Journal of the Neurological Sciences 348 (2015) 60–66
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Chronic migraine with medication overuse: Association between disability and quality of life measures, and impact of disease on patients' lives Alberto Raggi a,⁎, Silvia Schiavolin a, Matilde Leonardi a, Ambra Mara Giovannetti a, Gennaro Bussone b, Marcella Curone b, Paola Di Fiore b, Licia Grazzi b, Susanna Usai b, Domenico D'Amico b a b
Neurology, Public Health and Disability Unit, Neurological Institute “C. Besta” IRCCS Foundation, Milan, Italy Headache and Neuroalgology Unit, Neurological Institute “C. Besta” IRCCS Foundation, Milan, Italy
a r t i c l e
i n f o
Article history: Received 11 August 2014 Received in revised form 31 October 2014 Accepted 3 November 2014 Available online 8 November 2014 Keywords: Chronic migraine Medication overuse Health-related quality of life Disability Detoxification Withdrawal
a b s t r a c t Patients with chronic migraine with medication overuse (CM-MO) have decreased quality of life (QoL) and increased disability: the degree to which these outcomes are connected to disease severity and the pattern of MO towards disease severity are unclear. Patients under withdrawal were administered the Migraine Disability Assessment (MIDAS), the WHO Disability Assessment Schedule (WHODAS), and the Migraine-Specific Quality of Life Questionnaire (MSQ). They overused NSAIDs, triptans, NSAIDs and triptans, and other drugs (ergotamine, caffeine, opioids/barbiturates). We calculated the correlations between MIDAS, WHODAS, and MSQ; compared WHODAS to normative scores; compared MIDAS, WHODAS, and MSQ in patients with different CM-MO severity; and run a logistic regression to predict CM-MO severity based on overused drugs. One hundred ninety-four patients were enrolled: correlations between WHODAS, MSQ, and MIDAS were moderate; wide differences on WHODAS against normative were found; and no trend was found across severity groups. Compared to triptans overusers, patients overusing NSAID and other drugs had higher odds of severe CM-MO. Coupling different disability measures with QoL assessment offered different insights on the lived experience of CM-MO. Future studies are needed to clarify the relationship between overused drugs and CM-MO severity: we added evidence that NSAIDs do not have protective effect in high-frequency CM-MO. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Chronic migraine (CM) is characterized by headaches occurring 15 or more days per month and can be considered as a negative evolution of the migraine course in a subgroup of subjects [1]. CM prevalence in population studies ranges from 1.4% to 4% and from 2.5% to 3% of migraineurs (14% in clinical samples) progress to CM each year [2–5]. Several factors may play a role in the progression from episodic migraine (EM) to CM, such as lifestyle, comorbid conditions, metabolic dysfunction, mood, genetic terrain, and medication overuse (MO) [1,6, 7]. CM is strongly associated with the overuse of non-steroidal analgesic drugs (NSAIDs) as these drugs can widely be accessed as over the counter medications: the fact that frequent use of symptomatic drugs is a risk factor for progression of migraine marks an important difference with other pain-related conditions [8]. Indeed all acute drugs may be associated with chronification, but it is not clear if and how the disease's ⁎ Corresponding author at: Neurology, Public Health and Disability Unit, Neurological Institute C. Besta, IRCCS Foundation, Via Celoria 11, 20133, Milan, Italy. Tel.: + 39 02 2394 3105 (2521); fax: +39 02 2394 2442. E-mail address: [email protected] (A. Raggi).
http://dx.doi.org/10.1016/j.jns.2014.11.004 0022-510X/© 2014 Elsevier B.V. All rights reserved.
features are related to the different overused drugs [9] and the mechanisms for chronification are not completely understood yet. The pattern of overused medication varies across time and countries [10]. A community pharmacy-based study showed that one-fourth of patients overused acute medications and that combination of three or more agents was more frequent in overusers compared to nonoverusers: overusers also had higher headache-related, headaches frequency, and higher pain intensity [11]. Findings on chronification process connected to the class of overused medication are however partial and contrasting. Some evidence exists on the increased risk of chronification in patients using triptans, but only in those with a baseline frequency around 10–14 headaches/month, and of the protective effect of NSAIDs in low-frequency migraine sufferers [12,13]. Barbiturates and opioids have also been found to induce migraine progression, with a dose-dependent effect [12]. Recent reviews showed that health-related quality of life (HRQoL) is significantly impaired in persons suffering from chronic forms of headache, compared to non-headache sufferers, as well as compared to those with EM, and that this negative impact is more evident in those suffering from CM, particularly when MO is present [14,15]. CM causes physical health and emotional problems and determines increased disability,
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
measured in terms of absenteeism and days with reduced workplace effectiveness; therefore, it has remarkable consequences in terms of increased disease cost, which is three-fold that of EM (3561€/year vs 1222 €/year) [16]. Previous studies addressing disability in chronic migraine with medication overuse (CM-MO) [17–21] generally used the Migraine Disability Assessment (MIDAS) [22]. However, the use of MIDAS in CM patients poses some problems. First, the highest severity grade corresponds to a score N 21, which is intuitively overrepresented among patients reporting more than 45 headaches in the previous three months. In addition, the conceptualization of disability behind MIDAS is highly dependent on headaches frequency and does not account for other features of CM such as reduced mood or obsessive or anxiety disorders but also for the severity of pain, which are likely to impact on functioning in social and workplace roles, particularly as far as interpersonal relationships [17,23,24]. A model that recognizes disability in a multidimensional way is therefore needed. Such disability model should correspond to that endorsed by the World Health Organization (WHO) with the International Classification of Functioning, Disability and Health (ICF) [25]. The understanding of the possible relationships between disability—particularly when viewed as a multidimensional phenomenon—and HRQoL in patients with CM-MO may enhance the understanding of this disorder and help clinicians and researchers in choosing the most appropriate patient-reported outcome measure (PROMs) in specific studies. The WHO also produced an ICF-based disability assessment, the WHO Disability Assessment Schedule second version (WHODAS) [26]: previous studies with neurological patients (namely, EM, myasthenia gravis, epilepsy, stroke and Parkinson's disease) showed that it is suitable to address the increased disability level of patients as well as a moderate association with HRQoL assessments [27–31]. To our knowledge, the relationship between disability, assessed with an instrument like the WHODAS, and HRQoL measures in CM-MO was never addressed. The primary aim of this study was to assess the relationships between two disability measures, MIDAS and WHODAS, and between them and patients' HRQoL, in order to assess whether these different tools cover similar constructs or offer the possibility to evaluate different facets of CM-related impact: we expect to expand to the sample of CM-MO patients the findings of a previous study on EM patients, i.e., that MIDAS and WHODAS are only partially correlated and that the correlations with HRQoL are mild to moderate. The secondary aim was to address the impact of CM-MO on patients' disability. This was performed with three different procedures: (a) the evaluation of the impact of CM-MO on patients' disability compared to the general population, with the hypothesis that CM-MO patients report worse scores at WHODAS; (b) the evaluation of the extent to which differences in PROMs scores correspond to different severity profiles—based on frequency and intensity of headaches—with the hypothesis that PROMs scores cannot be completely explained by differences in CM-MO severity; and (c) the evaluation of the relative risk of having a more severe disease profile accounting for the type of overused drug, with the hypothesis that those overusing NSAIDs have are more likely to have the most severe profile. 2. Methods
61
during administration of the protocol. In any case, they were allowed to postpone it if they had an headache. The study was approved by the institute's ethical committee, and each patient signed an informed consent form prior to data collection. All enrolled patients completed the Migraine-Specific Quality of Life Questionnaire Version 2.1 (MSQ) [33] to evaluate HRQoL, the WHODAS [26,34], and the MIDAS [22,35] to evaluate disability. Headache frequency and average pain intensity were taken from the last two MIDAS items. Mood state was assessed with the Beck Depression Inventory [36]. MSQ is based on 14 items that investigate the impact of headaches on patients HRQoL during the previous 4 weeks. The items are grouped into three subscales: Role Restriction (RR), 7 items assessing how migraines limit one's daily social and work-related activities; Role Prevention (RP), 4 items assessing how migraines prevent social and workrelated activities; and Emotional Function (EF), 3 items assessing the emotions associated with headaches. Each item refers on how frequently headaches limits the activity under consideration and is rated on a 6-point scale from none of the time to all of the time. Each subscale has a 0–100 score, with lower scores indicating poor HRQoL. MSQ has been used as an outcome measure in clinical trials, mostly on preventive medications [37–41]: however, its use in research on patients with CM or other types of chronic headaches is limited [37–39]. In its Italian version, it has been used in one observational study in which patients with different type of migraine, of whom 2.5% had CM-MO, were evaluated [42]. Recently, three studies confirmed its validity in CM patients [43–45]: however, the presence of MO was explicitly excluded in the first study, which is based on the International Classification of Headache Disorders, 2nd version criteria for CM [46]; in the second study, only 60% of patients had concurrent MO; in the third study, all patients had CMMO. Therefore, the use of MSQ in patients with CM-MO is restricted to two studies [38,45]. WHODAS is a 36-item ICF-based disability assessment tool that examines the difficulties experienced by an individual due to a given health condition. Six domains are taken into account: understanding and communicating, getting around, self-care, getting along with people, life activities (divided into household and work), and finally participation in society. Patients are required to answer questions regarding how much difficulty, due to their health condition, they experienced during the previous 30 days. Answers are rated on a 5-point scale, from no problems to complete problems/cannot do the activity. Both total and subscale scores are available, ranging from 0 to 100, with higher scores reflecting greater disability. Total score can be calculated also on the basis of 32 items, excluding those related to work activities if the respondent is not currently employed. MIDAS is composed of 7 questions referred to the preceding 3 months: the first 5 investigate the influence of headaches on paid and school work, household work, and on leisure/family/social duties. MIDAS score is calculated by adding the individual scores of the first 5 questions and indicates the number of days in which migraine interfered with these activities. According to given intervals, 4 disability grades can be calculated: minimal (0–5), mild (6–10), moderate (11– 20), and severe (≥21) disability. BDI-2 is composed of 21 items, each rated on a 0–3 scale, that address cognitive and somatic-affective features of depression. Total score range is 0–63, with higher scores reflecting higher depressive mood.
2.1. Patients, procedure, and measures In this cross-sectional observational study, adult patients with CMMO according to Silberstein's criteria [32] were consecutively recruited at the Headache Centre of the Neurological Institute C. Besta of Milan between June 2011 and December 2012. Patients were enrolled on occasion of inpatient withdrawal treatment: the questionnaires were provided on the second or third day of hospitalisation. There are two reasons for this: first, to enable physicians to evaluate patients' eligibility; second, to make it more likely that patients were headache-free
2.2. Definition of medication overuse categories The different types of MO were defined as follows: overuse of nonsteroidal anti-inflammatory drugs (NSAIDs), when NSAIDS were used for at least 15 days/month; overuse of triptans, when any drug of this class was used for at least 10 days/month; overuse of both NSAIDs and triptans, when NSAIDs were used for at least 15 days/month and triptans for at least 10 days/month; overuse of ergotamine, caffeine,
62
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
Table 1 Spearman's rank correlations between WHODAS, MIDAS, MSQ, headaches frequency and average pain intensity.
