|Ahead of print publication
The hindi version of mcgill pain questionnaire: A cross-cultural adaptation study in rheumatoid arthritis
Riddhi A Shroff, Twinkle Y Dabholkar
Department of Musculoskeletal Physiotherapy, School of Physiotherapy, D.Y. Patil University, Navi Mumbai, Maharashtra, India
|Date of Submission||20-Jul-2020|
|Date of Acceptance||01-Dec-2020|
Riddhi A Shroff,
B/19, Modi Nagar, Mathuradas Road, Kandivali West, Mumbai - 400 067, Maharashtra
Source of Support: None, Conflict of Interest: None
Background: Pain assessment plays a major role in guiding treatment decisions in rheumatoid arthritis (RA). The long-form McGill Pain Questionnaire (LF-MPQ) is a valid and reliable tool for multidimensional pain assessment. However, a validated Hindi version of the LF-MPQ is not available for use in India. The objective was to develop an equivalent version of the LF-MPQ in native Hindi language that is cross-culturally adapted to suit Indian patients.
Methodology: After acquiring permission from the University Ethics Board, guidelines given by the Mapi Research Trust were followed and LF-MPQ was translated to culturally appropriate Hindi version. After translation, the questionnaire was subjected to testing of clinimetric properties in fifty patients with RA.
Results: The Hindi version of the LF-MPQ demonstrated high levels of internal consistency (Cronbach's alpha range: 0.74–0.80) and reliability (intraclass correlation coefficient range: 0.76–0.84). The Hindi version of LF-MPQ also demonstrated moderate construct and concurrent validity when tested with Visual Analog Scale (Pearson r: 0.78) and Disease Activity Score in 28 joints (Pearson r: 0.77), respectively.
Conclusion: The Hindi version of the LF-MPQ was found to be reproducible and valid for the assessment of pain in patients with RA.
Keywords: Hindi version, McGill Pain Questionnaire, pain, rheumatoid arthritis, validity and reliability
| Introduction|| |
Pathology and pain intensity do not exhibit a direct relationship, but instead, chronic pain is shaped by a myriad of biomedical, behavioral factors and psychosocial factors (eg. patients' beliefs, expectations, mood). Thus, assessment each of these domains through a comprehensive evaluation of the person with chronic pain is essential for treatment decisions and to facilitate optimal outcomes. Arthritis pain is traditionally evaluated from a biomedical perspective, but there is increasing evidence that psychological factors have an important role in patients' adjustment to arthritis pain. Hence, there is a need to adequately address the psychosocial impact of illness by health-care practitioners in patients with arthritis-related pain.
The long-form McGill Pain Questionnaire (LF-MPQ) developed by Melzack and Torgerson is considered as a gold standard multidimensional tool, which isuseful for comprehensive pain assessment. The MPQ, originally published in English language, comprises 78 pain descriptors that are categorized into 4 pain domains (sensory, affective, evaluative, and miscellaneous) and total 20 subclasses. Each subclass contains two to six descriptors that have an assigned rank of 1–6 reflecting the intensity. MPQ has been translated into many languages. Hindi is spoken by over 400 million people in India. Although the English version is available, an equivalent version of LF-MPQ in Hindi would allow the patients to convey adequately about their pain experience to clinicians.
Hence, the primary objective was to develop a version of the LF-MPQ in native Hindi language that is cross-culturally adapted to suit Indian patients. This will enable clinicians and researchers to assess patient's pain in their own cultural context with questions which are relevant and meaningful . The clinimetric properties of the translated Hindi version such as internal consistency, reproducibility, validity, and responsiveness were also tested to determine its usability in RA patients.
| Methodology|| |
The study protocol was approved by the university ethics committee. The study was conducted in two stages: the first stage was formulation of a cross-culturally adapted Hindi version of the LF-MPQ and the second stage was testing the clinimetric properties of the Hindi version of LF-MPQ.
