The Manchester Respiratory-related Sleep Symptoms scale for patients with COPD: development and validation

Figures & data

Table 1 Participant characteristics for Stages 1 and 2

Characteristics	Control focus groups (n=12)	COPD focus groups (n=36)	COPD cognitive debriefing (n=9)
Male n (%)	6 (50.0)	18 (50.0)	3 (33.3)
Age mean (SD)	66.0 (7.2)	68.4 (5.2)	66.4 (6.2)
BMI mean (SD)	28.7 (3.7)	27.2 (4.8)	28.5 (7.7)
Comorbidities
Circulatory n (%)	6 (50.0)	24 (66.6)	6 (66.7)
Musculoskeletal n (%)	8 (66.7)	13 (36.1)	4 (44.4)
Endocrine n (%)	2 (16.7)	10 (27.7)	2 (22.2)
Digestive n (%)	0 (0.0)	13 (36.1)	4 (44.4)
Nervous n (%)	4 (33.3)	9 (25.0)	4 (44.4)
Urinary n (%)	0 (0.0)	7 (19.4)	0 (0.0)
Cancer n (%)	0 (0.0)	6 (16.6)	0 (0.0)
Reproductive n (%)	1 (8.3)	3 (8.3)	1 (11.1)
Skin n (%)	0 (0.0)	2 (5.5)	0 (0.0)
Smoking exposure
Current smokers n (%)	1 (8.3)	11 (30.6)	4 (44.4)
Past smoker n (%)	7 (58.3)	25 (69.4)	5 (55.6)
Never smoked n (%)	4 (33.3)	0 (0.0)	0 (0.0)
Pack years n (%)	20.0 (16.7)	40.9 (15.1)	43.7 (8.2)
Respiratory function
COPD duration in years mean (SD)	N/A	7.7 (4.5)	11.2 (4.1)
FEV1 % predicted mean (SD)	N/A	52.1 (18.7)	53.9 (14.1)
FEV1/FVC (%) (SD)	N/A	47.7 (12.3)	46.9 (19.2)
GOLD Grade 1 n (%)	N/A	2 (6.1)
GOLD Grade 2 n (%)	N/A	14 (42.4)	7 (77.8)
GOLD Grade 3 n (%)	N/A	14 (42.4)	2 (22.2)
GOLD Grade 4 n (%)	N/A	3 (9.1)	0 (0.0)
GOLD Grade missing data	N/A	3

Abbreviations: BMI, body mass index; FEV₁ % predicted, forced expiratory volume in 1 second expressed as a percentage of the predicted normal range for the individual; FEV₁/FVC, the proportion of forced expiratory volume in 1 second/forced vital capacity exhaled in the first second of expiration; N/A, not applicable.

Table 2 Participant characteristics for Stages 3 and 4

Characteristics	COPD (n=203)	Controls (n=50)
Male n (%)	128 (63.1)	28 (56.0)
Age (years)	64.8 (7.5)	65.8 (6.9)
BMI (mean (SD))	27.6 (5.2)	27.6 (4.5)
Current smokers n (%)	66 (32.5)	16 (32.0)
Pack years median (IQR)	41 (27–57)	26 (17–36)
COPD duration in years (SD)	7.5 (5.3)
MOS Sleep Scale mean (SD)
Sleep disturbance	39.8 (25.2)	32.2 (27.0)
Short of breath/headache	26.2 (27.9)	5.2 (13.2)
Sleep adequacy	52.3 (27.4)	45.2 (27.2)
Sleep problem index II	41.7 (20.0)	38.3 (3.4)
FACIT-Fatigue mean (SD)	31.5 (13.1)	38.3 (8.4)
SGRQ-C mean (SD)
Total	51.9 (20.9)
Symptom	59.6 (19.5)
Activity	66.2 (27.7)
Impact	40.1 (22.8)
mMRC Dyspnea Scale n (%)
0	20 (9.9)
1	64 (31.5)
2	40 (19.7)
3	56 (27.6)
4	23 (11.3)
FEV1 % predicted (%)	56.6 (18.6)
FEV1/FVC (%)	48.3 (13.4)
GOLD (n=176) n (%)
Grade 1	28 (15.9)
Grade 2	83 (47.2)
Grade 3	50 (28.4)
Grade 4	15 (8.5)

Abbreviations: FEV₁ % predicted, forced expiratory volume in 1 second expressed as a percentage of the predicted normal range for the individual; FEV₁/FVC, the proportion of forced expiratory volume in 1 second/forced vital capacity; mMRC, modified Medical Research Council Dyspnea Scale; MOS, Medical Outcomes Survey; SGRQ-C, St George’s Respiratory Questionnaire-COPD.

