Cluster Headache Quality of Life scale (CH-QoL) in CCH and ECH.
CH-QoL total score and the 4 sub-domains scores for patients with CCH and ECH are presented in Table 2 and Figure 1. Patients with CCH had a significantly higher CH-QoL total score (Mdn= 71, IQR = 31) than ECH patients (Mdn= 65, IQR = 28, p < 0.01), indicating poorer health related QoL reported by CCH patients. The CH-QoL ‘restriction of daily activities’ score (Mdn = 25, IQR = 11) than ECH (Mdn = 23, IQR = 10.25), p = 0.03) and the CH-QoL ‘mood and interpersonal relationships’ score were also significantly higher for CCH (Mdn = 26, IQR = 17.5) than ECH (Mdn = 20, IQR = 17), p = 0.01). The median scores on the other two subscales ‘pain and anxiety’ and ‘lack of vitality’ subscales of CH-QoL were similar for patients with CCH (Mdn = 7, IQR = 2; Mdn = 15, IQR = 4) or ECH (Mdn = 7, IQR = 2; Mdn = 15, IQR = 6).
Table 2
Impact of chronic (CCH) and episodic (ECH) cluster headache on quality of life assessed on the Cluster Headache Quality of Life scale (CH-QoL).
CH-QoL subscale
|
CCH (SD)
|
ECH (SD)
|
Mdn Diff
|
Wilcoxon (95% CI)
|
p-values
|
r
|
ADL Restriction
|
25 (19-30)
|
23 (17-28)
|
-2
|
1.99 (0.1, 3.0)
|
0.03*
|
0.2
|
Mood & interpersonal relations
|
26 (16-33)
|
20 (11-28)
|
-4.8
|
5.01 (2.9, 7.9)
|
0.01*
|
0.5
|
Pain and anxiety
|
7 (6-8)
|
7 (6-8)
|
0
|
0.00 (-0.1, 0.0)
|
0.67
|
0.1
|
Lack of vitality
|
15 (13-17)
|
15 (11-17)
|
0
|
0.00 (-0.1,1.1)
|
0.20
|
0.2
|
CH-QoL total score
|
71 (55-86)
|
65 (49-77)
|
-6
|
- 2.97 (2.9,12)
|
0.01*
|
0.4
|
ADL=Activities of daily living, CCH= chronic cluster headache, ECH= episodic cluster headache, Mdn= Median, SD= Standard Deviation. |
[insert Figure.1 here]
Figure.1: Box plots depicting the distribution of CH-QoL scale total score and 4 sub-scores for chronic and episodic cluster headache. Black dots represent individual patients. Note: Higher scores on CH-QoL indicate poorer health related quality of life. CH-QoL cluster headache quality of life scale, F1 restriction of activities of daily living; F2 mood & interpersonal relations; F3 pain and anxiety; F4 lack of vitality.
Comparison of patients with CCH or ECH in terms of pain-related and psychosocial variables
The median and standard deviation values of the measures of pain, disability, co-morbid psychiatric symptoms, psychological health and functional well-being for the patients with chronic (CCH) and episodic cluster headache (ECH) are presented in Table 3.
Psychiatric Comorbidities
CCH had significantly higher scores than ECH patients on both HADS anxiety (CCH Mdn=12.0, IQR=7; ECH Mdn = 8.0, IQR= 6.5, Mdn diff = 4.00, p < 0.001) and depression (CCH Mdn =11.0, IQR=7; ECH Mdn = 6.0, IQR= 8, Mdn diff = 5.00, p < 0.001). More than half of CCH patients (56.4%) had anxiety scores of ≥11 on the HADS compared to 33.9% of ECH patients, while 50.9% of the former had HADS depression scores of ≥ 11 compared to 19.3% of the latter. On the SAS, CCH patients had higher apathy scores than ECH patients (CCH Mdn=16.0, IQR=12.0; ECH Mdn= 11.0, IQR= 9.0, Mdn diff = 5.00, p < 0.001), with 56.5% of CCH and 38.4% of ECH patients having clinically significant levels of apathy. CCH patients also had significantly higher BHS scores (CCH Mdn=9.5, IQR=13.0; ECH Mdn= 4.0, IQR= 6.0, Mdn diff = 5.50, p < 0.001). Based on the BHS standardized cutoffs ≥ 9, 52.1% of CCH patients reported moderate-severe hopelessness, while only 23.6% of ECH reported clinically significant levels of hopelessness. Patients with CCH also had significantly higher scores on the GHQ-28 than ECH patients (CCH Mdn=10.0, IQR=16.0; ECH Mdn= 5.0, IQR= 11.0, Mdn diff = 5.00, p < 0.001), considerably exceeding the cut-off score > 4 for this scale. Scores above the cut-off had been reported by 73.2% of CCH and 56.1% of ECH patients, suggesting a high proportion of CCH patients experienced distress due to their headaches.
