Validity, reliability, and sensitivity to change of the Thai version of the Migraine-Specific Quality of Life Questionnaire version 2.1
Article Category: Original article
Published Online: Mar 21, 2018
Page range: 331 - 342
DOI: https://doi.org/10.1515/abm-2018-0005
Keywords
© 2017 Thanin Asawavichienjinda, Pongpat Vorasayan, Jirawadee Noiwattanakul, Kammant Phanthumchinda
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 License.
Migraine is a common neurological disorder characterized by unilateral throbbing headache with moderate-to-severe pain intensity lasting for 4–72 h and aggravated by head motion, bright light, and noise. Migraine is also associated with nausea and vomiting [1]. These migraine symptoms lead to substantial disability, including physical, psychological, and social aspects [2]. Moderate-to-severe head pain of migraine and its associated symptoms during attacks disable and reduce the ability of migraineurs to work and function effectively. The migraine attack interval, even during the headache-free period, still causes migraineurs to be frustrated, worried, and afraid of upcoming migraine attacks. Migraineurs may be reluctant to participate in social activities that may trigger their migraine attacks. Migraine therefore has a significant impact on the quality of life (QoL) of migraineurs [3]. Most migraineurs will seek medicine to terminate pain during the attack and some to prevent attacks if necessary [4].
Currently, there are at least 7 scales to measure QoL, specifically for people with migraine, and these scales can be categorized into 3 groups. The first group is applied to adults and includes (1) the QoL questionnaire created ad hoc by Cavallini et al. without any validity and reliability test [5], (2) the Migraine-Specific Quality of Life Questionnaire (MSQOL) for long-term impact of migraine without a specific period [6], (3) the MSQ developed by Jhingran et al. from GlaxoWellcome for long-term impact of migraine for a specific period of 4 weeks [7], (4) the 6-item Headache Impact Test (HIT-6) for nonspecific type of headache including migraine at the initial development [8], and (5) the Comprehensive Headache-Specific Quality of Life Questionnaire for migraine and tension-type headache only tested for validity [9]. The second group is applied to adolescents only, and uses the (6) Quality of Life Headache in Youth (QLH-Y) scale for chronic headache, including migraine with a 1 week impact [10]. The third group is applied to measure the efficacy of acute treatment and uses the brief 24-hour (7) MSQ for short-term impact [11]. There have been other questionnaires measuring some parts of the QoL domains, such as the Headache Disability Inventory, Headache Impact Questionnaire, and Migraine Disability Assessment (MIDAS) questionnaire.
Migraine-specific QoL questionnaires that are most commonly used are the MSQ and HIT-6. The MSQ was first developed using a 4-phase approach, which included item selection starting with 48 items relevant to migraine, item reduction with 25 items retained, pretesting, and item finalization with 16 items covering 3 dimensions, and a test for validity and reliability [7]. This version of questionnaire is known as version 1.0. After clinical testing in people with episodic migraine, revisions were made to improve discriminant and convergent validity and reliability [12, 13]. This revised version of the questionnaire is known as version 2.0. The factorial structure of the version 2.0 of the MSQ was then evaluated by using confirmatory factor analysis [13, 14], and the results determined a 14-item version of the statistically improved MSQ [13]. This revised version is known as MSQ version 2.1 (MSQv2.1), and is composed of 14 items in 3 dimensions covering functional status specific to migraine role restrictive (RR), role preventive (RP), and emotional function (EF) [13]. Each item of the MSQv2.1 has a 6-point frequency scale ranging from none of the time, a little bit of the time, some of the time, a good bit of the time, most of the time, and all of the time, which are assigned scores of 1–6, respectively [15]. Construct validity was also tested with migraine symptoms with low-to-modest correlations (
The HIT-6 was originally composed of 54 items and was analyzed using an item response theory. Subsequently, 10 candidate items were selected and 35 more items were added as suggested by clinicians. This questionnaire was evaluated for clinical validity by telephone interview of people sampled from a migraine medication prescription database who were prescribed medication in the previous year. The people in the migraine medication prescription database might not all be people with migraine. Headache including other types of primary headache can be relieved by migraine medications. Six items then were selected for the new HIT short form (HIT-6) and were again tested for construct validity, including convergent validity and discriminant validity, and for reliability, including internal consistency and test–retest reliability [8]. The validity and reliability of the HIT-6 was also tested on participants with episodic and chronic migraines [19]. In addition, the HIT-6 was tested for chronic migraine [20] and applied in both clinical research [21, 22] and clinical practice [23, 24]. However, one clinical research study demonstrated that topiramate significantly reduced the mean number of monthly migraine days compared with placebo, but no significant intergroup difference was found using HIT-6 [21]. The other study demonstrated that botulinum toxin significantly improved headache severity and had improved HIT-6 score compared with placebo [22]. Among these 2 studies, one study did not evaluate how sensitive the score was to change [23]. The other study demonstrated that headache severity correlated with HIT-6 scale score; however, this study included people with episodic migraine, migrainous disorders, and other forms of episodic headache [24].
