Self-reported sleep quality and acute exacerbations of chronic obstructive pulmonary disease

Figures & data

Table 1 Baseline characteristics

Age, spirometry, and QoL on enrollment (mean ± SD)	PSQI <5 on enrollment (n=513)	PSQI >5 on enrollment (n=581)
Age	66.6±8.2	63.9±8.8*
FEV1 (% predicted)	38.96±15.56	40.04±15.59 (NS)
FEV1/FVC (%)	41.8±12.5	43.1±12.9 (NS)
SGRQ global score	44.4±14.9	55.7±15.7*
SF-36 domains 1–8
Physical function	42.2±24.0	32.6±22.6*
Role physical	44.1±39.9	26.4±36.1*
Role emotional	77.1±36.5	60.8±42.7*
Vitality	54.7±20.7	41.5±18.8*
Emotional well	81.0±14.7	69.8±19.4*
Social function	79.0±22.6	62.8±27.3*
Pain	77.4±23.0	62.1±27.4*
General health	45.6±21.6	35.1±18.7*

Note:

* P<0.0001.

Abbreviations: PSQI, Pittsburgh Sleep Quality Index; FEV₁, forced expired volume in one second; FVC, forced vital capacity; NS, not statistically significant; QoL, quality of life; SGRQ, St George’s Respiratory Questionnaire (lung disease-specific quality of life); SF-36, Medical Outcomes Short Form 36 (general quality of life); SD, standard deviation.

Table 2 Major classes of concomitant medical/psychiatric conditions at baseline associated with worse sleep quality

Disease class, n (%)	PSQI ≤5 on enrollment (n=513)	PSQI >5 on enrollment (n=581)
Baseline history of	n (%)	n (%)
Cardiovascular disease (all)	380 (74)	439 (76) (NS)
High blood pressure	279 (54)	299 (51) (NS)
Coronary artery disease	89 (17)	114 (20) (NS)
Angina/heart attack	97 (19)	170 (29)*
Murmur	35 (7)	49 (8) (NS)
Palpitations, irregular heartbeat	87 (17)	98 (17) (NS)
Valve disease	21 (4)	33 (6) (NS)
Congestive heart failure	37 (7)	69 (12)^+
Blood clots	39 (8)	46 (8) (NS)
Poor circulation (claudication)	49 (10)	90 (15)^+
Other cardiovascular	40 (8)	46 (8) (NS)
GI disease (all)	324 (63)	397 (68) (NS)
Neurologic disease (all)	117 (23)	172 (30)#
Musculoskeletal disease (all)	354 (69)	462 (80)*
Psychiatric disease
Depression	99 (19)	187 (32)*
Anxiety	127 (25)	254 (44)*
Other	15 (3)	37 (6)^+

Notes: Number and proportion of participants with certain major disease classes in good and poor sleepers. Disease categories not associated with differences in sleep quality included cardiovascular and GI disease. Neurologic disease included stroke, headaches, seizure, and other. Musculoskeletal disease included rheumatoid arthritis, gout, osteoporosis, fractures, joint pain, and osteoarthritis. Although for the overall category “cardiovascular disease”, the difference in incidence between good and poor sleepers was NS, certain subcategories of cardiovascular disease showed great proportions among poor sleepers. The table shows all of the cardiovascular disease categories.

# P<0.05

+ P<0.01

* P<0.0001.

Abbreviations: GI, gastrointestinal; NS, not statistically significant; PSQI, Pittsburgh Sleep Quality Index.

Table 3 Pulmonary treatments

Characteristic	n (%)
Oxygen use at home, n (%)	661 (59.3)
Hours of oxygen/day, mean (SD)	15.4 (8.4)
Inhaled medications, n (%)
Glucocorticoids only	57 (5.1)
LABAs only	21 (1.9)
LAMAs only	77 (6.9)
Glucocorticoids and LABAs	229 (20.5)
Glucocorticoids and LAMAs	51 (4.6)
LABAs and LAMAs	53 (4.7)
Glucocorticoids, LABAs and LAMAs	528 (47.3)
No inhaled medications	101 (9.0)

Notes: Number and percent of participants using specific classes of inhaled medications. Oxygen use per day is expressed as mean (SD) hours. The rest are expressed as number (percent of total). None of these inhaled treatments was associated with PSQI that was significantly different for participants using it versus those not using it.

Abbreviations: LABA, long-acting beta agonist; LAMA, long-acting muscarinic agonist; PSQI, Pittsburgh Sleep Quality Index; SD, standard deviation.

Table 4 Concomitant use of noninhaled medications in good and poor sleepers

Medication class, n (%)	PSQI ≤5 on enrollment (n=513)	PSQI >5 on enrollment (n=581)
Antidepressants	114 (22)	230 (40)*
Antipsychotics	7 (1)	25 (4)+
Anxiolytics	52 (10)	117 (20)*
Hypnotics	16 (3)	54 (9)*
Anti-seizure medications	32 (6)	60 (10)#
Opioids	45 (9)	107 (18)*
Dopaminergics	8 (2)	20 (3)#
Randomized to active drug	257 (50)	289 (50) (NS)

Notes:

# P<0.05

+ P<0.01

* P<0.0001.

Abbreviations: NS, not statistically significant; PSQI, Pittsburgh Sleep Quality Index.

Table 5 Exacerbation data

Exacerbations	Good sleep (PSQI ≤5) at baseline (n=513)	Poor sleep (PSQI >5) at baseline (n=581)	Hazard ratio (95% CI)	P-value
Time to first exacerbation median (days, 95% CI)	239 (213–294)	190 (159–225)	0.810 (0.697–0.941)1 0.838 (0.720–0.975)2 0.889 (0.760–1.039)3	0.00581 0.02222 0.13963 Cox proportional
Rate, mean ± SD	1.37±1.72 (unadjusted rates)	1.70±2.79		0.00341 0.01072 0.08583 (negative binomial)

Notes:

1 Unadjusted

2 adjusted for age and sex

3 adjusted for age, sex, neurologic disease, musculoskeletal disease, psychiatric disease, psychiatric medications, anti-seizure medications, opioids, dopaminergics.

Abbreviations: CI, confidence interval; SD, standard deviation; PSQI, Pittsburgh Sleep Quality Index.

Figure 1 Time to first exacerbation by sleep quality at baseline in 1,117 patients with moderate to severe chronic obstructive pulmonary disease (P=0.006).

Abbreviation: COPD, chronic obstructive pulmonary disease.