Disability
Patients with CCH had significantly higher scores on the MIDAS than ECH patients (CCH Mdn=83.0, IQR=139.5; ECH Mdn= 13.0, IQR= 50.0, Mdn diff = 70.00, p < 0.001). Based on the MIDAS disability grades, over half of the total sample of patients reported being severely disabled (56.8%), 8.1% were moderately disabled, 5.3% were mildly disabled, and 29.8% reported minimal disability. Patients with CCH had a significantly higher HIT-6 score than ECH patients (CCH Mdn=65.0, IQR=9.0; ECH Mdn= 62.0, IQR= 12.0, Mdn diff = 3.00, p < 0.001). The HIT-6 classification shows that the cluster headache severely impacts most patients (70.5%), 10.6% reported substantial impact, 11.9% had some impact, while only 7.0% had little or no impact. Patients with CCH had a significantly higher HDI total score than ECH patients (CCH Mdn=76.0, IQR=24.0; ECH Mdn= 64.0, IQR= 24.5, Mdn diff = 12.00, p < 0.001). The HDI total ranges from 4 – 100, indicating profound disability due to their headaches for both CCH and ECH groups.
Functional Well-being
We evaluated how much CH changes the life of sufferers in the professional, social or private domains. In all three domains patients with CCH reported significantly greater impact than patients with ECH (professional, CCH Mdn=10.0, ECH Mdn= 8.0; private, CCH Mdn=9.0, ECH Mdn= 8.0; social, CCH Mdn=9.0, ECH Mdn= 7.0, p < 0.001).
Psychological Health
The degree of perceived stigma associated with CH is also significantly greater in CCH patients than ECH (CCH Mdn=12.0, IQR=5.0; ECH Mdn= 8.0, IQR= 4.0, Mdn diff = 7.00, p < 0.001). Patients with CCH had significantly lower scores than ECH patients on the Rosenberg Self-Esteem scale (CCH Mdn=15.0, IQR=10.0; ECH Mdn= 20.0, IQR= 8.3, Mdn diff = 5.00, p < 0.001), with a higher proportion of CCH patients reporting low self-esteem on the Rosenberg Self-Esteem scale (CCH 44.8% vs. ECH 19.1%, p < 0.001) compared to the group with ECH. Low scores on the Acceptance of Illness scale reflect a lack of acceptance and poor adjustment to their CH. CCH reported significantly lower scores than ECH patients on the Acceptance of Illness scale (CCH Mdn=17.0, IQR=11.0; ECH Mdn= 24.0, IQR= 11.0, Mdn diff = 7.00, p < 0.001) indicating that CCH had poorer adjustment to their CH.
Pain And Pain-related Behaviours
We found no significant difference on the McGill Pain Questionnaire, CCH patients and ECH reported similar pain severity ratings (CCH Mdn=48.0, IQR=21.0; ECH Mdn= 43.0, IQR= 22.5, Mdn diff = 4.00)), suggesting that CCH and ECH experience similar levels of pain.
The PBC total score revealed significant differences in the patients' pain-related coping behaviours, with the CCH patients having higher total scores (CCH Mdn=34.0, IQR=12.0; ECH Mdn= 27.0, IQR= 23.0, Mdn diff = 7.00, p < 0.001), as well as on the helpseeking ( CCH Mdn=3.0, IQR=2.0; ECH Mdn= 2.0, IQR= 3.0, Mdn diff = 1.00, p < 0.001 ) and avoidance ( CCH Mdn=19.0, IQR=12.0; ECH Mdn= 13.0, IQR= 16.0, Mdn diff = 6.00, p < 0.001 ) subscales.