Translation and cross-cultural adaptation
After securing permission from the Mapi Research Trust (https://mapi-trust.org/), the English LF-MPQ was translated into Hindi following linguistic validation guidelines by Mapi for translation comprising the following stages. The first step in the process was forward translation, wherein two independent translators translated the questionnaire to Hindi. The second step was synthesis; at this stage, a third unbiased person – coordinator – resolved the discrepancy and a common version (Version 1) of the Hindi questionnaire was produced with consensus of all three. The third step was back translation; at this stage, Version 1 of the Hindi questionnaire was translated into English by two translators blinded to the original version producing back translation 1 and back translation 2. The fourth step was expert committee review, and in this step, four health professionals and translators reviewed the original version; all the translated versions and developed a prefinal version for pilot testing. The translators were asked to achieve the same version of the source language (English) and target (Hindi) words, and translation was taking into consideration four areas: semantic equivalence, idiomatic equivalence, experiential equivalence, and conceptual equivalence. The final questionnaire was formulated after examining all such equivalences. Fifth step: the prefinal version was tested in 30 patients with pain due to musculoskeletal disorders. The results of the pilot study showed that the Hindi version of LF-MPQ was well understood by subjects. The final version contains 76 pain descriptors as compared to 78 descriptors in the original LF-MPQ.
Testing of clinimetric properties
The final cross-culturally adapted Hindi version of the questionnaire was administered to 50 patients with rheumatoid arthritis (RA), who were recruited from the Rheumatology Outpatient Department of D.Y. Patil Hospital, Nerul, Navi Mumbai. The patients were enrolled after obtaining informed consent. To be eligible to participate in the study, the patients had to be able to speak, read, and write in Hindi, be diagnosed case of RA according to ACR/EULAR 2010 criteria with score of >6 and be in the age group of 18–60 years. As this was a pilot study, we enrolled 50 RA patients. Moreover, as per the guidelines issued by the quality criteria for health status questionnaires, a minimum of 50 patients was required for an appropriate analysis of construct, criterion validity, and reproducibility. All patients completed the Hindi version of LF-MPQ and pain Visual Analog Scale (VAS) on a 10-cm scale. The patient rated the intensity of pain on the VAS on a score between 0 and 10: score of 0 meaning “no pain” and score of 10 being “worst possible pain.” The Disease Activity Score in 28 joints (DAS-28) was administered to the patient by the clinician. To test reliability, the Hindi version of LF-MPQ was repeated at 48 h.
The clinimetric assessment consisted of validity, reliability, and internal consistency. Validity refers to the extent to which an instrument measures what it intends to measure. Construct validity establishes the ability of an instrument to measure an abstract construct (in this case pain) and the degree to which the instrument reflects theoretical components of the construct. Construct validity is assessed by demonstrating a positive correlation between measures of the original concept (in this case scores on the Hindi LF-MPQ) and those of other concepts (here in this study pain VAS score) to which the original concept is known to be positively related. Hence, in our study, pain VAS scores were utilized for testing construct validity of Hindi LF-MPQ, as both assess the similar construct of pain. Similar to pain VAS, a high total LF-MPQ score indicates a higher level of pain.
Concurrent validity establishes validity when two measures are taken at the relatively same time.
The DAS-28, a measure of RA disease activity, was employed to test concurrent validity. Lower DAS-28 indicates better disease control, while active disease and severe pain are often associated with a higher score. Pearson's test for correlation was used for assessing the construct and concurrent validity of the Hindi LF-MPQ scores, namely Pain Rating Index (PRI) total and present pain intensity (PPI) scores with VAS score and DAS-28, respectively.
Internal consistency was calculated for each domain of the Hindi version of LF-MPQ separately, i.e., sensory, affective, and miscellaneous using Cronbach's alpha. Cronbach's alpha values are considered adequate if they range between 0.70 and 0.95.
Internal consistency means the intercorrelation among a set of test items that are used to measure a single construct/domain of a questionnaire. Internal consistency statistics estimate how consistently individuals respond to the items within a scale. The word “scale” is used here to reflect a collection of survey items that are intended to measure the unobserved concept.