Not all migraine characteristics have clinically important outcomes. QoL assessment is a clinically important outcome for assessing a treatment effect. If the migraine characteristics improve, the MSQv2.1 score also improves, and this is known as sensitivity to change.
In Thailand, there were no specific questionnaires to evaluate QoL for people with migraine. The present study had 2 objectives. The primary objective aimed to evaluate concurrent validity and test–retest reliability of the Thai version of MSQv2.1. The secondary objective aimed to evaluate internal consistency and sensitivity to change of the Thai version of MSQv2.1 as a pilot study.
This was a prospective study conducted at the Chulalongkorn Comprehensive Headache Centre of King Chulalongkorn Memorial Hospital, Faculty of Medicine, Chulalongkorn University, in 2009 and was approved by the institutional review board of Faculty of Medicine, Chulalongkorn University (No. 034/52).
Two groups of patients were included in the present study. One group was used to test for concurrent validity, test–retest reliability, and internal consistency of the Thai version of MSQ 2.1, and the other group was used to test clinical sensitivity of the Thai version of MSQ 2.1 to change. Both groups of patients were those who visited the Chulalongkorn Comprehensive Headache Centre and were diagnosed as having migraine with or without aura. The inclusion criteria were patients aged 18–65 years who were literate in Thai and gave their written informed consent to participate in this study after receiving an invitation in a consecutive manner. Migraine diagnostic criteria and classification were based on the International Classification of Headache Disorders, first or second edition, that depended on headache characteristics received before or after 2004 [25, 26].
We received permission from GSK to use the original English questionnaire of the MSQv2.1 for translation and for use and validation of the Thai version. The MSQv2.1 was translated into Thai by the first author (TA). The following process was conducted to validate the Thai version of MSQv2.1 as equivalent to the original English version of MSQv2.1. The Thai version was translated back into English by a person with a high degree of English proficiency who was blinded to the original version. Subsequently, the original English version and the English version from the back translation were compared. If at any point the 2 English versions were different, the processes of translation into Thai and back translation into English were repeated until both English versions were identical in content, so-called “content equivalence”. Ultimately, the Thai version of MSQv2.1 was used with cultural modifications by an active informant experienced with migraine headache to simplify the Thai wording for ease of understanding and application. This process was called language equivalence. The Thai version of MSQv2.1 was then tested for concurrent validity with migraine characteristics including frequency of attacks per 4 weeks, average pain during per attack (measured in hours), average pain severity score measured using a numerical rating scale (NRS), and associated symptoms and headache severity in the previous 4 weeks. The headache severity was on a scale of mild (does not inhibit work or other activities), moderate (inhibits but does not prohibit), or severe (prohibits work and other activities). Each item on the Thai version of MSQv2.1 had a 6-point scale ranging from none of the time, a little bit of the time, some of the time, a good bit of the time, most of the time, and all of the time, which were assigned scores of 1–6, respectively, corresponding to the original version [15]. Raw dimension scores for RR, RP, and EF were computed as the sum of item responses, and overall QoL score were computed as the sum of the total item responses, and all were rescaled to a 0–100 scale, where higher scores indicated a better health-related QoL [15].
