Mandarin version of the Leeds Dyspepsia Questionnaire: A valid instrument for assessing symptoms in Asians

bs_bs_banner doi: 10.1111/1751-2980.12183 Journal of Digestive Diseases 2014; 15; 591–596 Original article Mandarin version of the Leeds Dyspepsia Questionnaire: A valid instrument for assessing symptoms in Asians Hwong-Ruey LEOW*, Siew-Mooi CHING†, Ramanujam SUJARITA‡, Choon-Fong YAP§, Yook-Chin CHIA‡,§, Shiaw-Hooi HO*, Suresh SITHAMBARAM*, Huck-Joo TAN¶, Khean-Lee GOH* & Sanjiv MAHADEVA* *Division of Gastroenterology, Department of Medicine, ‡Department of Primary Care Medicine, University of Malaya Primary Care Research Group, Faculty of Medicine, University of Malaya, †Department of Family Medicine, Faculty of Medicine and Health Sciences, Universiti Putra Malaysia, Kuala Lumpur, ¶Sunway Medical Centre, Selangor, Malaysia, and §Curtin Health Innovation Research Institute, Faculty of Health Sciences, Curtin University, Perth, Western Australia, Australia OBJECTIVE: To develop and validate a Mandarin all components of the M-LDQ (Cronbach’s α 0.79) version of the Leeds Dyspepsia Questionnaire and test–retest reliability (Spearman’s correlation (M-LDQ) in Asian patients with dyspepsia. coefficient 0.78) were good. The M-LDQ was valid in diagnosing dyspepsia in primary care (area under the METHODS: The M-LDQ was developed according receiver operating characteristics curve 0.84) and was to standardized methods. The validity, internal consis- able to discriminate between secondary and primary tency, test–retest reliability and responsiveness of care patients (median cumulative LDQ score 13.0 vs the instrument were evaluated in both primary and 3.0, P < 0.0001). Among eight patients with organic secondary care patients. dyspepsia, the median M-LDQ score reduced signifi- cantly from 21.0 (pretreatment) to 9.5 (4 weeks post- RESULTS: A total of 184 patients (mean age treatment) (P < 0.0001). 54.0 ± 15.8 years, of whom 59% were women and 72.3% of whom had at least secondary level educa- CONCLUSION: The M-LDQ is a valid and respon- tion) were recruited between August 2012 and March sive instrument for assessing ethnic Chinese adults 2013, from both primary (n = 100) and secondary with dyspepsia. care clinics (n = 84). Both the internal consistency of KEY WORDS: ethnic Chinese, functional dyspepsia, Mandarin, outcome measure, questionnaire, validation. Correspondence to: Sanjiv MAHADEVA, Division of Gastroenterology, INTRODUCTION Department of Medicine, Faculty of Medicine, University Malaya, Lembah Pantai, Kuala Lumpur 50603, Malaysia, Email: Dyspepsia, a collection of symptoms referring to the [email protected] upper gastrointestinal (GI) tract, is a global condi- tion.1 In South-East Asia, population-based studies Conflict of interest: None. have reported its prevalence ranging from 12.2% to 24.3%,2,3 which is similar to those reported in the Far © 2014 Chinese Medical Association Shanghai Branch, Chinese East.4–6 Functional dyspepsia (FD) is the commonest Society of Gastroenterology, Renji Hospital Affiliated to Shanghai cause of dyspepsia in both Western7 and Eastern Jiaotong University School of Medicine and Wiley Publishing Asia countries8,9. Unfortunately, treatment remains unsat- Pty Ltd isfactory in FD, resulting in an impact on patients’ 591

592 H-R Leow et al. Journal of Digestive Diseases 2014; 15; 591–596 health-related quality of life (QoL)3 and an economic process were resolved through discussion. Any prob- burden for healthcare services due to frequent medical lems in the forward translation were documented and consultation for those with persistent symptoms.10,11 an independent backward translation (Mandarin to Malaysian English) was then produced from the An effective treatment for FD is badly required, but consensus forward translation as a quality check. A clinical trials assessing the efficacy of new treatments consensus Mandarin version was derived after the need an appropriate outcome measure. As FD is approval by the original instrument developer. Cogni- mainly a symptom-driven condition, questionnaires tive interviews were conducted with 10 participants with established psychometrics are relied upon by fluent in Mandarin, but of varied age and educational researchers.12 While many such instruments exist in backgrounds. If any problems with items in the the English language, there is a paucity of validated M-LDQ were identified during the cognitive debrief- dyspepsia questionnaires in Asian languages. The ing, further changes were made and the initial steps Leeds Dyspepsia Questionnaire (LDQ) is an eight- repeated until a conceptually