Reproducibility was assessed in our study with a test–retest design using measures of reliability. Reliability is the extent to which a measurement is consistent and free from error. A reliable instrument is one that will perform with predictable consistency under set conditions. To assess reliability, the Hindi version of the LF-MPQ scale was re-administered at an interval of 24–48 h. Reliability was evaluated in our study using the intraclass correlation coefficient (Type 2-1 agreement) with 95% confidence interval as suggested by Tewree et al. where 0.75–0.90 as substantial reproducibility.
| Results|| |
The final form of cross-culturally adapted Hindi version of LF-MPQ consisted of 76 descriptors, separated into four domains (categories), namely sensory, affective, evaluative, and miscellaneous. The Pain Rating Index sensory (PRI-S) comprises 1–10 groups, with a total score from 0 to 40. The PRI affective (PRI-A) comprises 11–15 groups, with a total score from 0 to 14. The PRI evaluative (PRI-E) comprises group 16, with a total score from 0-5. The PRI miscellaneous (PRI-M) comprises 17–20 groups, with a total score from 0 to 17. The PRI total (PRI-T) comprises a total score of 0 to 76. During cross-cultural adaptation, the word boring and drilling had semantic equivalence in Hindi language, due to which a common descriptor was retained. Furthermore, Hindi descriptor drawing was removed from the final version of the questionnaire due to no appropriate word for translation. Hence, two words were deleted and the final Hindi version had total 76 descriptors. The translation process reflected equivalence between Hindi and English versions.
The Hindi version of LF-MPQ was administered to 50 RA patients whose mean age was 42 years and 37 were women. The baseline demographic characteristics are given in [Table 1]. As per the original scoring system of LF-MPQ, there are three score types, i.e., number of words chosen (NWC), PRI, and PPI. All the three scoring types were utilized in our analysis.
|Table 1: Demographic and baseline characteristics of study participants (n=50)|
Click here to view
NWC was obtained by counting the number of words selected by the respondent. In our study, the NWC ranged from 0 to 76.
The PRI was calculated on the basis of the rank values of the words. In this scoring system, the word in each subclass inferring the least pain is assigned a value of 1, the next word assigned a value of 2, and so forth. The rank values of the words chosen by a patient were added to obtain separate scores for PRI-S (subclasses 1–10), PRI-A (subclasses 11–15), PRI-E (subclass 16), and PRI-M (subclasses 17–20) words, in addition to providing a total score, i.e., PRI-T (subclasses 1–20).
PPI score was obtained by scoring the response to the question “Which word describes your pain right now?” and this, therefore, ranges from 1 (mild) to 5 (excruciating).
The mean ± standard deviation scores of LF-MPQ (NWC, PRI-T, and PPI), DAS-28, and pain VAS of the participants at baseline and after 48 h are shown in [Table 1]. The results of construct validity and concurrent validity testing are shown in [Table 2]. Correlation analysis indicated a good positive correlation (r = 0.78) between the Hindi version of LF-MPQ and VAS scores. A Pearson r of >0.70 indicated good construct validity.
|Table 2: Pearson correlation between the Hindi version of the long-form McGill Pain Questionnaire, Visual Analog Scale score, and Disease Activity Score-28|
Click here to view
Result of correlation analysis between the Hindi version of LF-MPQ and DAS-28 scores using Pearson correlation coefficient showed a good positive correlation (r value = 0.77) between the two demonstrating good concurrent validity [Table 2].
Results of the test for internal consistency showed that Cronbach's alpha (total as well as individual domains) ranged from 0.74 to 0.80 demonstrating adequate internal consistency [Table 3]. For testing reproducibility, the intraclass correlation coefficient was measured to show how strongly test and retest resemble each other. This type of reliability assumes that there will be no change in the quality or construct (pain scores on Hindi LF-MPQ) being tested in an interval time of 48 h. In our study, the reliability coefficients of the LF-MPQ ranged from 0.76 to 0.84 [Table 3], suggesting substantial reproducibility.
|Table 3: Internal consistency and reproducibility (reliability) of the Hindi version of long-form McGill Pain Questionnaire|
Click here to view
Finally, the most commonly chosen descriptors by our study participants to describe pain were gnawing , hot , heavy and stabbing from sensory domain, tiring from affective domain, troublesome from evaluative domain, and agonizing from miscellaneous domain [Figure 1].