The Chulalongkorn Comprehensive Headache Centre of King Chulalongkorn Memorial Hospital has been established since 2000. This center registers every patient who visits and records every visit using a standardized patient record form. There are 2 types of patient record forms: initial and follow-up. The initial form records patient demographics and socioeconomic data and history of headache including frequency (episodes per 4 weeks), pain duration (hours), pain severity score measured using NRS [zero (no pain) to 10 (maximum pain)], associated symptoms and headache severity, comorbidity, and previous treatments. The follow-up form records the history of headache, efficacy, adverse effects of acute pain/headache prophylactic medications (if any), and current treatments. The center also treats patients individually with prescription for acute pain and prophylactic medications (if any) and gives them specific advice for their migraine management problems. The patients also receive a headache diary that is used to educate them. The diary contains information on headache characteristics that can differentiate primary from secondary headache, the characteristics of migraine headache, a list of precipitating factors, how to control pain during an attack with nonpharmacological and pharmacological treatments, how to prevent pain with nonpharmacological and pharmacological treatments, an explanation of medication overuse and why compliance with medications is needed, and a schedule for follow-up.
The Thai version of MSQv2.1 was applied to migraine patients in a consecutive manner. The participants completed the first questionnaire by self-administration. The first questionnaire included 2 parts. The first part asked about their history of headache in the previous 4 weeks, and the second part was the Thai version of MSQv2.1. Before patients left the center, they received a second questionnaire to be completed at home within the following 2 weeks. The second questionnaire was similar to the first questionnaire with the exception of an additional question inquiring about whether the QoL at the time of the first and second questionnaires was the same. Enclosed along with the blank second questionnaires were stamped envelopes addressed to the third author (JN) for participants to conveniently mail back the second questionnaire after completion. At 2 weeks from the time of the first completed questionnaire, the third author (JN) reminded participants via telephone call to complete and return the second questionnaire. Standard post in Thailand takes an average of 2 weeks for delivery. Therefore, if the second questionnaire was not received or in the event that the received second questionnaire was incomplete or the answer was different in the additional question, the other consecutively eligible patients at the center were invited to replace previous patients until the number of participants met the assigned sample size. For the reliability test, the test–retest reliability was compared between the first and the second completed Thai version of MSQv2.1. The internal consistency of the 2 completed Thai versions of MSQv2.1 was tested for each occasion.
After tests for concurrent validity, test–retest reliability, and internal consistency of the Thai version of MSQv2.1 were complete, the Thai version of MSQv2.1 was then tested for sensitivity to change to history of headache in another group of migraineurs at the Chulalongkorn Comprehensive Headache Centre. The history of headache of the people who visited the center was applied at an arbitrary point that we called time zero and was followed for 4 weeks (so-called “week 4 application”) and 8 weeks (so-called “week 8 application”). The history of headache and the Thai version of MSQv2.1 scores including those for RR, RP, and EP and overall QoL score were compared between 2 periods: week 4 and time zero, week 8 and time zero, and week 8 and week 4. The difference in the history of headache between the 2 periods was compared to difference in scores for the dimensions RR, RP, and EF and overall QoL score of the Thai version of MSQv2.1 between the 2 periods to test whether or not the Thai version of MSQv2.1 was sensitive to change.
Concurrent validity of the Thai version of MSQv2.1 was tested with migraine characteristics using Spearman’s correlation coefficient for ordinal data and a rank-biserial correlation coefficient for correlation between nominal and ordinal data. Reliability of the questionnaire was tested using Spearman’s correlation coefficient for test–retest reliability and using Cronbach’s α for internal consistency. The history of headache 4 weeks before completion of the first and second questionnaires and the scores of the Thai version of MSQv2.1 were summarized with median and interquartile range. Associated symptoms and adverse effects of acute pain or prophylactic medications were categorized into dichotomous data on the basis of “yes” or “no” answers. The headache severity with the scale of mild (does not inhibit work or other activities), moderate (inhibits but does not prohibit), or severe (prohibits work and other activities) was categorized into dichotomous data as follows: mild and moderate or severe, with data summary of percentage. Clinical testing for the difference in the history of headache and the QoL scores between 2 periods was conducted using the McNemar
The sample size for test–retest reliability was based on the Spearman’s correlation coefficient with a type I error of 0.05 and power of 0.88 (equivalent to 0.8 for a Pearson correlation coefficient), and anticipated correlation coefficient of 0.5 was calculated according to Siegel’s book of nonparametric statistics [27]. Therefore, the total sample size required was 29 participants. We did not calculate the sample size for the internal consistency and clinical testing for sensitivity to change. These were analyzed in a pilot study.