Table 3 Details of item removal during hierarchical (n=203) and Rasch (n=178) analysis

Item	Reason for removal	P-value or correlation
Trouble getting to sleep	Unable to discriminate between COPD and controls	0.8*
Interrupted sleep	Unable to discriminate between COPD and controls	0.18*
Difficulty sleeping because of worry	Unable to discriminate between COPD and controls	0.09*
Bad tempered because unable to sleep	Correlation with age	0.01*
Use of COPD medication to help with sleep	Item redundant (item–item correlation)	>0.8**
Night-time use of medication	Item redundant (item–item correlation)	>0.8**
Cough while lying down	Item redundant (item–item correlation)	>0.8**
Difficulty sleeping because of coughing	Item redundant (item–item correlation)	>0.8**
Difficulty sleeping because of chest symptoms	Item redundant (item–item correlation)	>0.8**
Problems with sleeping position	Significant gender difference	0.004*
Difficulty sleeping because of breathlessness	Rasch analysis	<0.01**
Difficulty going to sleep while lying	Rasch analysis	<0.01*

Note:

* denotes P-values;

** denotes correlations.

Figure 1 Association between mean MaRSS and mMRC scores (standard error and CI).

Abbreviations: MaRSS, Manchester Respiratory Sleep Symptoms; mMRC, modified Medical Research Council Dyspnea Scale.

Table 4 Differences in mean item scores on the MaRSS between COPD patients and controls

Final MaRSS items	Mean score (SD) COPD (n=186)	Mean score (SD) controls (n=50)	t-test P-value
I woke up because of breathlessness	1.01 (1.0)	0.20 (0.6)	<0.001
Problems with sleep initiation due to phlegm	0.99 (0.99)	0.28 (0.7)	<0.001
Sleep disruption due to phlegm	1.57 (1.2)	0.84 (0.9)	0.002
Sleep disruption due to coughing	1.04 (1.1)	0.08 (0.4)	<0.001
Sleep disruption due to chest symptoms	1.28 (1.1)	0.32 (0.6)	<0.001
Tiredness	0.91 (0.9)	0.28 (0.6)	<0.001
Disruption of morning activities	0.88 (0.9)	0.12 (0.4)	<0.001
Medication use	2.02 (1.2)	1.36 (1.1)	0.01
Total	9.7 (6.6)	3.5 (4.1)	<0.001

Abbreviation: MaRSS, Manchester Respiratory-related Sleep Symptoms scale.

Figure 2 Comparison between MaRSS and CAT scores for each time point (scores are scaled 0–100) (standard error and CI).

Abbreviations: CAT, COPD Test; E0, day of exacerbation; E2, 2 weeks post-exacerbation; E6, 6 weeks post-exacerbation; MaRSS, Manchester Respiratory Sleep Symptoms; Pre, baseline visit at study entry.

Table S1 Symptom areas produced from focus group discussions

General symptom area	Number of items	Specific symptoms
Respiratory	12	Breathlessness, chest tightness, wheezing, cough, phlegm
Sleeping position	4	Getting comfortable, sitting, laying down
Medication use	4	COPD medication to help with sleep and breathing
Consequences of lack of sleep	6	Tiredness, mood affected

Table S2 Individual fit of the eight-item MaRSS to the Rasch unidimensional model

Item	Residual item fit (±2.5)	Chi-square P-value*	Logit (item severity)**
Item 4: I had difficulty falling asleep because of phlegm (sputum)	0.103	3.582	1.712
Item 17: I woke up because of tightness in my chest	−1.46	6.143	0.534
Item 16: I woke up because of phlegm (sputum)	−0.259	2.429	0.458
Item 1: I woke up because of breathlessness	−2.133	6.288	0.232
Item 9: During the night, I have needed to take puffs/inhalations of a short-acting bronchodilator (eg, Ventolin/Bricanyl)	−0.7	0.948	−0.147
Item 13: I woke up because of coughing	1.208	0.778	−0.371
Item 5: Lack of sleep interfered with my morning activities	1.297	3.138	−0.88
Item 20: I woke up feeling tired	2.06	4.619	−1.536

Notes:

* Chi-square P-value >0.05 meets unidimensional Rasch odel assumptions.

** The last column (logit) shows the mean level of severity for each item. Items are presented in descending order of severity (−ve logit indicates less severe and +ve indicates more severe).

Abbreviation: MaRSS, Manchester Respiratory-related Sleep Symptoms scale.

Table S3 Exacerbation-related change scores for MaRSS and CAT questionnaires

Measurement period	MaRSS (mean difference; P-value)	Cohen’s d effect size	CAT (mean difference; P-value)	Cohen’s d effect size
Baseline – E0*	4.36; P=0.007	0.64	4.21; P=0.006	0.49
Baseline – E2**	0.87; P=0.44 (NS)	0.13	2.09; P=0.14	0.24
Baseline – E6***	0.38; P=0.66 (NS)	0.06	0.71; P=0.54	0.08

Notes:

* E0: Day of exacerbation;

** E2: 2 weeks post exacerbation;

*** E6: 6 weeks post exacerbation.

Abbreviations: CAT, COPD Assessment Test; MaRSS, Manchester Respiratory-related Sleep Symptoms scale; NS, no significant change.