and semantically accept- item symptom-based questionnaire assessing the able Mandarin translation of the LDQ was developed. severity of dyspepsia according to the frequency and severity of various upper GI symptoms.13 Although it Participants was developed for the utility in native English speak- ers, the LDQ has been successfully translated and The inclusion criteria for participants were adults aged recently proven to be valid for assessing dyspepsia ≥18 years, of Chinese ethnicity and fluency in Manda- symptoms in several Southeast Asian languages.14 rin. The setting for the study included the primary care clinics of the University of Malaya and the Endoscopy Malaysia has a unique multi-ethnic Asian population, Units of both institutions (as a representative of sec- of which almost 30% are ethnic Chinese.15 Despite ondary care). Consecutive patients attending the being descendants of migrants from southern China primary care clinics (for any indications) for an index over several generations, the ethnic Chinese in Malay- upper GI endoscopy were invited to participate in the sia have preserved both the traditions and written lan- study. Patients were excluded if they were unable to guage (i.e. Mandarin), of natives from southern provide written informed consent or were not fluent China. As a prelude to several pharmaceutical trials in Mandarin. among ethnic Chinese patients with FD we performed this study to translate the LDQ into Mandarin and Evaluation of M-LDQ validate it among adult ethnic Chinese patients with dyspepsia. The psychometric properties of M-LDQ were evalu- ated by assessing its internal consistency, reliability, MATERIALS AND METHODS validity and responsiveness. All consenting partici- pants were interviewed by native Mandarin speakers The study was conducted at two institutions with a with the M-LDQ in both the primary care and the large catchment of ethnic Chinese patients, the Uni- Endoscopy Units, prior to being assessed by a physi- versity Malaya Medical Centre and Sunway Medical cian. The latter, who was blinded to the results of the Centre. Approval was obtained from the Institutional M-LDQ, would then make a clinical assessment as to Review Board at both institutions prior to the com- whether the patients interviewed had dyspepsia or not mencement of this study. and recorded this information separately. Concurrent validity, that is, assessing an instrument against the Translation of the LDQ into Mandarin gold standard,16 was determined by comparing the M-LDQ summary score with the presence or absence A Mandarin version of the LDQ (M-LDQ) was of dyspepsia as diagnosed by the primary care physi- developed using the standard forward–backwards cian only, using the receiver operating curve (ROC) translation technique. An independent forward trans- analysis. As patients in secondary care were expected lation (source Malaysian English version14 to target to have more severe dyspepsia than those in primary Mandarin version) was first produced with the aim of care, the discriminant validity of the M-LDQ was achieving equivalence in concepts (i.e. conceptual assessed by comparing the mean total scores of the equivalence) and meaning (i.e. semantic equivalence), LDQ between primary and secondary care patients. from which a consensus forward Mandarin translation Physicians in both primary and secondary care used a was obtained. Differences which had arisen from this similardefinition of dyspepsia, that is, predominant © 2014 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and Wiley Publishing Asia Pty Ltd

Journal of Digestive Diseases 2014; 15; 591–596 Mandarin LDQ in Asians 593 epigastric discomfort lasting for more than one Patients’ characteristics and LDQ item response month, together with any other associated symptoms referable to the upper GI tract (e.g., nausea, post- From August 2012 to March 2013, a total of 184 (100 prandial bloating and belching). This definition is in from the primary care and 84 from the endoscopy line with the recently published Asian Consensus unit) patients were recruited in the study. Table 1 high- report on FD.17 lights the basic characteristics of the study partici- pants. In particular, only 28.4% of the participants Internal consistency evaluates the correlation across had had tertiary education. The prevalence of dyspep- items in different domains and across all items of a sia, by physicians’ diagnosis, was 24.0% in primary questionnaire. Cronbach’s α was used to assess inter- care and 67.5% in secondary care. The response rate nal consistency, with a value of 0.7 being taken as for all items