There were 46 participants who completed the first questionnaire. A total of 37 participants mailed back the second questionnaire. Of them, 30 participants reported the same QoL between the first and second questionnaires and were eligible for reliability test analyses (
Baseline characteristics of the 30 eligible participants Based on UNESCO’s International Standard Classification of Education 1997 Counted more than once SD, standard deviation; UNESCO, United Nations Educational, Scientific and Cultural OrganizationCharacteristic Value Age, years (SD) 42.4 (13) Female sex, n (%) 29 (97) Educational level Primary education 8 (27) Upper secondary education 3 (10) Postsecondary nontertiary education 5 (17) First and second stage of tertiary education 14 (47) Classification of migraine, n (%) Migraine without aura 27 (90) Migraine with aura 3 (10) Duration of having migraine (years) Median 5.3 Interquartile range 8.0 Average pain duration per attack (hours) Median 3.8 Interquartile range 4.0 Average pain severity, n (%) Mild 1 (3) Moderate 23 (77) Severe 6 (20) Associated symptoms level Nausea 19 (63) Photophobia 17 (57) Phonophobia 15 (50) Vomiting 13 (43)
History of headache 4 weeks before completion of the first and second questionnaires MSQv2.1, Migraine-Specific Quality of Life Questionnaire version 2.1; NRS, numerical rating scale; NS, not significantHeadache information/score of Thai version of the MSQv2.1 First questionnaire Second questionnaire Frequency of attacks 4 weeks before (episodes) Median 4.0 4.0 NS Interquartile range 7.2 7.0 Average pain duration per attack (hours) Median 2.0 2.0 NS Interquartile range 3.6 2.9 Average pain severity score (NRS) Median 4.5 5.0 NS Interquartile range 4.0 5.2 No. of associated symptoms, n (%) 15.0 (50) 16.0 (53) NS Headache severity, n (%) Mild 22 (73) 20 (67) NS Moderate or severe 8 (27) 10 (33)
Concurrent validity of the Thai version of MSQv2.1 score with the migraine characteristics 4 weeks before completion of the first questionnaire EF, emotional function; MSQv2.1, Migraine-Specific Quality of Life Questionnaire version 2.1; NRS, numerical rating scale; QoL, quality of life; RP, role function – preventive; RR, role function – restrictiveMigraine characteristic Thai version of the MSQv2.1 score RR RP EF Overall QoL Frequency of attacks 4 weeks before (episodes) −0.42 −0.41 −0.39 −0.50 Average pain duration per attack (hours) −0.47 −0.34 −0.48 −0.48 Average pain severity score (NRS) −0.58 −0.53 −0.50 −0.62 No. of associated symptoms 0.24 0.35 0.29 0.30 Mild degree of headache severity −0.27 −0.43 −0.50 −0.42
Score of each item and domain of the Thai version of the MSQv2.1 between the first and second questionnaires Spearman’s correlation coefficient EF, emotional function; MSQv2.1, Migraine-Specific Quality of Life Questionnaire version 2.1; QoL, quality of life; RP, role function – preventive; RR, role function – restrictiveScore of Thai version of the MSQ 2.1 First questionnaire Second questionnaire RR – median score (interquartile range) Item 1 family interfered with how well you dealt with family, friends, and others 3.5 (2.2) 4.0 (2.0) 0.68 <0.001 Item 2 leisure interfered with your leisure time activities, such as reading and exercising 3.0 (3.0) 4.0 (2.0) 0.58 0.001 Item 3 activity – had difficulty in performing work or daily activities 4.0 (2.2) 4.0 (2.0) 0.77 <0.001 Item 4 work – kept you from