	238 (13)	0.02	0.28
CHF	979 (27)	522 (24)	0.08	0.006	510 (28)	463 (25)	0.06	0.08
CKD	640 (18)	369 (17)	0.02	0.39	342 (19)	332 (18)	0.01	0.67
Cancer	101 (3)	41 (2)	0.06	0.03	38 (2)	39 (2)	0.004	0.91
HIV	41 (1)	21 (1)	0.02	0.52	21 (1)	20 (1)	0.005	0.88
Medications, n (%)
Aspirin	1681 (47)	716 (33)	0.29	<0.001	710 (39)	692 (38)	0.02	0.54
Statins	2624 (73)	1206 (55)	0.38	<0.001	1156 (63)	1171 (64)	0.01	0.61
Beta-blockers	2057 (57)	941 (43)	0.28	<0.001	897 (49)	911 (50)	0.01	0.64
ACEI/ARB	1814 (51)	784 (36)	0.27	<0.001	774 (42)	773 (42)	0.001	0.97
LABA	2187 (61)	1022 (47)	0.29	<0.001	864 (47)	827 (45)	0.03	0.22
LAMA	1900 (53)	891 (41)	0.24	<0.001	798 (43)	786 (43)	0.01	0.68
ICS	1864 (48)	783 (36)	0.18	<0.001	701 (38)	657 (36)	0.06	0.13
AECOPD	614 (17)	321 (15)	0.06	0.02	327 (18)	300 (16)	0.03	0.23

Abbreviations: ACEI/ARB, Angiotensin converting enzyme inhibitors/Angiotensin receptor blockers; LABA-LAMA, Long acting beta-agonist-long acting; ICS, Inhaled corticosteroid; BMI, Body mass index; CKD, Chronic kidney disease; CHF, Congestive heart failure; MI, Myocardial infarction; HIV, Human immunodeficiency virus; AECOPD, Acute exacerbation of COPD.

Figure 1 Selection of the study cohort.

Figure 2 Kaplan–Meier curves for the association between hypnotic use and all-cause mortality in propensity-matched COPD patients with insomnia.

Table 2 Incidence Rate Ratios of All-Cause Mortality for Propensity-Matched Cohorts with Varying Intervals of Follow-Up

Follow-Up	Person-Years	No of Events	IR (95% CI)*	IRR (95% CI)^†	P value
2 years
Hypnotic nonusers	3635	57	15.7 (12.1–20.3)	-	-
BZRAs	2209	77	34.8 (27.9–43.6)	2.2 (1.5–3.2)	<0.001
Non-BZRAs	1282	49	38.2 (28.9–50.6)	2.4 (1.6–3.6)	<0.001
4 years
Hypnotic nonusers	3451	131	37.9 (31.9–45.0)	-	-
BZRAs	1990	102	51.2 (42.2–62.2)	1.35 (1.1–1.7)	0.02
Non-BZRAs	1095	31	28.3 (19.9–40.2)	0.7 (0.5–1.1)	0.14
6 years
Hypnotic nonusers	3165	148	46.7 (39.8–54.9)	-	-
BZRAs	1695	141	83.2 (70.5–98.1)	1.8 (1.4–2.3)	<0.001
Non-BZRAs	952	48	50.4 (38.0–66.9)	1.1 (0.7–1.5)	0.64

Notes: *IR, Incidence rate per 1000 person-years; †IRR, Incidence rate ratio.

Abbreviation: BZRA, Benzodiazepines and non-benzodiazepine benzodiazepine receptor agonists.

Table 3 Stratified Analysis of the Risk of All-Cause Mortality for Propensity-Matched Cohort Associated with Hypnotic Use

Patient Age (Years)	Hypnotic Users			Hypnotic Nonusers			HR (95% CI)^†
	Person-Years	No of Events	IR (95% CI) *	Person-Years	No of Events	IR (95% CI) *
<65	3205	143	44.6 (37.9–52.6)	3499	141	40.3 (34.2–47.5)	1.14 (0.91–1.45)
≥65	9318	617	66.2 (61.2–71.6)	11,145	624	55.9 (51.8–60.6)	1.25 (1.12–1.39)

Notes: *IR, Incidence rate per 1000 person-years; †HR, Hazard ratio.

Figure 3 Kaplan–Meier curves for the association between hypnotic categories and all-cause mortality in propensity-matched COPD patients with insomnia.