WHODAS—summary score MIDAS Headaches frequency Pain intensity BDI-2
MSQ-RR
MSQ-RP
MSQ-EF
MIDAS
Headaches frequency
Pain intensity
BDI-2
−0.449* −0.569* −0.209 −0.355* −0.308*
−0.504* −0.510* −0.151 −0.359* −0.300*
−0.438* −0.289* −0.110 −0.262* −0.494*
0.417* NA 0.242* −0.161 0.328*
−0.100 NA NA 0.011 0.014
0.307* NA 0.NA NA 0.213
0.587* NA NA NA NA
Note. *Spearman's rank correlations significant at P b 0.0021; unmarked correlations are not significant; NA = not applicable. WHODAS = World Health Organization Disability Assessment Schedule; MSQ = Migraine-Specific Quality of Life Questionnaire; MSQ-RR = MSQ Role Restriction; MSQ-RP = MSQ Role Prevention; MSQ-EF = MSQ Emotional Function; MIDAS = Migraine Disability Assessment Score.
opioids/barbiturates, when any drug of this class was used for at least 10 days/month, alone or together with NSAIDs or triptans.
2.3. Data analysis Spearman's rank correlation analysis was used to assess the relationships between MIDAS, WHODAS summary score, and MSQ scales. Correlations were considered weak for coefficient values b 0.29, moderate for values between 0.30 and 0.59 and strong for values N 0.60. Correlations were also carried out between the three questionnaires' headaches frequency, pain intensity, and BDI-2 score. As 24 multiple correlations were carried out, we applied Bonferroni correction, and significance was set with Type I error level at α = 0.0021 and 2-tailed testing. With regard to secondary aims, the impact of CM-MO on patients' disability was assessed comparing WHODAS scores against normative Italian scores reported by Federici and colleagues [34] referred to the general population (271 “normal adults”, mean age 30 years, SD 9.9). Comparison against normative Italian values was made using onesample t-test and Hedges's g as a measure of effect size [47] and was calculated as the difference between our sample's means and the means observed by Federici divided by the pooled SD: g N 0.5 was considered to reflect medium effect size, and g N 0.8 to reflect large effect size. Comparison was not made with Life activities scale of the WHODAS because in the paper of Federici and colleagues [34], no distinction between household and work activities was made. As 18 six comparisons were carried out, we applied Bonferroni correction and significance was set with Type I error level at α = 0.0083. Disability and HRQoL were compared in patients with different CM severity. To address CM severity, we run an exploratory K-means cluster analysis using MIDAS questions regarding the total number of days with headaches in the last 3 months and the average pain intensity as grouping variables, and one-way ANOVA to evaluate the contribution of the two variables in the identification of clusters. We predefined 3 clusters without specifying initial cluster centroids and decided to have a maximum of 10 iterations for defining cluster centers and membership. Kruskal– Wallis H test, a nonparametric analog of one-way analysis of variance, was performed to compare WHODAS 2.0 summary score, MIDAS, and MSQ in patients with different migraine severity. Bonferroni correction Table 2 One-sample t-test between study and normative data for WHODAS. WHODAS Scale
Sample data
Normative data
Effect Size
t-test
Understanding and communicating Getting around Self-care Getting along with people Household activities Work activities Participation in society Summary score
31.8 (18.8)
11.84 (12.96)
1.27
t = 14.8
28.1 (22.6) 10.8 (15.7) 21.6 (19.6) 43.0 (26.6) 38.9 (24.6) 37.4 (15.5) 31.2 (13.8)
7.07 (13.51) 3.53 (8.00) 12.57 (16.97) NA NA 12.12 (13.86) 12.95 (11.77)
1.18 0.61 0.50 NA NA 1.74 1.44
t= t= t= NA NA t= t=
13.0 6.5 6.5
22.8 18.5
Note. Values are reported as means (SD). NA, not applicable. P-value was b 0.001 for all t-tests.
was used to control for multiple comparisons: as 5 comparisons were performed, significance was set with Type I error level at α = 0.01. To assess the relative risk of having the most severe disease profile, logistic regression was carried out entering the most severe disease severity as dependent variable, pattern of overused drug (NSAIDs, triptans, NSAIDs and triptans, other drugs, i.e., ergotamine, caffeine, and opioids/ barbiturate) as independent variables, and sex, age, employment, marital status, and educational level as covariates. Overuse of triptans was set as reference for independent variable. The −2 log-likelihood difference with chi-square was used to test the difference between the full model and the model based on the intercept only. Nagelkerke pseudo R2 was used to assess the whole explanatory power of the model. Data analysis was performed using PAWS statistic 18.0. 3. Results A total of 213 patients were screened, two declined to participate and 17 did not match inclusion criteria; therefore, 194 patients, of whom 160 females (82.5%), were enrolled. Age range was 18–75 years (mean age 43.9 years, SD 12.0 years); 124 patients (63.9%) completed high school or had academic education level, 143 (73.7%) were employed or students, and 139 (71.6%) were married or cohabitating. The mean number of days with headaches in the previous three months was 56.3 (SD 23.7) and average pain intensity was 7.8 (SD 1.3). Mean MIDAS score was 82.7 (SD 59.9), and in 175 patients (90.2%), it was higher than 21, thus indicative of severe disability. The mean number of workdays lost on a monthly basis was 4.3 (SD 6.1), and the mean number days of housework lost was 6.1 (SD 6.2). The mean MSQ-RR was 32.7 (SD 17.3), the mean MSQ-RP was 47.6 (SD 22.0), the mean MSQ-EF was 42.2 (SD 24.9), and the mean WHODAS summary score was 31.2 (SD 13.8). As far as different classes of overused drugs, 57 patients (29.4%) overused NSAIDs, 42 (21.7%) triptans, 60 (30.9%) both NSAIDs and triptans, and 35 (18%) overused ergotamine, caffeine, or opioids/barbiturates. The correlations between the scores obtained at different tools are reported in Table 1. Correlations between WHODAS, MSQ, and MIDAS were generally moderate; MIDAS correlated to frequency of headaches but not to pain intensity, while WHODAS correlated to pain intensity and not to headaches frequency. BDI-2 was more strongly correlated with WHODAS than with MIDAS; correlations with MSQ subscales were stronger with EF than with RR and RP. Table 2 shows results on the impact of CM-MO on ability to function. For all WHODAS scales, the scores reported by patients with CM-MO were significantly lower compared to normative data, with Hedges's g reporting medium to large effect sizes. All one-sample t-tests were significant a P b 0.001 level. Cluster analysis is reported in Table 3. The three severity clusters were identified only on the basis headaches frequency, while the average pain severity did not provide any significant contribution to clusters' definition, as shown by one-way ANOVA analysis. Table 4 reports the results of Kruskal–Wallis H test for MIDAS score, WHODAS summary score and MSQ in patients from different severity clusters: statistical difference was found only for MIDAS. Table 5 reports the results of logistic regression. The overall explicative power of the model was 11.4%, and − 2 log-likelihood difference
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
63
Table 3 K-means cluster analysis and Kruskal–Wallis H test: patients divided into three groups according to migraine frequency and pain intensity.
Migraine frequency Pain intensity
Cluster A (No. 53)
Cluster B (No. 74)
Cluster C (No. 67)
F statistic
28.6 (26.5–29.7) 7.9 (7.5–8.3)
51.4 (49.5–53.3) 7.6 (7.4–7.9)
85.0 (83.4–86.6) 7.9 (7.6–8.2)
976.8 (P b 0.0001) 0.86 (P = 0.425)
Note. For each group, cluster center and 95% confidence interval are reported. Cluster A: low CM-MO severity. Cluster B: intermediate CM-MO severity. Cluster C: high CM-MO severity. Migraine frequency: days with migraine in the last 3 months. Pain intensity: rated on a 0–10 scale. Bonferroni post hoc test is significant (P b 0.0001) for migraine frequency between each pair; mean difference between severe and moderate cluster was 33.6, between severe and mild cluster was 56.4, and between moderate and mild cluster was 22.8.
analysis show that the final model outperforms the intercept-based model. Patients overusing NSAIDs only or other drugs had higher likelihood of having the most severe disease pattern (approximately 2.5 and three-fold compared to the reference of patients overusing triptans). 4. Discussion With respect to the primary aim of this study, our results showed that MIDAS and WHODAS were significantly and moderately correlated and that the correlations with MSQ and BDI-2 were moderate too, thus showing that the underlying concepts they are intended to measure are different. With regard to secondary aims, our results showed the strong impact of CM-MO on patients' disability, while we failed in determining a clear increase of disability and decrease in HRQoL when differences were assessed in patients with different CM-MO severity: differences were found only for MIDAS. Finally, patients in the most severe cluster (i.e., with more frequent headaches) were more likely to be overusers of NSAIDs or of ergotamine, caffeine, and opioids/barbiturates. The two disability measures displayed a moderate correlation, thus giving confirmation of the hypothesis that the constructs behind them are not overlapping. This result is largely consistent with that previously found in a sample of patients with EM [27], that was similar for age (the present sample was 0.4 years older) and female prevalence (−2.8% in the present sample), but displayed clear differences at MIDAS: score N21 was in fact reported by two-fold of the patients in this study that in the previous on EM. Therefore, our results reinforce those previously reported on EM patients: MIDAS and WHODAS measure different facets of the same construct, as the first provides a measure based upon frequency of migraine-related problems while the second provides a measure of the intensity of disease-related problems. Such a hypothesis was also reinforced by the different pattern of correlations between the two disability measures and patient-reported headaches frequency and intensity: MIDAS correlated to headaches frequency, and WHODAS to average pain intensity. Mild to moderate associations were also found between disability and HRQoL scores as well as between these scales and mood state. The association was stronger for scales referred to restriction and prevention of social activities rather than for the issues of emotions associated to headaches, thus showing that the concepts behind the three assessment instruments were neither overlapping nor transposable. The same kind of results were also found in the paper on EM, where the Medical Outcome Survey 36-Item Short-Form Health Survey (SF36) [48] was used. The correlations between SF-36 and MSQ were reported in two studies and ranged between 0.19 and 0.38 [33] and between 0.26 and 0.35 [49]. In the present study, the association between the WHODAS and MSQ scale exploring emotional components (MSQ-EF) was weaker than that with MSQ-RR and MSQ-RP: again, this is similar to what was found in the previous study on EM patients, where WHODAS correlated more with physical component than with mental component of SF-36 [27]. The opposite was instead found for the correlations between MIDAS and MSQ, which was stronger with MSQ-RR and MSQ-RP than with MSQ-EF: this can be explained considering that MIDAS specifically reflects the impact of migraine on productivity/performance in daily activities, while MSQ-EF assesses the emotions associated with headaches. Here we also added a measure of mood state: results show that depression correlated more with
disability when measured with WHODAS than when measured with MIDAS and that—as expectable—it correlated more with MSQ-EF than with MSQ-RR and MSQ-RP. Taken together, these results suggest that the use of both HRQoL and disability assessment instruments is important to couple information on aspects connected to physical health (e.g., mobility, self-care), relationships and participation to social situations, with those connected to emotional issues. Patients with chronic forms of headaches display an increased burden connected to all these aspects, as shown in several studies [17,24,50], that should be measured to understand patients perspective and to assess the outcomes after treatments. Our results also point out that WHODAS addresses in a better way than MIDAS the issues connected to mood state. Our results confirmed the relevant impact of CM-MO on patients' daily lives, with a relevant burden on the individual in terms of difficulties in dealing with all the domains of life, from those most frequently investigated—such as employment, where the number of workdays lost by CM-MO patients is more than six-fold those observed in EM patients—to those less frequently considered, such as mobility and selfcare. In a previous study with age and gender-matched patients with EM and CM-MO, we showed that the latter have higher disability [17], and the scores observed in the present paper at WHODAS are indicative of a higher disability if contrasted to those reported in a sample of EM patients [27]. However, similarly to what observed in these patients, severity of problems in WHODAS mobility domain (e.g., getting out of your home or walking a long distance) was similar to that of problems with understanding and communicating (e.g., generally understanding what people say, or starting and maintaining a conversation) and was higher than problems with getting along with people (e.g., dealing with friends, or dealing with unknown people). Moreover, our patients reported more problems in activities connected to participation in social activities (e.g., joining community activities, consequences of health problems on family members) than in relational situations. This analysis of WHODAS scales suggests that social situations may represent the context in which persons suffering from CM-MO experience emotional difficulties, while interpersonal interactions do not seem to be a problematic issue per se. Similar results were found, in addition to EM patients, also in other patients suffering from myasthenia gravis and controlled epilepsy [27–29]: the fact that these diseases are “invisible” to other persons might explain this similarity. Patients are likely to suffer in interpersonal situations because they may find little understanding by relevant others on their conditions and on the difficulties associated to them. These results, however, need to be read also in consideration of the different average age of this sample (43.9, SD 12.0) and of the Italian validation sample (30, SD 9.9), which might have a confounding effect. A recent population study showed that younger age was associated to lower odds of severe disability [51], which reflects the life span accumulation of health conditions (i.e., multimorbidity) that is associated with ageing [52]. When we addressed the WHODAS, MIDAS and MSQ scores in patients with different degree of disease severity, we failed to find significant differences other than in MIDAS, that is, however, based on headaches frequency, like the three clusters. A further proof to this is that in our study MIDAS score correlated, although weakly, with headaches frequency. Less surprising is the absence of significance found for WHODAS and MSQ as they did not correlate to headaches frequency.