|Figure 1: The most common descriptors chosen by rheumatoid arthritis patients in this study to describe their pain|
Click here to view
| Discussion|| |
In this study, a Hindi version of the LF-MPQ was finalized with 76 descriptors. We found that the Hindi version of LF-MPQ demonstrated adequate reliability and validity in Hindi speaking RA patients. The approach we used to validity testing is the same as other studies that used VAS to assess construct and DAS-28 disease-specific scale to assess concurrent validity., We observed high correlations between the Hindi version of LF-MPQ and VAS and the Hindi version of LF-MPQ and DAS-28, which denotes good construct and concurrent validity, respectively.
In terms of reproducibility, the reliability coefficient of this Hindi questionnaire was 0.81. The 48-h interval was chosen in our study as it was the most appropriate duration to measure reliability, and a longer interval between the tests would bias the results because true changes in pain in patients with musculoskeletal conditions receiving treatment would become more likely. Our reliability estimates for LF-MPQ are comparable to those previously reported for the Brazilian version.
To test the internal consistency of the questionnaire, Cronbach's alpha for each dimension ranged from 0.74 to 0.80, which is considered adequate if it is between 0.70 and 0.95. The translated Brazilian and Japanese version of LF-MPQ has demonstrated ICC values of 0.70–0.79 and 0.59–0.81, respectively., Although our values of ICC are comparable, both these studies have been done in a diverse population comprising spinal pain, extremity pain, and fibromyalgia.,
The cultural background can influence pain perception among people. It is usually observed in clinical practice that patients find difficult to convey pain experience, which is a major barrier in treatment. With respect to the descriptors, majority of our study participants selected the sensory discriminative words referring to the mechanical and spatial properties of pain. The most commonly chosen descriptors for pain among patients in our study from sensory category were “stabbing” from subclass punctuate pressure, “gnawing” from subclass constrictive pressure, “hot” from subclass thermal, and “heavy” from subclass dullness. From the affective and miscellaneous category, our patients chose “tiring” and “agonizing,” respectively. Roche et al. have reported that the commonly chosen descriptors in RA patients were gnawing, heavy, aching, annoying, and exhausting.
Sensory descriptors of gnawing and heavy indicate nociceptive pain linked to inflammation and hot indicates neuropathic mechanism of pain. Our patients chose “tiring” which indicates the presence of affective component of pain. With the present tool, i.e., the Hindi version of LF-MPQ questionnaire, we were able to identify descriptors related to both sensory (inflammatory and neuropathic) and affective mechanisms of pain generation in patients with RA. Clinical guidelines for managing arthritis pain recognize the importance of thoughts, feelings , emotions and they recommend cognitive behaviour therapy to reduce pain, psychological distress, and improve coping.
The limitation of our study was the small sample size, and since we studied only RA patients, the results cannot be generalized to other rheumatic conditions. Nonetheless, in chronic pain syndromes like fibromyalgia, the Hindi version of LF-MPQ can be useful to assess the affective domains of pain and effectiveness of therapy. Early identification of psychosomatic pain factors will be beneficial in adopting multidisciplinary pain approach. This will enhance the quality of life of patients.
| Conclusion|| |
The Hindi version of LF-MPQ can be considered a valid and reliable multidimensional pain assessment tool in Indian patients with RA. The use of this Hindi version of LF-MPQ can assist in comprehensive pain assessment aiding management and improved patient care in India.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Turk DC, Robinson JP. Assessment of patients with chronic pain: A comprehensive approach. In: Turk DC, Melzac R, editors. Handbook of Pain Assessment. 3rd
ed. New York, NY: The Guilford Press; 2011. p. 88-210.
Keefe FJ, Somers TJ. Psychological approaches to understanding and treating arthritis pain. Nat Rev Rheumatol 2010;6:210-16.
Backman CL. Arthritis and pain. Psychosocial aspects in the management of arthritis pain. Arthritis Res Ther 2006;8:1-7.