in both versions of the LDQ was 100%, adequate for group comparisons.16 Test–retest reliabil- and Table 2 highlights the response frequencies of ity, an assessment of error caused by repeated mea- each category for patients from both primary and sec- surement under a similar condition, was performed by ondary care. In general, item responses were skewed, a repeat interview by telephone 3 days after the initial reflecting a lower prevalence of dyspepsia in primary interview. The summed M-LDQ scores at day 0 and care patients than in secondary care. Individual items day 3 were then compared using Spearman’s rank cor- of the LDQ had a response rate of > 5% in most cat- relation coefficient. egories, apart from the vomiting and nausea items, perhaps indicating less severe dyspepsia symptoms in Responsiveness, that is, the ability of an instrument to primary care. detect changes in patients over a period of time,16 was assessed in secondary care. All dyspeptic patients with Reliability either reflux esophagitis or peptic ulcer disease, known to have received proven beneficial therapy in the form Internal consistency, assessed by Cronbach’s α, for of proton pump inhibitors and Helicobacter pylori combined primary and secondary care patients was eradication, were asked to return for a repeat interview 0.79, indicating a sufficiently high level of internal with the M-LDQ 4 weeks after therapy. A comparison consistency. All 184 (100%) patients, that is, 100%, of of summed total scores of the M-LDQ before and after patients, participated in the follow-up telephone inter- treatment in these selected patients was performed to view after 3 days. Spearman’s correlation coefficient assess its responsiveness to change. between the first and second summed total scores of the M-LDQ was 0.78, representing a good degree of Statistical analysis reliability on retesting the questionnaire. Statistical analyses were performed using SPSS 16.0 Validity (SPSS Inc., Chicago, IL, USA). Mann–Whitney U–test, Wilcoxon rank matched pair test, Fisher’s exact test Concurrent validity was assessed in the primary care and Spearman’s rank correlation were used where setting only. The area under the ROC curve (AUROC) appropriate. A two-sided P ≤ 0.05 was considered to indicate statistical significance. Table 1. Characteristics of patients who completed the Mandarin version of the Leeds dyspepsia questionnaire RESULTS Characteristics Patients (N = 184) Translation and development of the M-LDQ Age (mean ± SD) 54.0 ± 15.8 M-LDQ was developed according to the standard pro- Gender tocol detailed above. The final version of the M-LDQ 41:59 underwent cognitive debriefing in 10 native Mandarin Male : female (%) speakers from the lay public, three of whom were Education level#, n (%) 2 (1.1) males. Their median age was 31.5 years (range 25–42 48 (26.2) years) and 7 of the 10 individuals had tertiary educa- None 81 (44.3) tion. No problems were encountered with the original Primary 52 (28.4) translation. Secondary Tertiary 100 (54.3) Location, n (%) 84 (45.7) Primary care Secondary care #Data is available in 183 participants. SD, standard deviation. © 2014 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and Wiley Publishing Asia Pty Ltd

594 H-R Leow et al. Journal of Digestive Diseases 2014; 15; 591–596 Table 2. Endorsement frequencies of each response category of the Mandarin version of the Leeds dyspepsia questionnaire Response category (%) Symptoms: None Less than Between monthly More than At least No frequency of monthly and weekly weekly daily response Indigestion 55.9 8.2 9.8 12.5 13.6 0 Heartburn 80.5 5.4 6.5 6.0 1.6 0 Regurgitation 73.9 7.1 9.2 6.0 3.8 0 Burping 60.8 7.6 6.0 9.8 15.8 0 Nausea 79.3 4.3 8.2 4.9 3.3 0 Vomiting 90.2 4.9 3.3 1.1 0.5 0 Bloating 63.7 5.4 9.2 14.1 7.6 0 40 40 LDQ cumulative score 30 30 20 20 10 10 0 Primary care Endoscopic unit 0 Pretreatment Study setting Posttreatment Figure 1. Box plot showing the median values of the Figure 2. Box plot showing the median values of the summed total score of the Mandarin version of the Leeds summed total score of the Mandarin version of the Leeds Dyspepsia Questionnaire (LDQ) in primary and secondary Dyspepsia Questionnaire (LDQ) in secondary care patients care patients (P < 0.0001). before and after 30-day treatment with proven efficacy. for the summed total score of the LDQ plotted against Responsiveness physician diagnosis of dyspepsia was 0.84 (95% CI 0.75–0.94). The point along the ROC curve that Of 84 patients, 12 were diagnosed with either reflux provided the closest agreement between sensitivity esophagitis or peptic