64
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
Table 4 Kruskal–Wallis H test for MIDAS, WHODAS, and MSQ scales in patients with different disease severity.
MIDAS WHODAS MSQ-RR MSQ-RP MSQ-EF
Cluster A (No. 53)
Cluster B (No. 74)
Cluster C (No. 67)
Kruskal–Wallis H test
65.8 (54.5–77.2) 33.1 (28.8–37.3) 37.7 (33.0–42.4) 51.8 (45.7–57.9) 47.7 (40.5–55.0)
72.1 (61.6–82.5) 31.4 (28.4–34.5) 32.1 (28.7–35.4) 48.6 (44.1–53.1) 42.1 (36.7–47.5)
107.8 (89.0–125.5) 29.6 (26.4–32.8) 29.4 (24.7–34.2) 43.1 (37.3–49.0) 38.1 (32.0–44.2)
P P P P P
= = = = =
0.005 0.511 0.014 0.084 0.229
Note. For each group and scale, mean and 95% confidence interval are reported. MIDAS = Migraine Disability Assessment Score; WHODAS = World Health Organization Disability Assessment Schedule; MSQ-RR = MSQ Role Restriction; MSQ-RP = MSQ Role Prevention; MSQ-EF = MSQ Emotional Function.
Taken as a whole, the results of this analysis show that the burden caused by CM-MO is not only due to the severity of the headache disorder per se. Rather, it is possible to presume that the overuse of acute medications and the emotional burden associated to such a disabling condition play a role in the definition of its outcome: in fact, the score observed at MSQ-EF was 20.7 lower that that found by Cevoli and colleagues [42] in a sample of patients with migraine, of whom only 2.5% had CM-MO, and 6.1 lower compared to a study of Bagley and colleagues [43] that was based on a sample of patients without MO. Finally, our analysis of the relative risk of being included in the most severe cluster according to the overused medication showed that patients overusing NSAIDs only and those overusing ergotamine, caffeine, and opioids/barbiturates had higher odds of having a more severe CM compared to those overusing triptans, while no increased risk was found for patients overusing both NSAIDs and triptans. Previous research on the impact of different drugs categories on CM severity is lacking. It is known that NSAIDs may exert a sort of protective effect that, however, is restricted to low-frequency headaches and has been observed in the transition between episodic and chronic migraine [12, 13], while other studies point out the strongly risk of chronification associated with the overuse of NSAIDs and of opioid-containing products, with a dose-dependent effect [12]. Other reports showed that patients overusing ergotamine and NSAIDs typically have a daily tension-typelike headache, whereas patients with triptan-induced overuse are more likely to describe a migraine-like headache, i.e., CM [9]. This aspect deserves further research considering that NSAIDs are the most commonly overused drugs also because they are easily available as overthe-counter medications [8] and that opioids are associated to an increased risk of migraine chronification [53] may lead to abuse and dependence [54] and might be the result of an inappropriate long-term use of an emergency department prescription [55]. Moreover, the fact that patients overusing both NSAIDs and triptans did not have an increased risk of higher disease severity, compared to those overusing triptans only, might suggest a sort of “protective” effect of consuming a kind of drug that is not available as an over-the-counter medication, and is therefore more likely to be under the control of a clinician. The main limitation is that sample, entirely derived from a single specialty center, is likely to represent the most severe spectrum of CM,
Table 5 Logistic regression analyses for the association between overused drugs class and inclusion in the higher severity cluster.
as all included patients had very high headache frequency and were overusing symptomatic medications. It is also possible that the context of a structured withdrawal might have influenced the scores of disability and HRQoL instruments: thus, our results cannot be automatically generalized to all CM patients. Second, the cross-sectional research design does not allow us to define causal relationships between clinical data and patient-reported outcomes. Moreover, a “restriction of range” issue is also to be taken into account: in fact, as patients had a severe clinical condition, it is very likely that the range of reported problems and the scores at outcome measures chiefly reflected the upper range of WHODAS and MIDAS as well as the lower range of MSQ scores. Therefore, caution is recommended in the interpretation of correlation levels, which might be attenuated also in consideration of the restricted scores range. Third, the impact of CM over patients' disability should be evaluated considering the higher average age of this sample compared to the Italian validation sample. 5. Conclusions In conclusion, we evaluated disability and HRQoL in patients with CM-MO, showing a remarkable impact of the disease on such dimensions of patients' lives. Coupling different disability measures, such as the WHODAS and the MIDAS together with an HRQoL tool such as the MSQ may offer different insights on the lived experience of CM-MO: these measures in fact represent non-transposable psychosocial facets. MIDAS and WHODAS reflect different aspects of disability, connected to the two main features of the disease, i.e., the frequency of headaches and pain intensity, respectively. The inclusion of an HRQoL measure enabled to add information on emotional aspects connected to CM-MO, which are remarkable and not captured by MIDAS. On the other hand, WHODAS showed that persons suffering from CM-MO experience emotional difficulties in social situations, while interpersonal interactions do not seem to be problematic per se: it is possible to presume that the high frequency of headaches that force CM-MO patients to avoid social situation might play a role in the increased emotional burden of these patients. Future studies are needed to shed light on the complex relationship between overused drugs and CM-MO severity: our study seem to add evidence to the fact that NSAIDs overuse does not have protective effect in high-frequency CM and poses some elements for debate on use of opioids/barbiturates, caffeine, and ergotamine agents. Conflict of interest
Parameters estimates
Intercept Triptans NSAIDs + Triptans NSAIDs Other drugs
B (SE)
OR (95% CI)
P
−3.26 (1.00) – 0.16 (0.47) 0.94 (0.46) 1.12 (0.51)
– – 1.17 (0.46–2.97) 2.56 (1.04–6.32) 3.06 (1.13–8.30)
0.001 – 0.733 0.042 0.028
Note. NSAIDs = non-steroidal anti-inflammatory drugs; other = ergotamine, caffeine, opioids/barbiturates; OR = odds ratio; 95% CI = 95% confidence interval; df = degrees of freedom. Model based on intercept only −2 log likelihood = 240.95; final model −2 log likelihood = 224.16; chi-squared = 16.70, df = 5, P = 0.005. Nagelkerke pseudoR2 = 0.114.
The authors declare that they have no conflict of interest. Acknowledgments This paper is based on the results of DIS.CHRONIC project, sponsored by the Neurological Institute C. Besta IRCCS Foundation. References [1] Bigal ME, Lipton RB. Migraine chronification. Curr Neurol Neurosci Rep 2011;11: 139–48. http://dx.doi.org/10.1007/s11910-010-0175-6.
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66 [2] Manack AN, Buse DC, Lipton RB. Chronic migraine: epidemiology and disease burden. Curr Pain Headache Rep 2011;15:70–8. http://dx.doi.org/10.1007/s11916010-0157-z. [3] Natoli JL, Manack A, Dean B, Butler Q, Turkel CC, Stovner L, et al. Global prevalence of chronic migraine: a systematic review. Cephalalgia 2010;30:599–609. http://dx.doi. org/10.1111/j.1468-2982.2009.01941.x. [4] Lipton RB. Tracing transformation: chronic migraine classification, progression, and epidemiology. Neurology 2009;72:S3–7. http://dx.doi.org/10.1212/WNL. 0b013e3181974b19. [5] Katsarava Z, Schneeweiss S, Kurth T, Kroener U, Fritsche G, Eikermann A, et al. Incidence and predictors for chronicity of headache in patients with episodic migraine. Neurology 2004;62:788–90. http://dx.doi.org/10.1212/01.WNL.0000113747.18760. D2. [6] Bigal ME, Serrano D, Buse D, Scher A, Stewart WF, Lipton RB, et al. Acute migraine medications and evolution from episodic to chronic migraine: a longitudinal population-based study. Headache 2008;48:1157–68. http://dx.doi.org/10.1111/j. 1526-4610.2008.01217.x. [7] D'Andrea G, D'Amico D, Bussone G, Bolner A, Aguggia M, Saracco MG, et al. The role of tyrosine metabolism in the pathogenesis of chronic migraine. Cephalalgia 2013; 33:932–7. http://dx.doi.org/10.1177/0333102413480755. [8] Zwart JA, Dyb G, Hagen K, Svebak S, Stovner LJ, Holmen J, et al. Analgesic overuse among subjects with headache, neck, and low-back pain. Neurology 2004;62: 1540–4. http://dx.doi.org/10.1212/01.WNL.0000123262.96132.FC. [9] Limmroth V, Katsarava Z, Fritsche G, Przywara S, Diener HC. Features of medication overuse headache following overuse of different acute headache drugs. Neurology 2002;59:1011–4. http://dx.doi.org/10.1212/WNL.59.7.1011. [10] Evers S, Marziniak M. Clinical features, pathophysiology, and treatment of medication-overuse headache. Lancet Neurol 2010;9:391–401. http://dx.doi.org/ 10.1016/S1474-4422(10)70008-9. [11] Mehuys E, Paemeleire K, Van Hees T, Christiaens T, Van Bortel LM, Van Tongelen I, et al. Self-medication of regular headache: a community pharmacy-based survey. Eur J Neurol 2012;19:1093–9. http://dx.doi.org/10.1111/j.1468-1331.2012. 03681.x. [12] Bigal ME, Lipton RB. Overuse of acute migraine medications and migraine chronification. Curr Pain Headache Rep 2009;13:301–7. http://dx.doi.org/10.1007/ s11916-009-0048-3. [13] Starling AJ, Hoffman-Snyder C, Halker RB, Wellik KE, Vargas BB, Dodick DW, et al. Risk of development of medication overuse headache with nonsteroidal antiinflammatory drug therapy for migraine: a critically appraised topic. Neurologist 2011;17:297–9. http://dx.doi.org/10.1097/NRL.0b013e31822d109c. [14] Lantéri-Minet M, Duru G, Mudge M, Cottrell S. Quality of life impairment, disability and economic burden associated with chronic daily headache, focusing on chronic migraine with or without medication overuse: a systematic review. Cephalalgia 2011;31:837–50. http://dx.doi.org/10.1177/0333102411398400. [15] D'Amico D, Grazzi L, Usai S, Raggi A, Leonardi M, Bussone G, et al. Disability in chronic daily headache: state of the art and future directions. Neurol Sci 2011;32:S71–6. http://dx.doi.org/10.1007/s10072-011-0552-1. [16] Linde M, Gustavsson A, Stovner LJ, Steiner TJ, Barré J, Katsarava Z, et al. The cost of headache disorders in Europe: the Eurolight project. Eur J Neurol 2012;19:703–11. http://dx.doi.org/10.1111/j.1468-1331.2011.03612.x. [17] Raggi A, Giovannetti AM, Leonardi M, Schiavolin S, D'Amico D, Curone M, et al. Disability and mood state in patients with episodic and chronic migraine associated to medication overuse. Neurol Sci 2012;33:S169–71. http://dx.doi.org/10.1007/ s10072-012-1076-z. [18] Andrasik F, Grazzi L, Usai S, Kass S, Bussone G. Disability in chronic migraine with medication overuse: treatment effects through 5 years. Cephalalgia 2010;30: 610–4. http://dx.doi.org/10.1111/j.1468-2982.2009.01932.x. [19] Raggi A, Leonardi M, Giovannetti AM, Schiavolin S, Bussone G, Grazzi L, et al. A 14month study of change in disability and mood state in patients with chronic migraine associated to medication overuse. Neurol Sci 2013;34:S139–40. http://dx. doi.org/10.1007/s10072-013-1371-3. [20] Curone M, Tullo V, Savino M, Proietti-Cecchini A, Bussone G, D'Amico D, et al. Outcome of patients with chronic migraine with medication overuse and depression after duloxetine: influence of coexisting obsessive compulsive disorder. Neurol Sci 2013;34:S175–7. http://dx.doi.org/10.1007/s10072-013-1376-y. [21] Bendtsen L, Munksgaard S, Tassorelli C, Nappi G, Katsarava Z, Lainez M, et al. Disability, anxiety and depression associated with medication-overuse headache can be considerably reduced by detoxification and prophylactic treatment. Results from a multicentre, multinational study (COMOESTAS project). Cephalalgia 2014;34: 426–33. http://dx.doi.org/10.1177/0333102413515338. [22] Stewart WF, Lipton RB, Whyte J, Dowson A, Kolodner K, Liberman JN, et al. An international study to assess reliability of the Migraine Disability Assessment (MIDAS) score. Neurology 1999;53:988–94. http://dx.doi.org/10.1212/WNL.53.5.988. [23] Curone M, Tullo V, Mea E, Proietti-Cecchini A, Peccarisi C, Bussone G, et al. Psychopathological profile of patients with chronic migraine and medication overuse: study and findings in 50 cases. Neurol Sci 2011;32:S177–9. http://dx.doi.org/10. 1007/s10072-011-0527-2. [24] Autret A, Roux S, Rimbaux-Lepage S, Valade D, Debiais S. West Migraine Study Group Psychopathology and quality of life burden in chronic daily headache: influence of migraine symptoms. J Headache Pain 2010;11:247–53. http://dx.doi.org/10. 1007/s10194-010-0208-1. [25] World Health Organization. The International Classification of Functioning, Disability and Health-ICF. Geneva: WHO; 2001. [26] Ustun TB, Chatterji S, Kostanjsek N, Rehm J, Kennedy C, Epping-Jordan J, et al. Developing the World Health Organization Disability Assessment Schedule 2.0. Bull World Health Organ 2010;88:815–23. http://dx.doi.org/10.2471/BLT.09.067231.