Katz J, Melzack R. The McGill pain questionnaire: Development, psychometric properties, and usefulness of the long form, short form, and short form-2. In: Turk DC, Melzack R, editors. Handbook of Pain Assessment. 3rd
ed. New York, NY: The Guilford Press; 2011. p. 45-66.
Costa LD, Maher CG, McAuley JH, Costa LO. Systematic review of cross-cultural adaptations of McGill Pain Questionnaire reveals a paucity of clinimetric testing. J Clin Epidemiol 2009;62:934-43.
Linguistic Validation Guidance of a Clinical Outcome Assessment – Mapi Research Trust 2018. Available from: https:// mapi-trust.org/
. [Last accessed on 2018 Jul 16].
Terwee CB, Bot SD, de Boer MR, van der Windt DA, Knol DL, Dekker J, et al
. Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol 2007;60:34-42.
Hawker GA, Mian S, Kendzerska T, French M. Measures of adult pain: Visual analog scale for pain (vas pain), numeric rating scale for pain (nrs pain), mcgill pain questionnaire (mpq), short-form mcgill pain questionnaire (sf-mpq), chronic pain grade scale (cpgs), short form-36 bodily pain scale (sf-36 bps), and measure of intermittent and constant osteoarthritis pain (icoap). Arthritis Res Ther 2011;63:240-52.
Sim J, Arnell P. Measurement validity in physical therapy research. Phys Ther 1993;73:102-10.
Portney LG, Watkins MP. Validity of measurements. In: Portney LG, Watkins MP, editors. Foundations of Clinical Research Applications to Practice. 2nd
ed. New Jersey: Pearson Education Limited; 2014. p. 75-105.
Anderson JK, Zimmerman L, Caplan L, Michaud K. Measures of rheumatoid arthritis disease activity: Patient (PtGA) and provider (PrGA) global assessment of disease activity, disease activity score (DAS) and disease activity score with 28-joint counts (DAS28), simplified disease activity index (SDAI), clinical disease activity index (CDAI), patient activity score (PAS) and patient activity score-II (PASII), routine assessment of patient index data (RAPID), rheumatoid arthritis disease activity index (RADAI) and rheumatoid arthritis disease activity. Arthritis Care Res 2011;63:14-36.
Vaske JJ, Beaman J, Sponarski CC. Rethinking internal consistency in Cronbach's alpha. Leis Sci 2017;39:163-73.
Portney LG, Watkins MP. Reliability. In: Portney LG, Watkins MP, editors. Foundations of Clinical Research Applications to Practice. 2nd
ed. New Jersey: Pearson Education Limited; 2014. p. 61-75.
Costa LD, Maher CG, McAuley JH, Hancock MJ, de Melo Oliveira W, et al
. The Brazilian-Portuguese versions of the McGill Pain Questionnaire were reproducible, valid, and responsive in patients with musculoskeletal pain. J Clin Epidemiol 2011;64:903-12.
Hasegawa M, Hattori S, Ishizaki K, Suzuki S, Goto F. The mcgill pain questionnaire-Japanese version, reconsidered: Confirming the reliability and validity. Pain Res Manag 1996;1:233-37.
Kachooei AR, Ebrahimzadeh MH, Sayyar ER, Salehi M, Salimi E, Razi S. Short Form-McGill Pain Questionnaire-2 (SF-MPQ-2): A cross-cultural adaptation and validation study of the Persian version in patients with knee osteoarthritis. Arch Bone Jt Surg 2015;3:45-50.
Ronald Melzack. The McGill pain questionnaire. In: Melzack R, editor. Pain Measurement and Assessment. New York: Raven Press; 1983.
Roche PA, Klestov AC, Heim HM. Description of stable pain in rheumatoid arthritis: A 6 year study. J Rheumatol 2003;30:1733-38.
Ito S, Kobayashi D, Murasawa A, Narita I, Nakazono K. An analysis of the neuropathic pain components in rheumatoid arthritis patients. Intern Med 2018;57:479-85.
Marques AP, Rhoden L, Siqueira JD, Joao SM. Pain evaluation of patients with fibromyalgia, osteoarthritis, and low back pain. Clinics (Sao Paulo) 2001;56:5-10.
[Table 1], [Table 2], [Table 3]