ulcer disease in secondary care. and specificity was chosen as the cut-off value for diag- Eight of these 12 (75.0%) patients returned after 30 nosing the presence of dyspepsia. A score of 6.5 on the days of treatment with proven efficacy for a follow-up M-LDQ yielded a sensitivity of 75.0% and a specificity interview using the LDQ. The median LDQ score of 82.9% for diagnosing dyspepsia in the community. among the patients who had used the M-LDQ (n = 8), Discriminant validity was determined by comparing reduced from 21.0 (IQR 12.0–24.0) prior to treatment the summed total score between patients in primary to 9.5 (IQR 2.0–13.0) after treatment (P < 0.0001, and secondary care. The median LDQ score in primary Wilcoxon rank test; Figure 2). care was 3.0 (interquartile range [IQR] 1.0–8.8), while that in secondary care was 13.0 (IQR 5.0–17.0) DISCUSSION (P < 0.0001, Mann–Whitney U-test) (Fig. 1). These differences in median scores demonstrated that the Dyspepsia is as prevalent in Asia as in Western coun- M-LDQ was able to discriminate between two popu- tries.17 However, there are recognized differences in lations with different prevalences and severity of clinical features and etiology of dyspepsia between dyspepsia. Western and Asian patients.18 As the management of © 2014 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and Wiley Publishing Asia Pty Ltd

Journal of Digestive Diseases 2014; 15; 591–596 Mandarin LDQ in Asians 595 dyspepsia, particularly FD, is mainly symptom-driven, disease (GERD), may appear to conflict with the Rome it is important that questionnaires for dyspepsia with process.27 However, the Rome definition of dyspepsia, adequate psychometric properties are developed in and its specification that any symptoms of GERD be local Asian languages among native speakers. Instead excluded, has been criticised for excluding many of developing an instrument from the beginning, we patients with genuine FD in both Asian28 and non- decided to validate a translated version of the LDQ, an Asian communities.29 It has even less applicability for internationally recognized dyspepsia questionnaire,12 uninvestigated dyspepsia in the community, where to enable cross-cultural comparisons between Asian both reflux symptoms and epigastric pain commonly and Western patients with dyspepsia in the future. coexist. We conclude that M-LDQ is a valid, reliable and responsive instrument for assessing dyspepsia in ethnic In this study the M-LDQ was validated in both Chinese patients who communicate in Mandarin. primary and secondary care patients, as most patients However, further validation of the M-LDQ in non- consult at these levels of healthcare services.19 As the Malaysian Chinese may be necessary due to dialect severity of patients’ symptoms tends to differ between variations in Mandarin in other Asian regions. We primary and secondary, or even tertiary care, a dyspep- envisage that the utilisation of the M-LDQ in clinical sia questionnaire assessing symptoms should ideally trials of ethnic Chinese patients may subsequently be validated in all settings.13 A previous study in Hong enable direct comparisons of treatment efficacy with Kong SAR, China did develop and validate a ‘Chinese- non-Chinese patients with FD. language’ dyspepsia questionnaire,20 but this was con- ducted in secondary-care patients only and it was not ACKNOWLEDGMENTS clear which type of Chinese language was used. We would like to thank Professor Paul MOAYYEDI The assessment of concurrent validity for the M-LDQ from the McMaster’s University (Hamilton, Canada) was based on the physician’s diagnosis as a ‘gold stan- for granting us permission to translate and validate the dard’ for dyspepsia. Various studies have used different Leeds Dyspepsia Questionnaire locally. standards for the evaluation of symptom question- naires, as a ‘gold standard’ diagnosis of dyspepsia does We would like to thank Mme Khin-Choo GO for her not exist.21,22 While many other studies have used a invaluable assistance in developing the Mandarin similar method to ours,13,23,24 self-assessment diaries,25 translation of the LDQ. This study was funded by an dyspepsia adverse events26 and generic quality of life educational grant from the Malaysian Society of Gas- scores20 have all been used to demonstrate concurrent troenterology and Hepatology. validity. We believe that for pragmatic reasons the use of a primary care physician’s diagnosis in this study REFERENCES was more appropriate for determining the validity of the M-LDQ. 1 Mahadeva S, Goh KL. Epidemiology of functional dyspepsia: a global perspective. World J Gastroenterol 