65
[27] Leonardi M, Raggi A, Bussone G, D'Amico D. Health-related quality of life, disability and severity of disease in patients with migraine attending to a specialty headache center. Headache 2010;50:1576–86. http://dx.doi.org/10.1111/j.1526-4610.2010.01770.x. [28] Leonardi M, Raggi A, Antozzi C, Confalonieri P, Maggi L, Cornelio F, et al. The relationship between health, disability and quality of life in Myasthenia gravis: results from an Italian study. J Neurol 2010;257:98–102. http://dx.doi.org/10.1007/s00415-0095279-z. [29] Quintas R, Alvarez A, Koutsogeorgou E, Cerniauskaite M, Meucci P, Sattin D, et al. The relationship between health-related quality of life and disability in patients with controlled epilepsy: a cross-sectional observational study. Am J Phys Med Rehabil 2012;91:S31–8. http://dx.doi.org/10.1097/PHM.0b013e31823d4db9. [30] Cerniauskaite M, Quintas R, Koutsogeorgou E, Meucci P, Sattin D, Leonardi M, et al. Quality of life and disability in patients with Stroke. Am J Phys Med Rehabil 2012; 91:S39–47. http://dx.doi.org/10.1097/PHM.0b013e31823d4df7. [31] Leonardi M, Raggi A, Pagani M, Carella F, Soliveri P, Albanese A, et al. Relationships between disability, quality of life and prevalence of nonmotor symptoms in Parkinson's disease. Parkinsonism Relat Disord 2012;18:35–9. http://dx.doi.org/10. 1016/j.parkreldis.2011.08.011. [32] Silberstein SD, Lipton RB, Sliwinski M. Classification of daily and near-daily headaches. Neurology 1996;47:871–5. [33] Martin BC, Pathak DS, Sharfman MI, Adelman JU, Taylor F, Kwong WJ, et al. Validity and reliability of the Migraine-Specific Quality of Life Questionnaire (MSQ Version 2.1). Headache 2000;40:204–15. http://dx.doi.org/10.1046/j.1526-4610.2000. 00030.x. [34] Federici S, Meloni F, Mancini A, Lauriola M, Olivetti Belardinelli M. World Health Organisation Disability Assessment Schedule II: contribution to the Italian validation. Disabil Rehabil 2009;31:553–64. http://dx.doi.org/10.1080/ 09638280802240498. [35] D'Amico D, Mosconi P, Genco S, Usai S, Prudenzano AM, Grazzi L, et al. The Migraine Disability Assessment (MIDAS) questionnaire: translation and reliability of the Italian version. Cephalalgia 2001;21:947–52. http://dx.doi.org/10.1046/j.0333-1024. 2001.00277.x. [36] Beck AT, Steer RA, Brown GK. Manual for the Beck Depression Inventory-II. San Antonio: Psychological Corporation; 1996. [37] Silberstein SD, Blumenfeld AM, Cady RK, Turner IM, Lipton RB, Diener HC, et al. OnabotulinumtoxinA for treatment of chronic migraine: PREEMPT 24-week pooled subgroup analysis of patients who had acute headache medication overuse at baseline. J Neurol Sci 2013;331:48–56. http://dx.doi.org/10.1016/j.jns. 2013.05.003. [38] Diener HC, Bussone G, Van Oene JC, Lahaye M, Schwalen S, Goadsby PJ, et al. Topiramate reduces headache days in chronic migraine: a randomized, doubleblind, placebo-controlled study. Cephalalgia 2007;27:814–23. http://dx.doi.org/10. 1111/j.1468-2982.2007.01326.x. [39] Lipton RB, Varon SF, Grosberg B, McAllister PJ, Freitag F, Aurora SK, et al. Onabotulinumtoxin A improves quality of life and reduces impact of chronic migraine. Neurology 2011;77:1465–72. http://dx.doi.org/10.1212/WNL. 0b013e318232ab65. [40] Silberstein SD, Loder E, Forde G, Papadopoulos G, Fairclough D, Greenberg S, et al. The impact of migraine on daily activities: effect of topiramate compared with placebo. Curr Med Res Opin 2006;22:1021–9. http://dx.doi.org/10.1185/ 030079906X104731. [41] Dodick DW, Freitag F, Banks J, Saper J, Xiang J, Rupnow M, et al. Topiramate versus amitriptyline in migraine prevention: a 26-week, multicenter, randomized, double-blind, double-dummy, parallel-group noninferiority trial in adult migraineurs. Clin Ther 2009;31:542–59. http://dx.doi.org/10.1016/j.clinthera. 2009.03.020. [42] Cevoli S, D'Amico D, Martelletti P, Valguarnera F, Del Bene E, De Simone R, et al. Underdiagnosis and undertreatment of migraine in Italy: a survey of patients attending for the first time 10 headache centres. Cephalalgia 2009;29:1285–93. http://dx.doi. org/10.1111/j.1468-2982.2009.01874.x. [43] Bagley CL, Rendas-Baum R, Maglinte GA, Yang M, Varon SF, Lee J, et al. Validating Migraine-Specific Quality of Life Questionnaire v2.1 in episodic and chronic migraine. Headache 2012;52:409–21. http://dx.doi.org/10.1111/j.1526-4610.2011. 01997.x. [44] Rendas-Baum R, Bloudek LM, Maglinte GA, Varon SF. The psychometric properties of the Migraine-Specific Quality of Life Questionnaire version 2.1 (MSQ) in chronic migraine patients. Qual Life Res 2013;22:1123–33. http://dx.doi.org/10.1007/s11136012-0230-7. [45] Raggi A, Giovannetti AM, Schiavolin S, Leonardi M, Bussone G, Grazzi L, et al. Validating the Migraine-Specific Quality of Life Questionnaire v2.1 (MSQ) in Italian inpatients with chronic migraine with a history of medication overuse. Qual Life Res 2014;23:1273–7. http://dx.doi.org/10.1007/s11136-013-0556-9. [46] Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders: 2nd edition. Cephalalgia 2004; 24:S9–S160. [47] Hedges LV. Distribution theory for Glass's estimator of effect size and related estimators. J Educ Behav Stat 1981;6:107–28. [48] Ware JE. SF-36 health survey update. Spine 2000;25:3130–9. [49] Cole JC, Lin P, Rupnow MF. Validation of the Migraine-Specific Quality of Life Questionnaire version 2.1 (MSQ v. 2.1) for patients undergoing prophylactic migraine treatment. Qual Life Res 2007;16:1231–7. http://dx.doi.org/10.1007/s11136-0079217-1. [50] Yalug I, Selekler M, Erdogan A, Kutlu A, Dundar G, Ankarali H, et al. Correlations between alexithymia and pain severity, depression, and anxiety among patients with chronic and episodic migraine. Psychiatry Clin Neurosci 2010;64:231–8. http://dx. doi.org/10.1111/j.1440-1819.2010.02093.x.
66
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
[51] Raggi A, Covelli V, Pagani M, Meucci P, Martinuzzi A, Buffoni M, et al. Sociodemographic features and diagnoses as predictors of severe disability in a sample of adults applying for disability certification. Int J Rehabil Res 2014;37:180–6. http://dx.doi.org/10.1097/MRR.0000000000000054. [52] Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet 2012;380:37–43. http://dx.doi.org/10.1016/S01406736(12)60240-2.