[insert Table.3 here]
Table3. Measures of co-morbid psychiatric symptoms, disability, pain and social function for patients with chronic and episodic cluster headache.
Variable
|
CCH (IQR)
|
Cases %
|
ECH (IQR)
|
Cases %
|
Mdn Diff
|
Wilcoxon test (95% CI)
|
p-values
|
Mood
|
|
|
|
|
|
|
|
HADS-A (cut-off ≥11)
|
12 (8-15)
|
56.4
|
8 (5-11)
|
33.9
|
-4
|
3.01 (2.0, 4.0)
|
0.01*
|
HADS-D (cut-off ≥11)
|
11 (7-14)
|
50.9
|
6 (2-10)
|
19.3
|
-5
|
4.99 (4.0, 6.0)
|
0.01*
|
SAS (cut-off ≥14)
|
16 (10-22)
|
56.5
|
11 (7-16)
|
38.4
|
-5
|
3.99 (2.0,5.0)
|
0.01*
|
BHS (cut-off ≥ 9)
|
9.5 (4-17)
|
52.1
|
4 (2-8)
|
23.6
|
-5.5
|
4.00 (3.0, 6.0)
|
0.01*
|
Minimal (0-3)
|
|
21.3
|
|
47.6
|
|
|
|
Mild (4-8)
|
|
25.6
|
|
28.8
|
|
|
|
Moderate (9-14)
|
|
17.7
|
|
12.7
|
|
|
|
Severe (15-20)
|
|
35.4
|
|
10.9
|
|
|
|
Disability
|
|
|
|
|
|
|
|
MIDAS
|
84 (28-168)
|
|
13 (0-50)
|
|
-71
|
49.99 (34.0,67.0)
|
0.01*
|
HIT-6
|
65 (61-70)
|
|
62 (55-67)
|
|
-3
|
3.99 (2.0, 6.0)
|
0.01*
|
HDI total score
|
76 (64-88)
|
|
64 (52-76)
|
|
-12
|
11.99 (7.9, 15.9)
|
0.01*
|
Pain
|
|
|
|
|
|
|
|
McGill Pain Questionnaire
|
|
Total
|
48 (37-58)
|
|
43 (34-56)
|
|
-5
|
3.99 (0.1, 7.0)
|
0.02
|
Evaluative
|
5 (4-5)
|
|
5 (4-5)
|
|
0
|
0.01 (-0.1, 0.1)
|
0.49
|
Affective
|
9 (6-11)
|
|
9 (5-11)
|
|
0
|
0.99 (-0.01,1.0)
|
0.08
|
Sensory
|
25 (17-31)
|
|
21 (14-28)
|
|
-4
|
2.99 (0.0, 4.9)
|
0.02
|
Miscellaneous
|
10 (7-13)
|
|
9 (7-13)
|
|
-1
|
0.01 ( -0.01, 1.0)
|
0.45
|
Pain report behaviour
|
|
|
|
|
|
|
|
PBC Total
|
34 (26-38)
|
|
27 (13-36)
|
|
-7
|
6.01 (3.9,9.0)
|
0.01*
|
Help Seeking
|
3 (2-4)
|
|
2 (1-4)
|
|
-1
|
1 (0.1, 1.1)
|
0.01*
|
Avoidance
|
19 (12-24)
|
|
13 (4-20)
|
|
-6
|
5.0 (3.0,7.0)
|
0.01*
|
Complaint
|
7 (6-9)
|
|
7 (3-8)
|
|
0
|
0.9 (0.1, 1.1)
|
0.07
|
Psychological Health
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Stigma scale
|
12 (9-14)
|
|
8 (6-11)
|
|
-4
|
3.01 (2.0,4.0)
|
0.01*
|
Life Satisfaction Rate
|
2.6 (1.1-5.1)
|
|
5.5 (2.2-7.8)
|
|
2.9
|
-1.8 (-2.4, -1.1)
|
0.01*
|
Rosenberg Self-Esteem scale
|
15 (10-20)
|
|
20 (16-24)
|
|
5
|
-5.01 (-6.9, -3.9)
|
0.01*
|
Acceptance of illness scale
|
17 (12-23)
|
|
24 (19-30)
|
|
7
|
-6.01 (-8.0, -4.9)
|
0.01*
|
Functional well-being
|
|
|
Social Affected
|
9 (8-10)
|
|
7 (5-8)
|
|
-2
|
1.99 (1.0, 2.0)
|
0.01*
|
Professional Affected
|
10 (8-10)
|
|
8 (6-9)
|
|
-2
|
1.0 (1.0, 2.0)
|
0.01*
|
Private Affected
|
9 (7-10)
|
|
8 (5-9)
|
|
-1
|
1.0 (1.0, 2.0)
|
0.01*
|
BHS= Beck Hopelessness Scale, CCH= chronic cluster headache, ECH= episodic cluster headache, HADS-A & D= Hospital Anxiety and Depression Scale, HDI= The Henry Ford Hospital Headache Disability Inventory, HIT-6= The Headache Impact Test, MIDAS= The Migraine Disability Assessment Scale, SAS= Starkstein apathy scale, Mdn= Median, IQR= Interquartile range, % percent above cut-offs.