getting as much done at work or at home 4.0 (2.0) 3.5 (2.0) 0.66 <0.001 Item 5 concentration – limited your ability to concentrate on work or daily activities 3.5 (3.0) 4.0 (2.2) 0.61 <0.001 Item 6 tired – left you too tired to do work or daily activities 3.0 (2.3) 4.0 (2.0) 0.71 <0.001 Item 7 energy – limited the number of days you have felt energetic 3.0 (2.2) 4.0 (2.0) 0.72 <0.001 RR dimension scores (rescale to 0–100) – median (interquartile range) 71.4 (39.2) 77.1 (23.6) 0.83 <0.001 RP – median score (interquartile range) Item 8 cancel – cancel work or daily activities 4.0 (2.0) 4.0 (3.0) 0.65 <0.001 Item 9 help – need help in handling routine tasks 5.0 (1.0) 4.0 (1.0) 0.60 <0.001 Item 10 stop – stop work or daily activities 4.0 (2.0) 4.0 (2.0) 0.62 <0.001 Item 11 social – not able to go to social activities 4.0 (2.0) 4.5 (2.0) 0.66 <0.001 RP dimension scores (rescale to 0–100) – median (interquartile range) 87.5 (35.0) 80.0 (35.0) 0.76 <0.001 EF – median score (interquartile range) Item 12 frustrated – felt fed up or frustrated 3.5 (3.0) 4.0 (2.0) 0.68 <0.001 Item 13 burden – felt like you were a burden on others 5.0 (2.0) 4.0 (1.0) 0.68 <0.001 Item 14 afraid – afraid of letting others down 5.0 (1.0) 5.0 (1.0) 0.56 0.001 EF dimension scores (rescale to 0–100) – median (interquartile range) 80.0 (30.0) 83.3 (21.6) 0.65 <0.001 Overall QoL score (rescale to 0–100) – median (interquartile range) 76.7 (29.3) 80.0 (24.6) 0.83 <0.001
Flow diagram for recruitment of participants to test for concurrent validity, test–retest reliability, and internal consistency of the Thai version of the Migraine-Specific Quality of life Questionnaire version 2.1; QoL, quality of lifeFigure 1
Clinical testing of 12 patients at the center with a diagnosis of migraine was conducted to determine the sensitivity of the Thai version of MSQv2.1 to change. Of the 12 patients, 1 patient was lost to follow-up. Data from 11 migraineurs were ultimately included for analysis (a flow diagram is shown in
Flow diagram for recruitment of participants to test for sensitivity of the Thai version of the Migraine-Specific Quality of Life Questionnaire version 2.1 to changeFigure 2
Baseline characteristics of the 11 participants for clinical testing SD, standard deviationCharacteristic Value Age, years (SD) 39.6 (14.7) Female sex, n (%) 10 (91) Educational level, n (%) Primary education 2 (18) Upper secondary education 3 (27) Postsecondary nontertiary education 5 (46) First stage of tertiary education 1 (9) Occupation, n (%) Employee 7 (64) Studying 3 (27) Business 1 (9) Duration of having migraine, years Median 3.0 Interquartile range 8.0
Comparison of the change in the history of headache between 2 periods of 4 weeks and the score change in the Thai version of the MSQv2.1 Counted more than once Counted more than once Counted more than once Counted more than once EF, emotional function; MSQv2.1, Migraine-Specific Quality of Life Questionnaire version 2.1; NA, not applicable because of a zero value; NRS, numerical rating scale; NSAIDs, nonsteroidal anti-inflammatory drugs; QoL, quality of life; RP, role function – preventive; RR, role function – restrictiveHeadache history Time zero (0) Week 4 (4) Week 8 (8) Frequency of attacks 4 weeks before (episodes) Median 12.0 8.0 2.0 0.75 0.18 0.06 Interquartile range 12.0 18.0 11.0 Average pain duration per attack, hours Median 3.0 1.5 1.0 1.00 Interquartile