[53] Bigal ME, Lipton RB. Excessive opioid use and the development of chronic migraine. Pain 2009;142:179–82. http://dx.doi.org/10.1016/j.pain.2009.01.013. [54] Buse DC, Pearlman SH, Reed ML, Serrano D, Ng-Mak DS, Lipton RB, et al. Opioid use and dependence among persons with migraine: results of the AMPP study. Headache 2012;52:18–36. http://dx.doi.org/10.1111/j.1526-4610.2011.02050.x. [55] Vinson DR. Treatment patterns of isolated benign headache in US emergency departments. Ann Emerg Med 2002;39:215–22. http://dx.doi.org/10.1067/mem. 2002.121400.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences journal homepage: www.elsevier.com/locate/jns
Chronic migraine with medication overuse: Association between disability and quality of life measures, and impact of disease on patients' lives Alberto Raggi a,⁎, Silvia Schiavolin a, Matilde Leonardi a, Ambra Mara Giovannetti a, Gennaro Bussone b, Marcella Curone b, Paola Di Fiore b, Licia Grazzi b, Susanna Usai b, Domenico D'Amico b a b
Neurology, Public Health and Disability Unit, Neurological Institute “C. Besta” IRCCS Foundation, Milan, Italy Headache and Neuroalgology Unit, Neurological Institute “C. Besta” IRCCS Foundation, Milan, Italy
a r t i c l e
i n f o
Article history: Received 11 August 2014 Received in revised form 31 October 2014 Accepted 3 November 2014 Available online 8 November 2014 Keywords: Chronic migraine Medication overuse Health-related quality of life Disability Detoxification Withdrawal
a b s t r a c t Patients with chronic migraine with medication overuse (CM-MO) have decreased quality of life (QoL) and increased disability: the degree to which these outcomes are connected to disease severity and the pattern of MO towards disease severity are unclear. Patients under withdrawal were administered the Migraine Disability Assessment (MIDAS), the WHO Disability Assessment Schedule (WHODAS), and the Migraine-Specific Quality of Life Questionnaire (MSQ). They overused NSAIDs, triptans, NSAIDs and triptans, and other drugs (ergotamine, caffeine, opioids/barbiturates). We calculated the correlations between MIDAS, WHODAS, and MSQ; compared WHODAS to normative scores; compared MIDAS, WHODAS, and MSQ in patients with different CM-MO severity; and run a logistic regression to predict CM-MO severity based on overused drugs. One hundred ninety-four patients were enrolled: correlations between WHODAS, MSQ, and MIDAS were moderate; wide differences on WHODAS against normative were found; and no trend was found across severity groups. Compared to triptans overusers, patients overusing NSAID and other drugs had higher odds of severe CM-MO. Coupling different disability measures with QoL assessment offered different insights on the lived experience of CM-MO. Future studies are needed to clarify the relationship between overused drugs and CM-MO severity: we added evidence that NSAIDs do not have protective effect in high-frequency CM-MO. © 2014 Elsevier B.V. All rights reserved.
1. Introduction Chronic migraine (CM) is characterized by headaches occurring 15 or more days per month and can be considered as a negative evolution of the migraine course in a subgroup of subjects [1]. CM prevalence in population studies ranges from 1.4% to 4% and from 2.5% to 3% of migraineurs (14% in clinical samples) progress to CM each year [2–5]. Several factors may play a role in the progression from episodic migraine (EM) to CM, such as lifestyle, comorbid conditions, metabolic dysfunction, mood, genetic terrain, and medication overuse (MO) [1,6, 7]. CM is strongly associated with the overuse of non-steroidal analgesic drugs (NSAIDs) as these drugs can widely be accessed as over the counter medications: the fact that frequent use of symptomatic drugs is a risk factor for progression of migraine marks an important difference with other pain-related conditions [8]. Indeed all acute drugs may be associated with chronification, but it is not clear if and how the disease's ⁎ Corresponding author at: Neurology, Public Health and Disability Unit, Neurological Institute C. Besta, IRCCS Foundation, Via Celoria 11, 20133, Milan, Italy. Tel.: + 39 02 2394 3105 (2521); fax: +39 02 2394 2442. E-mail address: [email protected] (A. Raggi).
http://dx.doi.org/10.1016/j.jns.2014.11.004 0022-510X/© 2014 Elsevier B.V. All rights reserved.
features are related to the different overused drugs [9] and the mechanisms for chronification are not completely understood yet. The pattern of overused medication varies across time and countries [10]. A community pharmacy-based study showed that one-fourth of patients overused acute medications and that combination of three or more agents was more frequent in overusers compared to nonoverusers: overusers also had higher headache-related, headaches frequency, and higher pain intensity [11]. Findings on chronification process connected to the class of overused medication are however partial and contrasting. Some evidence exists on the increased risk of chronification in patients using triptans, but only in those with a baseline frequency around 10–14 headaches/month, and of the protective effect of NSAIDs in low-frequency migraine sufferers [12,13]. Barbiturates and opioids have also been found to induce migraine progression, with a dose-dependent effect [12]. Recent reviews showed that health-related quality of life (HRQoL) is significantly impaired in persons suffering from chronic forms of headache, compared to non-headache sufferers, as well as compared to those with EM, and that this negative impact is more evident in those suffering from CM, particularly when MO is present [14,15]. CM causes physical health and emotional problems and determines increased disability,
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
measured in terms of absenteeism and days with reduced workplace effectiveness; therefore, it has remarkable consequences in terms of increased disease cost, which is three-fold that of EM (3561€/year vs 1222 €/year) [16]. Previous studies addressing disability in chronic migraine with medication overuse (CM-MO) [17–21] generally used the Migraine Disability Assessment (MIDAS) [22]. However, the use of MIDAS in CM patients poses some problems. First, the highest severity grade corresponds to a score N 21, which is intuitively overrepresented among patients reporting more than 45 headaches in the previous three months. In addition, the conceptualization of disability behind MIDAS is highly dependent on headaches frequency and does not account for other features of CM such as reduced mood or obsessive or anxiety disorders but also for the severity of pain, which are likely to impact on functioning in social and workplace roles, particularly as far as interpersonal relationships [17,23,24]. A model that recognizes disability in a multidimensional way is therefore needed. Such disability model should correspond to that endorsed by the World Health Organization (WHO) with the International Classification of Functioning, Disability and Health (ICF) [25]. The understanding of the possible relationships between disability—particularly when viewed as a multidimensional phenomenon—and HRQoL in patients with CM-MO may enhance the understanding of this disorder and help clinicians and researchers in choosing the most appropriate patient-reported outcome measure (PROMs) in specific studies. The WHO also produced an ICF-based disability assessment, the WHO Disability Assessment Schedule second version (WHODAS) [26]: previous studies with neurological patients (namely, EM, myasthenia gravis, epilepsy, stroke and Parkinson's disease) showed that it is suitable to address the increased disability level of patients as well as a moderate association with HRQoL assessments [27–31]. To our knowledge, the relationship between disability, assessed with an instrument like the WHODAS, and HRQoL measures in CM-MO was never addressed. The primary aim of this study was to assess the relationships between two disability measures, MIDAS and WHODAS, and between them and patients' HRQoL, in order to assess whether these different tools cover similar constructs or offer the possibility to evaluate different facets of CM-related impact: we expect to expand to the sample of CM-MO patients the findings of a previous study on EM patients, i.e., that MIDAS and WHODAS are only partially correlated and that the correlations with HRQoL are mild to moderate. The secondary aim was to address the impact of CM-MO on patients' disability. This was performed with three different procedures: (a) the evaluation of the impact of CM-MO on patients' disability compared to the general population, with the hypothesis that CM-MO patients report worse scores at WHODAS; (b) the evaluation of the extent to which differences in PROMs scores correspond to different severity profiles—based on frequency and intensity of headaches—with the hypothesis that PROMs scores cannot be completely explained by differences in CM-MO severity; and (c) the evaluation of the relative risk of having a more severe disease profile accounting for the type of overused drug, with the hypothesis that those overusing NSAIDs have are more likely to have the most severe profile. 2. Methods
61
during administration of the protocol. In any case, they were allowed to postpone it if they had an headache. The study was approved by the institute's ethical committee, and each patient signed an informed consent form prior to data collection. All enrolled patients completed the Migraine-Specific Quality of Life Questionnaire Version 2.1 (MSQ) [33] to evaluate HRQoL, the WHODAS [26,34], and the MIDAS [22,35] to evaluate disability. Headache frequency and average pain intensity were taken from the last two MIDAS items. Mood state was assessed with the Beck Depression Inventory [36]. MSQ is based on 14 items that investigate the impact of headaches on patients HRQoL during the previous 4 weeks. The items are grouped into three subscales: Role Restriction (RR), 7 items assessing how migraines limit one's daily social and work-related activities; Role Prevention (RP), 4 items assessing how migraines prevent social and workrelated activities; and Emotional Function (EF), 3 items assessing the emotions associated with headaches. Each item refers on how frequently headaches limits the activity under consideration and is rated on a 6-point scale from none of the time to all of the time. Each subscale has a 0–100 score, with lower scores indicating poor HRQoL. MSQ has been used as an outcome measure in clinical trials, mostly on preventive medications [37–41]: however, its use in research on patients with CM or other types of chronic headaches is limited [37–39]. In its Italian version, it has been used in one observational study in which patients with different type of migraine, of whom 2.5% had CM-MO, were evaluated [42]. Recently, three studies confirmed its validity in CM patients [43–45]: however, the presence of MO was explicitly excluded in the first study, which is based on the International Classification of Headache Disorders, 2nd version criteria for CM [46]; in the second study, only 60% of patients had concurrent MO; in the third study, all patients had CMMO. Therefore, the use of MSQ in patients with CM-MO is restricted to two studies [38,45]. WHODAS is a 36-item ICF-based disability assessment tool that examines the difficulties experienced by an individual due to a given health condition. Six domains are taken into account: understanding and communicating, getting around, self-care, getting along with people, life activities (divided into household and work), and finally participation in society. Patients are required to answer questions regarding how much difficulty, due to their health condition, they experienced during the previous 30 days. Answers are rated on a 5-point scale, from no problems to complete problems/cannot do the activity. Both total and subscale scores are available, ranging from 0 to 100, with higher scores reflecting greater disability. Total score can be calculated also on the basis of 32 items, excluding those related to work activities if the respondent is not currently employed. MIDAS is composed of 7 questions referred to the preceding 3 months: the first 5 investigate the influence of headaches on paid and school work, household work, and on leisure/family/social duties. MIDAS score is calculated by adding the individual scores of the first 5 questions and indicates the number of days in which migraine interfered with these activities. According to given intervals, 4 disability grades can be calculated: minimal (0–5), mild (6–10), moderate (11– 20), and severe (≥21) disability. BDI-2 is composed of 21 items, each rated on a 0–3 scale, that address cognitive and somatic-affective features of depression. Total score range is 0–63, with higher scores reflecting higher depressive mood.
2.1. Patients, procedure, and measures In this cross-sectional observational study, adult patients with CMMO according to Silberstein's criteria [32] were consecutively recruited at the Headache Centre of the Neurological Institute C. Besta of Milan between June 2011 and December 2012. Patients were enrolled on occasion of inpatient withdrawal treatment: the questionnaires were provided on the second or third day of hospitalisation. There are two reasons for this: first, to enable physicians to evaluate patients' eligibility; second, to make it more likely that patients were headache-free
2.2. Definition of medication overuse categories The different types of MO were defined as follows: overuse of nonsteroidal anti-inflammatory drugs (NSAIDs), when NSAIDS were used for at least 15 days/month; overuse of triptans, when any drug of this class was used for at least 10 days/month; overuse of both NSAIDs and triptans, when NSAIDs were used for at least 15 days/month and triptans for at least 10 days/month; overuse of ergotamine, caffeine,
62
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
Table 1 Spearman's rank correlations between WHODAS, MIDAS, MSQ, headaches frequency and average pain intensity.