Relationship of pain, disability, psychiatric symptoms, and psychological health, with Cluster Headache Quality of Life questionnaire (CH-QoL) total score in CCH and ECH.
Spearman correlational analyses, corrected for multiple comparisons, were performed to explore the relationship between measures of comorbid psychiatric symptoms, disability, pain, and social function and the CH-QoL total scores. The results are presented in Table 4 and Figure 2, 3 and 4.
[insert Table.4 here]
Table.4: Correlations of comorbid psychiatric symptoms, disability, pain, and social function with the cluster headache quality of life scale (CH-QoL) total score in CCH and ECH
|
CCH
|
ECH
|
|
Spearman r
|
p-value
adjusted <0.001
|
Spearman r
|
p-value
adjusted <0.001
|
Mood
|
|
|
|
|
HADS anxiety
|
0.68
|
<0.001*
|
0.58
|
<0.001*
|
HADS depression
|
0.64
|
<0.001*
|
0.52
|
<0.001*
|
SAS
|
0.59
|
<0.001*
|
0.57
|
<0.001*
|
GHQ28 total score
|
0.60
|
<0.001*
|
0.34
|
<0.001*
|
BHS
|
0.62
|
<0.001*
|
0.42
|
<0.001*
|
Disability
|
|
|
|
|
MIDAS
|
0.45
|
<0.001*
|
0.28
|
<0.001*
|
HIT 6
|
0.48
|
<0.001*
|
0.40
|
<0.001*
|
HDI total
|
0.76
|
<0.001*
|
0.72
|
<0.001*
|
Pain
|
|
|
|
|
McGill Pain Questionnaire
|
|
|
|
|
Total
|
0.36
|
<0.001*
|
0.24
|
0.002
|
Evaluative
|
0.29
|
<0.001*
|
0.20
|
0.008
|
Affective
|
0.47
|
<0.001*
|
0.28
|
<0.001*
|
Sensory
|
0.22
|
0.01
|
0.16
|
0.04
|
Miscellaneous
|
0.33
|
<0.001*
|
0.31
|
<0.001*
|
Pain Behavior Checklist
|
|
|
|
|
PBC Total
|
0.46
|
<0.001*
|
0.25
|
<0.001*
|
Help Seeking
|
0.03
|
0.71
|
0.15
|
0.03
|
Avoidance
|
0,49
|
<0.001*
|
0.28
|
<0.001*
|
Complaint
|
0,22
|
0,02
|
0.13
|
0.06
|
Psychological Health
|
|
|
|
|
Stigma total score
|
0.74
|
<0.001*
|
0.57
|
<0.01*
|
Life Satisfaction Rate
|
-0.69
|
<0.001*
|
-0.49
|
<0.01*
|
Rosenberg Self-Esteem scale
|
-0.71
|
<0.001*
|
-0.54
|
<0.01*
|
Acceptance of illness
|
-0.76
|
<0.001*
|
-0.70
|
<0.01*
|
Functioning in life
|
|
|
|
|
Social Life
|
0.59
|
<0.001*
|
0.44
|
<0.01*
|
Professional Life
|
0.54
|
<0.001*
|
0.57
|
<0.01*
|
Private Life
|
0.47
|
<0.001*
|
0.44
|
<0.01*
|
BHS= Beck Hopelessness Scale, CCH= chronic cluster headache, ECH= episodic cluster headache, HADS-A & D= Hospital Anxiety and Depression Scale, HDI= The Henry Ford Hospital Headache Disability Inventory, HIT-6= The Headache Impact Test, MIDAS= The Migraine Disability Assessment Scale, SAS= Starkstein apathy scale.