range 3.5 10.5 2.0 Average pain severity score (NRS) Median 6.0 5.0 5.0 0.34 1.00 Interquartile range 1.5 4.0 6.0 Headache severity, n (%) Mild 6 (55) 9 (82) 11 (100) 0.38 0.06 0.50 Moderate or severe 5 (45) 2 (18) 0 (0) Location of pain Ocular 8 (73) 7 (64) 3 (27) 1.00 0.09 0.19 Back of neck 6 (55) 3 (27) 1 (9) 0.38 0.06 0.58 Forehead 2 (18) 0 (0) 2 (18) NA 1.00 NA Temple 10 (91) 8 (73) 8 (73) 0.58 0.58 1.00 Occiput 8 (73) 5 (45) 5 (45) 0.38 0.36 1.00 Vertex 2 (18) 0 (0) 0 (0) NA NA NA Whole head 5 (45) 2 (18) 1 (9) 0.36 0.14 1.00 Side Unilateral, constant 2 (18) 6 (55) 4 (36) 0.10 0.63 0.66 Bilateral 5 (45) 5 (45) 2 (18) 1.00 0.36 0.36 Unilateral with alternating 6 (55) 2 (18) 3 (27) 0.10 0.36 1.00 Unilateral and then bilateral 2 (18) 0 (0) 0 (0) NA NA NA Pain quality Dull aching/pressing 4 (36) 4 (36) 2 (18) 1.00 0.63 0.63 Throbbing 9 (82) 6 (55) 6 (55) 0.36 0.36 1.00 Sharp or stabbing 1 (9) 0 (0) 3 (27) NA 0.58 NA Associated symptoms Nausea/vomiting 9 (82) 2 (18) 1 (9) 0.01 0.002 1.00 Photophobia/phonophobia 10 (91) 2 (18) 3 (27) 0.002 0.007 1.00 Precipitating factors, n (%) No 2 (18) 2 (45) 5 (45) 1.00 0.36 0.36 Taking acute pain medication, n (%) Yes 11 (100) 11 (100) 10 (91) NA NA NA Type of acute pain medication, n (%) Simple analgesic/NSAIDs 10 (91) 10 (91) 10 (91) 1.00 1.00 1.00 Migraine-specific medication 1 (9) 1 (9) 1 (9) 1.00 NA NA Effect of acute pain medication, n (%) Pain improvement 8 (73) 9 (82) 9 (82) 1.00 1.00 1.00 Adverse effect of acute pain medication, n (%) Yes 2 (18) 0 (0) 0 (0) NA NA NA Taking prophylactic medication, n (%) Yes 3 (27) 10 (91) 11 (100) 0.007 NA NA Type of prophylactic medication, n (%) Antiepileptic drugs 1 (9) 7 (64) 7 (64) 0.02 0.02 1.00 Calcium antagonists 2 (18) 1 (9) 1 (9) 1.00 1.00 1.00 Antidepressants 0 (0) 2 (18) 3 (27) NA NA 1.00 Adverse effect of prophylactic medication, n (%) Yes 0 (0) 2 (18) 1 (9) NA NA 1.00 Adverse effect of both acute pain and prophylactic medications, n (%) Yes 2 (18) 2 (18) 1 (9) 1.00 1.00 1.00 RR – median score (interquartile range) 48.6 (28.6) 85.7 (22.9) 85.7 (20.0) 0.02 0.002 0.06 RP – median score (interquartile range) 70.0 (25.0) 95.0 (15.0) 100.0 (15.0) 0.02 0.002 1.00 EF – median score (interquartile range) 73.3 (38.0) 86.7 (20.0) 93.3 (20.0) 0.22 0.12 1.00 Overall score of MSQv2.1 – median score (interquartile range) 62.9 (27.1) 88.6 (18.6) 88.6 (18.6) 0.02 0.001 0.11
Headache severity was reduced to mild with a higher proportion in week 4 (82%) and week 8 (100%) compared with time zero (55%), although the reductions were not significant. Migraine-associated symptoms, nausea or vomiting, and photophobia or phonophobia significantly decreased from time zero (82%
and 91%, respectively) to week 4 (18% and 18%, respectively) and week 8 (9% and 27%, respectively) with a significant difference between week 4 and time zero (
Clinical testing for sensitivity of the Thai version of the MSQv2.1 to change of the headache history between 2 periods of 4 weeks EF, emotional function dimension; MSQv2.1, Migraine-Specific Quality of Life Questionnaire version 2.1; NRS, numerical rating scale; QoL, quality of life; RP, role preventive dimension; RR, role restrictive dimensionHistory of headache Correlation coefficient for median improvement RR RP EF Overall QoL Between week 4 and time zero Median reduction in frequency of attacks 4 weeks before −0.32 −0.23 −0.22 −0.24 Median reduction