WHODAS—summary score MIDAS Headaches frequency Pain intensity BDI-2
MSQ-RR
MSQ-RP
MSQ-EF
MIDAS
Headaches frequency
Pain intensity
BDI-2
−0.449* −0.569* −0.209 −0.355* −0.308*
−0.504* −0.510* −0.151 −0.359* −0.300*
−0.438* −0.289* −0.110 −0.262* −0.494*
0.417* NA 0.242* −0.161 0.328*
−0.100 NA NA 0.011 0.014
0.307* NA 0.NA NA 0.213
0.587* NA NA NA NA
Note. *Spearman's rank correlations significant at P b 0.0021; unmarked correlations are not significant; NA = not applicable. WHODAS = World Health Organization Disability Assessment Schedule; MSQ = Migraine-Specific Quality of Life Questionnaire; MSQ-RR = MSQ Role Restriction; MSQ-RP = MSQ Role Prevention; MSQ-EF = MSQ Emotional Function; MIDAS = Migraine Disability Assessment Score.
opioids/barbiturates, when any drug of this class was used for at least 10 days/month, alone or together with NSAIDs or triptans.
2.3. Data analysis Spearman's rank correlation analysis was used to assess the relationships between MIDAS, WHODAS summary score, and MSQ scales. Correlations were considered weak for coefficient values b 0.29, moderate for values between 0.30 and 0.59 and strong for values N 0.60. Correlations were also carried out between the three questionnaires' headaches frequency, pain intensity, and BDI-2 score. As 24 multiple correlations were carried out, we applied Bonferroni correction, and significance was set with Type I error level at α = 0.0021 and 2-tailed testing. With regard to secondary aims, the impact of CM-MO on patients' disability was assessed comparing WHODAS scores against normative Italian scores reported by Federici and colleagues [34] referred to the general population (271 “normal adults”, mean age 30 years, SD 9.9). Comparison against normative Italian values was made using onesample t-test and Hedges's g as a measure of effect size [47] and was calculated as the difference between our sample's means and the means observed by Federici divided by the pooled SD: g N 0.5 was considered to reflect medium effect size, and g N 0.8 to reflect large effect size. Comparison was not made with Life activities scale of the WHODAS because in the paper of Federici and colleagues [34], no distinction between household and work activities was made. As 18 six comparisons were carried out, we applied Bonferroni correction and significance was set with Type I error level at α = 0.0083. Disability and HRQoL were compared in patients with different CM severity. To address CM severity, we run an exploratory K-means cluster analysis using MIDAS questions regarding the total number of days with headaches in the last 3 months and the average pain intensity as grouping variables, and one-way ANOVA to evaluate the contribution of the two variables in the identification of clusters. We predefined 3 clusters without specifying initial cluster centroids and decided to have a maximum of 10 iterations for defining cluster centers and membership. Kruskal– Wallis H test, a nonparametric analog of one-way analysis of variance, was performed to compare WHODAS 2.0 summary score, MIDAS, and MSQ in patients with different migraine severity. Bonferroni correction Table 2 One-sample t-test between study and normative data for WHODAS. WHODAS Scale
Sample data
Normative data
Effect Size
t-test
Understanding and communicating Getting around Self-care Getting along with people Household activities Work activities Participation in society Summary score
31.8 (18.8)
11.84 (12.96)
1.27
t = 14.8
28.1 (22.6) 10.8 (15.7) 21.6 (19.6) 43.0 (26.6) 38.9 (24.6) 37.4 (15.5) 31.2 (13.8)
7.07 (13.51) 3.53 (8.00) 12.57 (16.97) NA NA 12.12 (13.86) 12.95 (11.77)
1.18 0.61 0.50 NA NA 1.74 1.44
t= t= t= NA NA t= t=
13.0 6.5 6.5
22.8 18.5
Note. Values are reported as means (SD). NA, not applicable. P-value was b 0.001 for all t-tests.
was used to control for multiple comparisons: as 5 comparisons were performed, significance was set with Type I error level at α = 0.01. To assess the relative risk of having the most severe disease profile, logistic regression was carried out entering the most severe disease severity as dependent variable, pattern of overused drug (NSAIDs, triptans, NSAIDs and triptans, other drugs, i.e., ergotamine, caffeine, and opioids/ barbiturate) as independent variables, and sex, age, employment, marital status, and educational level as covariates. Overuse of triptans was set as reference for independent variable. The −2 log-likelihood difference with chi-square was used to test the difference between the full model and the model based on the intercept only. Nagelkerke pseudo R2 was used to assess the whole explanatory power of the model. Data analysis was performed using PAWS statistic 18.0. 3. Results A total of 213 patients were screened, two declined to participate and 17 did not match inclusion criteria; therefore, 194 patients, of whom 160 females (82.5%), were enrolled. Age range was 18–75 years (mean age 43.9 years, SD 12.0 years); 124 patients (63.9%) completed high school or had academic education level, 143 (73.7%) were employed or students, and 139 (71.6%) were married or cohabitating. The mean number of days with headaches in the previous three months was 56.3 (SD 23.7) and average pain intensity was 7.8 (SD 1.3). Mean MIDAS score was 82.7 (SD 59.9), and in 175 patients (90.2%), it was higher than 21, thus indicative of severe disability. The mean number of workdays lost on a monthly basis was 4.3 (SD 6.1), and the mean number days of housework lost was 6.1 (SD 6.2). The mean MSQ-RR was 32.7 (SD 17.3), the mean MSQ-RP was 47.6 (SD 22.0), the mean MSQ-EF was 42.2 (SD 24.9), and the mean WHODAS summary score was 31.2 (SD 13.8). As far as different classes of overused drugs, 57 patients (29.4%) overused NSAIDs, 42 (21.7%) triptans, 60 (30.9%) both NSAIDs and triptans, and 35 (18%) overused ergotamine, caffeine, or opioids/barbiturates. The correlations between the scores obtained at different tools are reported in Table 1. Correlations between WHODAS, MSQ, and MIDAS were generally moderate; MIDAS correlated to frequency of headaches but not to pain intensity, while WHODAS correlated to pain intensity and not to headaches frequency. BDI-2 was more strongly correlated with WHODAS than with MIDAS; correlations with MSQ subscales were stronger with EF than with RR and RP. Table 2 shows results on the impact of CM-MO on ability to function. For all WHODAS scales, the scores reported by patients with CM-MO were significantly lower compared to normative data, with Hedges's g reporting medium to large effect sizes. All one-sample t-tests were significant a P b 0.001 level. Cluster analysis is reported in Table 3. The three severity clusters were identified only on the basis headaches frequency, while the average pain severity did not provide any significant contribution to clusters' definition, as shown by one-way ANOVA analysis. Table 4 reports the results of Kruskal–Wallis H test for MIDAS score, WHODAS summary score and MSQ in patients from different severity clusters: statistical difference was found only for MIDAS. Table 5 reports the results of logistic regression. The overall explicative power of the model was 11.4%, and − 2 log-likelihood difference
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
63
Table 3 K-means cluster analysis and Kruskal–Wallis H test: patients divided into three groups according to migraine frequency and pain intensity.
Migraine frequency Pain intensity
Cluster A (No. 53)
Cluster B (No. 74)
Cluster C (No. 67)
F statistic
28.6 (26.5–29.7) 7.9 (7.5–8.3)
51.4 (49.5–53.3) 7.6 (7.4–7.9)
85.0 (83.4–86.6) 7.9 (7.6–8.2)
976.8 (P b 0.0001) 0.86 (P = 0.425)
Note. For each group, cluster center and 95% confidence interval are reported. Cluster A: low CM-MO severity. Cluster B: intermediate CM-MO severity. Cluster C: high CM-MO severity. Migraine frequency: days with migraine in the last 3 months. Pain intensity: rated on a 0–10 scale. Bonferroni post hoc test is significant (P b 0.0001) for migraine frequency between each pair; mean difference between severe and moderate cluster was 33.6, between severe and mild cluster was 56.4, and between moderate and mild cluster was 22.8.
analysis show that the final model outperforms the intercept-based model. Patients overusing NSAIDs only or other drugs had higher likelihood of having the most severe disease pattern (approximately 2.5 and three-fold compared to the reference of patients overusing triptans). 4. Discussion With respect to the primary aim of this study, our results showed that MIDAS and WHODAS were significantly and moderately correlated and that the correlations with MSQ and BDI-2 were moderate too, thus showing that the underlying concepts they are intended to measure are different. With regard to secondary aims, our results showed the strong impact of CM-MO on patients' disability, while we failed in determining a clear increase of disability and decrease in HRQoL when differences were assessed in patients with different CM-MO severity: differences were found only for MIDAS. Finally, patients in the most severe cluster (i.e., with more frequent headaches) were more likely to be overusers of NSAIDs or of ergotamine, caffeine, and opioids/barbiturates. The two disability measures displayed a moderate correlation, thus giving confirmation of the hypothesis that the constructs behind them are not overlapping. This result is largely consistent with that previously found in a sample of patients with EM [27], that was similar for age (the present sample was 0.4 years older) and female prevalence (−2.8% in the present sample), but displayed clear differences at MIDAS: score N21 was in fact reported by two-fold of the patients in this study that in the previous on EM. Therefore, our results reinforce those previously reported on EM patients: MIDAS and WHODAS measure different facets of the same construct, as the first provides a measure based upon frequency of migraine-related problems while the second provides a measure of the intensity of disease-related problems. Such a hypothesis was also reinforced by the different pattern of correlations between the two disability measures and patient-reported headaches frequency and intensity: MIDAS correlated to headaches frequency, and WHODAS to average pain intensity. Mild to moderate associations were also found between disability and HRQoL scores as well as between these scales and mood state. The association was stronger for scales referred to restriction and prevention of social activities rather than for the issues of emotions associated to headaches, thus showing that the concepts behind the three assessment instruments were neither overlapping nor transposable. The same kind of results were also found in the paper on EM, where the Medical Outcome Survey 36-Item Short-Form Health Survey (SF36) [48] was used. The correlations between SF-36 and MSQ were reported in two studies and ranged between 0.19 and 0.38 [33] and between 0.26 and 0.35 [49]. In the present study, the association between the WHODAS and MSQ scale exploring emotional components (MSQ-EF) was weaker than that with MSQ-RR and MSQ-RP: again, this is similar to what was found in the previous study on EM patients, where WHODAS correlated more with physical component than with mental component of SF-36 [27]. The opposite was instead found for the correlations between MIDAS and MSQ, which was stronger with MSQ-RR and MSQ-RP than with MSQ-EF: this can be explained considering that MIDAS specifically reflects the impact of migraine on productivity/performance in daily activities, while MSQ-EF assesses the emotions associated with headaches. Here we also added a measure of mood state: results show that depression correlated more with
disability when measured with WHODAS than when measured with MIDAS and that—as expectable—it correlated more with MSQ-EF than with MSQ-RR and MSQ-RP. Taken together, these results suggest that the use of both HRQoL and disability assessment instruments is important to couple information on aspects connected to physical health (e.g., mobility, self-care), relationships and participation to social situations, with those connected to emotional issues. Patients with chronic forms of headaches display an increased burden connected to all these aspects, as shown in several studies [17,24,50], that should be measured to understand patients perspective and to assess the outcomes after treatments. Our results also point out that WHODAS addresses in a better way than MIDAS the issues connected to mood state. Our results confirmed the relevant impact of CM-MO on patients' daily lives, with a relevant burden on the individual in terms of difficulties in dealing with all the domains of life, from those most frequently investigated—such as employment, where the number of workdays lost by CM-MO patients is more than six-fold those observed in EM patients—to those less frequently considered, such as mobility and selfcare. In a previous study with age and gender-matched patients with EM and CM-MO, we showed that the latter have higher disability [17], and the scores observed in the present paper at WHODAS are indicative of a higher disability if contrasted to those reported in a sample of EM patients [27]. However, similarly to what observed in these patients, severity of problems in WHODAS mobility domain (e.g., getting out of your home or walking a long distance) was similar to that of problems with understanding and communicating (e.g., generally understanding what people say, or starting and maintaining a conversation) and was higher than problems with getting along with people (e.g., dealing with friends, or dealing with unknown people). Moreover, our patients reported more problems in activities connected to participation in social activities (e.g., joining community activities, consequences of health problems on family members) than in relational situations. This analysis of WHODAS scales suggests that social situations may represent the context in which persons suffering from CM-MO experience emotional difficulties, while interpersonal interactions do not seem to be a problematic issue per se. Similar results were found, in addition to EM patients, also in other patients suffering from myasthenia gravis and controlled epilepsy [27–29]: the fact that these diseases are “invisible” to other persons might explain this similarity. Patients are likely to suffer in interpersonal situations because they may find little understanding by relevant others on their conditions and on the difficulties associated to them. These results, however, need to be read also in consideration of the different average age of this sample (43.9, SD 12.0) and of the Italian validation sample (30, SD 9.9), which might have a confounding effect. A recent population study showed that younger age was associated to lower odds of severe disability [51], which reflects the life span accumulation of health conditions (i.e., multimorbidity) that is associated with ageing [52]. When we addressed the WHODAS, MIDAS and MSQ scores in patients with different degree of disease severity, we failed to find significant differences other than in MIDAS, that is, however, based on headaches frequency, like the three clusters. A further proof to this is that in our study MIDAS score correlated, although weakly, with headaches frequency. Less surprising is the absence of significance found for WHODAS and MSQ as they did not correlate to headaches frequency.