Psychiatric symptoms
For both CCH and ECH there was a significant positive correlation between CH-QoL total score and HAD anxiety (r = 0.68, p < 0.01; r = 0.58, p <0.01), HAD depression (r = 0.64, p < 0.01; r = 0.52, p <0.01), Starkstein apathy scale (r = 0.59, p < 0.01; r = 0.57, p <0.01), GHQ 28 (r = 0.60, p < 0.01; r = 0.34, p <0.01), and BHS (r = 0.62, p < 0.01; r = 0.42, p <0.01), indicating that an increase of symptoms of anxiety, depression, apathy and hopelessness are associated with higher scores on the CH-QoL, indicating poorer QoL with increased psychiatric morbidity.
[insert Figure.2 here]
Figure.2: Scatter plots representing significant Spearman correlations between CH-QoL total score and measures of psychiatric comorbidities, anxiety (HADS-A), depression (HADS-D), hopelessness (BHS) and apathy (Starkstein apathy scale). Note: CH-QoL total higher scores indicate poorer health related quality of life. HADS-A, HADS-D, BHS, Starkstein higher scores indicate poorer functioning. Individual observation, correlation lines, R values and p-values are showed for chronic cluster headache (CCH, purple circles) and episodic cluster headache (ECH, orange triangles).
Disability
For both CCH and ECH we found a significant positive correlation between the CH-QoL total score and the MIDAS (r = 0.45, p < 0.01; r = 0.28, p <0.01), HIT-6 (r = 0.48, p < 0.01; r = 0.40, p <0.01), and HDI (r = 0.76, p < 0.01; r = 0.72, p <0.01), suggesting that an increase of disability as a result of CH is associated with higher CH-QoL scores and poorer QoL.
[insert Figure.3 here]
Figure.3: Scatter plots representing significant Spearman correlations between CH-QoL total score and measures of disability (MIDAS, HIT-6, HDI). Note: CH-QoL total higher scores indicate poorer health related quality of life. MIDAS, HIT-6 and HDI higher scores indicate higher disability. Individual observation, correlation lines, R values and p-values are showed for chronic cluster headache (CCH, purple circles) and episodic cluster headache (ECH, orange triangles).
Psychological health
For both CCH and ECH, there was a significant negative correlation between CH-QoL total score and Life satisfaction rate (respectively r = -0.69, p < 0.01; r = -0.49, p <0.01), indicating that lower ratings of life satisfaction is associated with poorer quality of life.
For both CCH and ECH, there was a significant negative correlation between CH-QoL total score and the Rosenberg Self-Esteem scale (respectively r = -0.71, p < 0.01; r = -0.54, p <0.01); suggesting that lower self-esteem is associated with poorer QoL.
For both CCH and ECH, there was also a significant negative correlation between CH-QoL total and Acceptance of the Illness (respectively r = -0.76, p < 0.01; r = -0.70, p <0.01), suggesting that lower acceptance of CH is associated with poorer QoL.
Pain behavior
For both CCH and ECH, there was a significant positive correlation between CH-QoL total and PBC total (respectively r = 0.46, p < 0.01; r = 0.25, p <0.01) and also PBC avoidance (respectively r = 0.49, p < 0.01; r = 0.28, p <0.01), suggesting that more pain-related behaviours and particularly engaging in pain avoidance behaviours are associated with lower QoL.
[insert Figure.4 here]
Figure.4: Scatter plots representing significant Spearman correlations between CH-QoL total score and measures of psychological health (Self-esteem, Acceptance of illness, Stigma), pain (McGill pain questionnaire) and pain behavior (PBC total). Note: CH-QoL total higher scores indicate poorer health related quality of life. Self-esteem and Acceptance of the illness higher scores indicate better functioning. Individual observation, correlation lines, R values and p-values are showed for chronic cluster headache (CCH, purple circles) and episodic cluster headache (ECH, orange triangles).