in average pain duration per attack, hours −0.13 0.40 0.07 0.12 Median reduction in average pain severity score (NRS) 0.24 0.16 −0.33 0.16 Improvement in associated symptoms with nausea/vomiting to no nausea/vomiting −0.12 −0.18 0.31 −0.09 Improvement in associated symptoms with photophobia/phonophobia to no photophobia/phonophobia 0.32 0.09 0.27 0.29 Change in proportion to take prophylactic medications 0.24 −0.21 0.25 −0.21 Improvement in headache severity from moderate to severe to mild 0.71 0.77 0.40 0.77 Between week 8 and time zero Median reduction in frequency of attacks 4 weeks before 0.15 0.20 −0.21 0.15 Median reduction in average pain duration per attack, hours 0.02 −0.21 −0.59 −0.12 Median reduction in average pain severity score (NRS) 0.34 0.17 0.01 0.28 Improvement in associated symptoms with nausea/vomiting to no nausea/vomiting 0.33 0.13 −0.16 0.25 Improvement in associated symptoms with photophobia/phonophobia to no photophobia/phonophobia 0.00 0.24 −0.24 0.06 Change in proportion to take prophylactic medications −0.45 −0.26 0.66 −0.38 Improvement in headache severity from moderate to severe to mild 0.71 0.52 0.29 0.77 Between weeks 8 and 4 Median reduction in frequency of attacks 4 weeks before 0.07 −0.15 −0.61 −0.12 Median reduction in average pain duration per attack, hours 0.07 0.57 −0.02 0.11 Median reduction in average pain severity score (NRS) −0.39 −0.50 −0.35 −0.62 Improvement in associated symptoms with nausea/vomiting to no nausea/vomiting 0.15 0.15 0.03 0.22 Improvement in associated symptoms with photophobia/phonophobia to no photophobia/phonophobia 0.20 −0.05 0.40 0.30 Change in proportion to take prophylactic medications −0.50 −0.30 0.20 −0.40 Improvement in headache severity from moderate to severe to mild 0.59 0.69 0.26 0.67
To our knowledge, this is the third reported study for validity and reliability of a non-English version of the MSQv2.1. The first study was conducted in Iran using 106 patients with chronic migraine and episodic migraine to test the validity and reliability of a Persian version of MSQv2.1 in neurology clinics [28]. The investigators reported an internal consistency with Cronbach’s α of 0.92, test–retest reliability with intraclass correlation coefficients of 0.41–0.50, and convergent validity to a 36-Item Short Form Survey with a Pearson correlation coefficient of 0.41–0.46. The second study was conducted in Italy using 182 patients with chronic migraine and a history of medication overuse to test validity and reliability of an Italian version of MSQv2.1 in an inpatient department at a neurology institute [29]. The investigators reported internal consistency with a Cronbach’s α of between 0.85 for the EF domain and 0.92 for the RR domain and construct validity to the MIDAS questionnaire with Pearson correlation coefficients with
This study was limited by its small sample size for testing of internal consistency and sensitivity to change, lack of correction of
The Thai version of MSQv2.1 had concurrent validity, acceptable internal consistency, moderate-to-strong test–retest reliability, and strong correlation between improvement in headache severity and overall QoL score. The Thai version of MSQv2.1 would be helpful to measure clinical outcomes in clinical practice. A future study using a larger sample size and longer follow-up is required to fully test internal consistency and sensitivity to change.