64
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
Table 4 Kruskal–Wallis H test for MIDAS, WHODAS, and MSQ scales in patients with different disease severity.
MIDAS WHODAS MSQ-RR MSQ-RP MSQ-EF
Cluster A (No. 53)
Cluster B (No. 74)
Cluster C (No. 67)
Kruskal–Wallis H test
65.8 (54.5–77.2) 33.1 (28.8–37.3) 37.7 (33.0–42.4) 51.8 (45.7–57.9) 47.7 (40.5–55.0)
72.1 (61.6–82.5) 31.4 (28.4–34.5) 32.1 (28.7–35.4) 48.6 (44.1–53.1) 42.1 (36.7–47.5)
107.8 (89.0–125.5) 29.6 (26.4–32.8) 29.4 (24.7–34.2) 43.1 (37.3–49.0) 38.1 (32.0–44.2)
P P P P P
= = = = =
0.005 0.511 0.014 0.084 0.229
Note. For each group and scale, mean and 95% confidence interval are reported. MIDAS = Migraine Disability Assessment Score; WHODAS = World Health Organization Disability Assessment Schedule; MSQ-RR = MSQ Role Restriction; MSQ-RP = MSQ Role Prevention; MSQ-EF = MSQ Emotional Function.
Taken as a whole, the results of this analysis show that the burden caused by CM-MO is not only due to the severity of the headache disorder per se. Rather, it is possible to presume that the overuse of acute medications and the emotional burden associated to such a disabling condition play a role in the definition of its outcome: in fact, the score observed at MSQ-EF was 20.7 lower that that found by Cevoli and colleagues [42] in a sample of patients with migraine, of whom only 2.5% had CM-MO, and 6.1 lower compared to a study of Bagley and colleagues [43] that was based on a sample of patients without MO. Finally, our analysis of the relative risk of being included in the most severe cluster according to the overused medication showed that patients overusing NSAIDs only and those overusing ergotamine, caffeine, and opioids/barbiturates had higher odds of having a more severe CM compared to those overusing triptans, while no increased risk was found for patients overusing both NSAIDs and triptans. Previous research on the impact of different drugs categories on CM severity is lacking. It is known that NSAIDs may exert a sort of protective effect that, however, is restricted to low-frequency headaches and has been observed in the transition between episodic and chronic migraine [12, 13], while other studies point out the strongly risk of chronification associated with the overuse of NSAIDs and of opioid-containing products, with a dose-dependent effect [12]. Other reports showed that patients overusing ergotamine and NSAIDs typically have a daily tension-typelike headache, whereas patients with triptan-induced overuse are more likely to describe a migraine-like headache, i.e., CM [9]. This aspect deserves further research considering that NSAIDs are the most commonly overused drugs also because they are easily available as overthe-counter medications [8] and that opioids are associated to an increased risk of migraine chronification [53] may lead to abuse and dependence [54] and might be the result of an inappropriate long-term use of an emergency department prescription [55]. Moreover, the fact that patients overusing both NSAIDs and triptans did not have an increased risk of higher disease severity, compared to those overusing triptans only, might suggest a sort of “protective” effect of consuming a kind of drug that is not available as an over-the-counter medication, and is therefore more likely to be under the control of a clinician. The main limitation is that sample, entirely derived from a single specialty center, is likely to represent the most severe spectrum of CM,
Table 5 Logistic regression analyses for the association between overused drugs class and inclusion in the higher severity cluster.
as all included patients had very high headache frequency and were overusing symptomatic medications. It is also possible that the context of a structured withdrawal might have influenced the scores of disability and HRQoL instruments: thus, our results cannot be automatically generalized to all CM patients. Second, the cross-sectional research design does not allow us to define causal relationships between clinical data and patient-reported outcomes. Moreover, a “restriction of range” issue is also to be taken into account: in fact, as patients had a severe clinical condition, it is very likely that the range of reported problems and the scores at outcome measures chiefly reflected the upper range of WHODAS and MIDAS as well as the lower range of MSQ scores. Therefore, caution is recommended in the interpretation of correlation levels, which might be attenuated also in consideration of the restricted scores range. Third, the impact of CM over patients' disability should be evaluated considering the higher average age of this sample compared to the Italian validation sample. 5. Conclusions In conclusion, we evaluated disability and HRQoL in patients with CM-MO, showing a remarkable impact of the disease on such dimensions of patients' lives. Coupling different disability measures, such as the WHODAS and the MIDAS together with an HRQoL tool such as the MSQ may offer different insights on the lived experience of CM-MO: these measures in fact represent non-transposable psychosocial facets. MIDAS and WHODAS reflect different aspects of disability, connected to the two main features of the disease, i.e., the frequency of headaches and pain intensity, respectively. The inclusion of an HRQoL measure enabled to add information on emotional aspects connected to CM-MO, which are remarkable and not captured by MIDAS. On the other hand, WHODAS showed that persons suffering from CM-MO experience emotional difficulties in social situations, while interpersonal interactions do not seem to be problematic per se: it is possible to presume that the high frequency of headaches that force CM-MO patients to avoid social situation might play a role in the increased emotional burden of these patients. Future studies are needed to shed light on the complex relationship between overused drugs and CM-MO severity: our study seem to add evidence to the fact that NSAIDs overuse does not have protective effect in high-frequency CM and poses some elements for debate on use of opioids/barbiturates, caffeine, and ergotamine agents. Conflict of interest
Parameters estimates
Intercept Triptans NSAIDs + Triptans NSAIDs Other drugs
B (SE)
OR (95% CI)
P
−3.26 (1.00) – 0.16 (0.47) 0.94 (0.46) 1.12 (0.51)
– – 1.17 (0.46–2.97) 2.56 (1.04–6.32) 3.06 (1.13–8.30)
0.001 – 0.733 0.042 0.028
Note. NSAIDs = non-steroidal anti-inflammatory drugs; other = ergotamine, caffeine, opioids/barbiturates; OR = odds ratio; 95% CI = 95% confidence interval; df = degrees of freedom. Model based on intercept only −2 log likelihood = 240.95; final model −2 log likelihood = 224.16; chi-squared = 16.70, df = 5, P = 0.005. Nagelkerke pseudoR2 = 0.114.
The authors declare that they have no conflict of interest. Acknowledgments This paper is based on the results of DIS.CHRONIC project, sponsored by the Neurological Institute C. Besta IRCCS Foundation. References [1] Bigal ME, Lipton RB. Migraine chronification. Curr Neurol Neurosci Rep 2011;11: 139–48. http://dx.doi.org/10.1007/s11910-010-0175-6.
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66 [2] Manack AN, Buse DC, Lipton RB. Chronic migraine: epidemiology and disease burden. Curr Pain Headache Rep 2011;15:70–8. http://dx.doi.org/10.1007/s11916010-0157-z. [3] Natoli JL, Manack A, Dean B, Butler Q, Turkel CC, Stovner L, et al. Global prevalence of chronic migraine: a systematic review. Cephalalgia 2010;30:599–609. http://dx.doi. org/10.1111/j.1468-2982.2009.01941.x. [4] Lipton RB. Tracing transformation: chronic migraine classification, progression, and epidemiology. Neurology 2009;72:S3–7. http://dx.doi.org/10.1212/WNL. 0b013e3181974b19. [5] Katsarava Z, Schneeweiss S, Kurth T, Kroener U, Fritsche G, Eikermann A, et al. Incidence and predictors for chronicity of headache in patients with episodic migraine. Neurology 2004;62:788–90. http://dx.doi.org/10.1212/01.WNL.0000113747.18760. D2. [6] Bigal ME, Serrano D, Buse D, Scher A, Stewart WF, Lipton RB, et al. Acute migraine medications and evolution from episodic to chronic migraine: a longitudinal population-based study. Headache 2008;48:1157–68. http://dx.doi.org/10.1111/j. 1526-4610.2008.01217.x. [7] D'Andrea G, D'Amico D, Bussone G, Bolner A, Aguggia M, Saracco MG, et al. The role of tyrosine metabolism in the pathogenesis of chronic migraine. Cephalalgia 2013; 33:932–7. http://dx.doi.org/10.1177/0333102413480755. [8] Zwart JA, Dyb G, Hagen K, Svebak S, Stovner LJ, Holmen J, et al. Analgesic overuse among subjects with headache, neck, and low-back pain. Neurology 2004;62: 1540–4. http://dx.doi.org/10.1212/01.WNL.0000123262.96132.FC. [9] Limmroth V, Katsarava Z, Fritsche G, Przywara S, Diener HC. Features of medication overuse headache following overuse of different acute headache drugs. Neurology 2002;59:1011–4. http://dx.doi.org/10.1212/WNL.59.7.1011. [10] Evers S, Marziniak M. Clinical features, pathophysiology, and treatment of medication-overuse headache. Lancet Neurol 2010;9:391–401. http://dx.doi.org/ 10.1016/S1474-4422(10)70008-9. [11] Mehuys E, Paemeleire K, Van Hees T, Christiaens T, Van Bortel LM, Van Tongelen I, et al. Self-medication of regular headache: a community pharmacy-based survey. Eur J Neurol 2012;19:1093–9. http://dx.doi.org/10.1111/j.1468-1331.2012. 03681.x. [12] Bigal ME, Lipton RB. Overuse of acute migraine medications and migraine chronification. Curr Pain Headache Rep 2009;13:301–7. http://dx.doi.org/10.1007/ s11916-009-0048-3. [13] Starling AJ, Hoffman-Snyder C, Halker RB, Wellik KE, Vargas BB, Dodick DW, et al. Risk of development of medication overuse headache with nonsteroidal antiinflammatory drug therapy for migraine: a critically appraised topic. Neurologist 2011;17:297–9. http://dx.doi.org/10.1097/NRL.0b013e31822d109c. [14] Lantéri-Minet M, Duru G, Mudge M, Cottrell S. Quality of life impairment, disability and economic burden associated with chronic daily headache, focusing on chronic migraine with or without medication overuse: a systematic review. Cephalalgia 2011;31:837–50. http://dx.doi.org/10.1177/0333102411398400. [15] D'Amico D, Grazzi L, Usai S, Raggi A, Leonardi M, Bussone G, et al. Disability in chronic daily headache: state of the art and future directions. Neurol Sci 2011;32:S71–6. http://dx.doi.org/10.1007/s10072-011-0552-1. [16] Linde M, Gustavsson A, Stovner LJ, Steiner TJ, Barré J, Katsarava Z, et al. The cost of headache disorders in Europe: the Eurolight project. Eur J Neurol 2012;19:703–11. http://dx.doi.org/10.1111/j.1468-1331.2011.03612.x. [17] Raggi A, Giovannetti AM, Leonardi M, Schiavolin S, D'Amico D, Curone M, et al. Disability and mood state in patients with episodic and chronic migraine associated to medication overuse. Neurol Sci 2012;33:S169–71. http://dx.doi.org/10.1007/ s10072-012-1076-z. [18] Andrasik F, Grazzi L, Usai S, Kass S, Bussone G. Disability in chronic migraine with medication overuse: treatment effects through 5 years. Cephalalgia 2010;30: 610–4. http://dx.doi.org/10.1111/j.1468-2982.2009.01932.x. [19] Raggi A, Leonardi M, Giovannetti AM, Schiavolin S, Bussone G, Grazzi L, et al. A 14month study of change in disability and mood state in patients with chronic migraine associated to medication overuse. Neurol Sci 2013;34:S139–40. http://dx. doi.org/10.1007/s10072-013-1371-3. [20] Curone M, Tullo V, Savino M, Proietti-Cecchini A, Bussone G, D'Amico D, et al. Outcome of patients with chronic migraine with medication overuse and depression after duloxetine: influence of coexisting obsessive compulsive disorder. Neurol Sci 2013;34:S175–7. http://dx.doi.org/10.1007/s10072-013-1376-y. [21] Bendtsen L, Munksgaard S, Tassorelli C, Nappi G, Katsarava Z, Lainez M, et al. Disability, anxiety and depression associated with medication-overuse headache can be considerably reduced by detoxification and prophylactic treatment. Results from a multicentre, multinational study (COMOESTAS project). Cephalalgia 2014;34: 426–33. http://dx.doi.org/10.1177/0333102413515338. [22] Stewart WF, Lipton RB, Whyte J, Dowson A, Kolodner K, Liberman JN, et al. An international study to assess reliability of the Migraine Disability Assessment (MIDAS) score. Neurology 1999;53:988–94. http://dx.doi.org/10.1212/WNL.53.5.988. [23] Curone M, Tullo V, Mea E, Proietti-Cecchini A, Peccarisi C, Bussone G, et al. Psychopathological profile of patients with chronic migraine and medication overuse: study and findings in 50 cases. Neurol Sci 2011;32:S177–9. http://dx.doi.org/10. 1007/s10072-011-0527-2. [24] Autret A, Roux S, Rimbaux-Lepage S, Valade D, Debiais S. West Migraine Study Group Psychopathology and quality of life burden in chronic daily headache: influence of migraine symptoms. J Headache Pain 2010;11:247–53. http://dx.doi.org/10. 1007/s10194-010-0208-1. [25] World Health Organization. The International Classification of Functioning, Disability and Health-ICF. Geneva: WHO; 2001. [26] Ustun TB, Chatterji S, Kostanjsek N, Rehm J, Kennedy C, Epping-Jordan J, et al. Developing the World Health Organization Disability Assessment Schedule 2.0. Bull World Health Organ 2010;88:815–23. http://dx.doi.org/10.2471/BLT.09.067231.
65
[27] Leonardi M, Raggi A, Bussone G, D'Amico D. Health-related quality of life, disability and severity of disease in patients with migraine attending to a specialty headache center. Headache 2010;50:1576–86. http://dx.doi.org/10.1111/j.1526-4610.2010.01770.x. [28] Leonardi M, Raggi A, Antozzi C, Confalonieri P, Maggi L, Cornelio F, et al. The relationship between health, disability and quality of life in Myasthenia gravis: results from an Italian study. J Neurol 2010;257:98–102. http://dx.doi.org/10.1007/s00415-0095279-z. [29] Quintas R, Alvarez A, Koutsogeorgou E, Cerniauskaite M, Meucci P, Sattin D, et al. The relationship between health-related quality of life and disability in patients with controlled epilepsy: a cross-sectional observational study. Am J Phys Med Rehabil 2012;91:S31–8. http://dx.doi.org/10.1097/PHM.0b013e31823d4db9. [30] Cerniauskaite M, Quintas R, Koutsogeorgou E, Meucci P, Sattin D, Leonardi M, et al. Quality of life and disability in patients with Stroke. Am J Phys Med Rehabil 2012; 91:S39–47. http://dx.doi.org/10.1097/PHM.0b013e31823d4df7. [31] Leonardi M, Raggi A, Pagani M, Carella F, Soliveri P, Albanese A, et al. Relationships between disability, quality of life and prevalence of nonmotor symptoms in Parkinson's disease. Parkinsonism Relat Disord 2012;18:35–9. http://dx.doi.org/10. 1016/j.parkreldis.2011.08.011. [32] Silberstein SD, Lipton RB, Sliwinski M. Classification of daily and near-daily headaches. Neurology 1996;47:871–5. [33] Martin BC, Pathak DS, Sharfman MI, Adelman JU, Taylor F, Kwong WJ, et al. Validity and reliability of the Migraine-Specific Quality of Life Questionnaire (MSQ Version 2.1). Headache 2000;40:204–15. http://dx.doi.org/10.1046/j.1526-4610.2000. 00030.x. [34] Federici S, Meloni F, Mancini A, Lauriola M, Olivetti Belardinelli M. World Health Organisation Disability Assessment Schedule II: contribution to the Italian validation. Disabil Rehabil 2009;31:553–64. http://dx.doi.org/10.1080/ 09638280802240498. [35] D'Amico D, Mosconi P, Genco S, Usai S, Prudenzano AM, Grazzi L, et al. The Migraine Disability Assessment (MIDAS) questionnaire: translation and reliability of the Italian version. Cephalalgia 2001;21:947–52. http://dx.doi.org/10.1046/j.0333-1024. 2001.00277.x. [36] Beck AT, Steer RA, Brown GK. Manual for the Beck Depression Inventory-II. San Antonio: Psychological Corporation; 1996. [37] Silberstein SD, Blumenfeld AM, Cady RK, Turner IM, Lipton RB, Diener HC, et al. OnabotulinumtoxinA for treatment of chronic migraine: PREEMPT 24-week pooled subgroup analysis of patients who had acute headache medication overuse at baseline. J Neurol Sci 2013;331:48–56. http://dx.doi.org/10.1016/j.jns. 2013.05.003. [38] Diener HC, Bussone G, Van Oene JC, Lahaye M, Schwalen S, Goadsby PJ, et al. Topiramate reduces headache days in chronic migraine: a randomized, doubleblind, placebo-controlled study. Cephalalgia 2007;27:814–23. http://dx.doi.org/10. 1111/j.1468-2982.2007.01326.x. [39] Lipton RB, Varon SF, Grosberg B, McAllister PJ, Freitag F, Aurora SK, et al. Onabotulinumtoxin A improves quality of life and reduces impact of chronic migraine. Neurology 2011;77:1465–72. http://dx.doi.org/10.1212/WNL. 0b013e318232ab65. [40] Silberstein SD, Loder E, Forde G, Papadopoulos G, Fairclough D, Greenberg S, et al. The impact of migraine on daily activities: effect of topiramate compared with placebo. Curr Med Res Opin 2006;22:1021–9. http://dx.doi.org/10.1185/ 030079906X104731. [41] Dodick DW, Freitag F, Banks J, Saper J, Xiang J, Rupnow M, et al. Topiramate versus amitriptyline in migraine prevention: a 26-week, multicenter, randomized, double-blind, double-dummy, parallel-group noninferiority trial in adult migraineurs. Clin Ther 2009;31:542–59. http://dx.doi.org/10.1016/j.clinthera. 2009.03.020. [42] Cevoli S, D'Amico D, Martelletti P, Valguarnera F, Del Bene E, De Simone R, et al. Underdiagnosis and undertreatment of migraine in Italy: a survey of patients attending for the first time 10 headache centres. Cephalalgia 2009;29:1285–93. http://dx.doi. org/10.1111/j.1468-2982.2009.01874.x. [43] Bagley CL, Rendas-Baum R, Maglinte GA, Yang M, Varon SF, Lee J, et al. Validating Migraine-Specific Quality of Life Questionnaire v2.1 in episodic and chronic migraine. Headache 2012;52:409–21. http://dx.doi.org/10.1111/j.1526-4610.2011. 01997.x. [44] Rendas-Baum R, Bloudek LM, Maglinte GA, Varon SF. The psychometric properties of the Migraine-Specific Quality of Life Questionnaire version 2.1 (MSQ) in chronic migraine patients. Qual Life Res 2013;22:1123–33. http://dx.doi.org/10.1007/s11136012-0230-7. [45] Raggi A, Giovannetti AM, Schiavolin S, Leonardi M, Bussone G, Grazzi L, et al. Validating the Migraine-Specific Quality of Life Questionnaire v2.1 (MSQ) in Italian inpatients with chronic migraine with a history of medication overuse. Qual Life Res 2014;23:1273–7. http://dx.doi.org/10.1007/s11136-013-0556-9. [46] Headache Classification Subcommittee of the International Headache Society. The international classification of headache disorders: 2nd edition. Cephalalgia 2004; 24:S9–S160. [47] Hedges LV. Distribution theory for Glass's estimator of effect size and related estimators. J Educ Behav Stat 1981;6:107–28. [48] Ware JE. SF-36 health survey update. Spine 2000;25:3130–9. [49] Cole JC, Lin P, Rupnow MF. Validation of the Migraine-Specific Quality of Life Questionnaire version 2.1 (MSQ v. 2.1) for patients undergoing prophylactic migraine treatment. Qual Life Res 2007;16:1231–7. http://dx.doi.org/10.1007/s11136-0079217-1. [50] Yalug I, Selekler M, Erdogan A, Kutlu A, Dundar G, Ankarali H, et al. Correlations between alexithymia and pain severity, depression, and anxiety among patients with chronic and episodic migraine. Psychiatry Clin Neurosci 2010;64:231–8. http://dx. doi.org/10.1111/j.1440-1819.2010.02093.x.
66
A. Raggi et al. / Journal of the Neurological Sciences 348 (2015) 60–66
[51] Raggi A, Covelli V, Pagani M, Meucci P, Martinuzzi A, Buffoni M, et al. Sociodemographic features and diagnoses as predictors of severe disability in a sample of adults applying for disability certification. Int J Rehabil Res 2014;37:180–6. http://dx.doi.org/10.1097/MRR.0000000000000054. [52] Barnett K, Mercer SW, Norbury M, Watt G, Wyke S, Guthrie B. Epidemiology of multimorbidity and implications for health care, research, and medical education: a cross-sectional study. Lancet 2012;380:37–43. http://dx.doi.org/10.1016/S01406736(12)60240-2.
[53] Bigal ME, Lipton RB. Excessive opioid use and the development of chronic migraine. Pain 2009;142:179–82. http://dx.doi.org/10.1016/j.pain.2009.01.013. [54] Buse DC, Pearlman SH, Reed ML, Serrano D, Ng-Mak DS, Lipton RB, et al. Opioid use and dependence among persons with migraine: results of the AMPP study. Headache 2012;52:18–36. http://dx.doi.org/10.1111/j.1526-4610.2011.02050.x. [55] Vinson DR. Treatment patterns of isolated benign headache in US emergency departments. Ann Emerg Med 2002;39:215–22. http://dx.doi.org/10.1067/mem. 2002.121400.