2006; The original LDQ was designed to be administered in 12: 2661–6. a face-to-face interview.13 Face-to-face or telephone- based questionnaires may be potentially inferior to 2 Yang SY, Lee OY, Bak YT et al. Prevalence of self-administered questionnaires for several reasons. gastroesophageal reflux disease symptoms and An interview-based questionnaire is generally more uninvestigated dyspepsia in Korea: a population-based time consuming and may be prone to bias introduced study. Dig Dis Sci 2008; 53: 188–93. by misinterpretation with less well-trained interview- ers. However, in many Asian societies, where the edu- 3 Mahadeva S, Yadav H, Rampal S, Everett SM, Goh KL. cation levels are somewhat lower than in developed Ethnic variation, epidemiological factors and quality of life countries, an interview-based questionnaire has the impairment associated with dyspepsia in urban Malaysia. advantage of clarifying questions or terminology that Aliment Pharmacol Ther 2010; 31: 1141–51. may be less well understood. Furthermore, in the setting of an intervention trial, patients or participants 4 Hu WH, Wong WM, Lam CL et al. Anxiety but not are generally interviewed as part of the study protocol. depression determines health care-seeking behaviour in The extra time spent in using an interview-based ques- Chinese patients with dyspepsia and irritable bowel tionnaire is a minor consideration in this setting. syndrome: a population-based study. Aliment Pharmacol Ther 2002; 16: 2081–8. The inclusive definition of dyspepsia used in the LDQ, which includes symptoms of gastroesophageal reflux 5 Li Y, Nie Y, Sha W, Su H. The link between psychosocial factors and functional dyspepsia: an epidemiological study. Chin Med J (Engl) 2002; 115: 1082–4. 6 Lu CL, Lang HC, Chang FY et al. Prevalence and health/social impacts of functional dyspepsia in Taiwan: a study based on the Rome criteria questionnaire survey assisted by endoscopic exclusion among a physical check-up population. Scand J Gastroenterol 2005; 40: 402–11. © 2014 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and Wiley Publishing Asia Pty Ltd

596 H-R Leow et al. Journal of Digestive Diseases 2014; 15; 591–596 7 Ford AC, Marwaha A, Lim A, Moayyedi P. What is the 19 Mahadeva S, Yadav H, Everett SM, Goh KL. Factors prevalence of clinically significant endoscopic findings in influencing dyspepsia-related consultation: differences subjects with dyspepsia? Systematic review and between a rural and an urban population. Neurogastroenterol meta-analysis. Clin Gastroenterol Hepatol 2010; 8: 830–7.e2. Motil 2011; 23: 846–53. 8 Mahadeva S, Goh KL. Clinically significant endoscopic 20 Hu WH, Lam KF, Wong YH et al. The Hong Kong index of findings in a multi-ethnic population with uninvestigated dyspepsia: a validated symptom severity questionnaire for dyspepsia. Dig Dis Sci 2012; 57: 3205–12. patients with dyspepsia. J Gastroenterol Hepatol 2002; 17: 545–51. 9 Zhao Y, Zou D, Wang R et al. Dyspepsia and irritable bowel syndrome in China: a population-based endoscopy study of 21 Dent J. Definitions of reflux disease and its separation from prevalence and impact. Aliment Pharmacol Ther 2010; 32: dyspepsia. Gut 2002; 50 Suppl 4: iv17–20; discussion 562–72. iv21–2. 10 Mahadeva S, Yadav H, Everett SM, Goh KL. Economic 22 Fraser A, Delaney BC, Ford AC, Qume M, Moayyedi P. The impact of dyspepsia in rural and urban Malaysia: a short-form Leeds Dyspepsia Questionnaire validation study. population-based study. J Neurogastroenterol Motil 2012; 18: Aliment Pharmacol Ther 2007; 25: 477–86. 43–57. 23 Buckley MJ, Scanlon C, McGurgan P, O’Morain CA. A 11 Mahadeva S, Goh KL. Letter: East-West differences in the validated dyspepsia symptom score. Ital J Gastroenterol economic impact of functional dyspepsia. Aliment Pharmacol Hepatol 1997; 29: 495–500. Ther 2013; 38: 655. 24 Junghard O, Lauritsen K, Talley NJ, Wiklund IK. Validation 12 Fraser A, Delaney B, Moayyedi P. Symptom-based outcome of seven graded diary cards for severity of dyspeptic measures for dyspepsia and GERD trials: a systematic symptoms in patients with non ulcer dyspepsia. Eur J Surg review. Am J Gastroenterol 2005; 100: 442–52. Suppl 1998; 583: 106–11. 13 Moayyedi P, Duffett S, Braunholtz D et al. The Leeds 25 Leidy NK, Farup C, Rentz AM, Ganoczy D, Koch KL. Dyspepsia Questionnaire: a valid tool for measuring the Patient-based assessment in dyspepsia: development and presence and severity of dyspepsia. Aliment Pharmacol Ther validation of Dyspepsia Symptom Severity Index (DSSI). Dig 1998; 12: 1257–62. Dis Sci 2000; 45: 1172–9. 14 Mahadeva S, Chan WK, Mohazmi M, Sujarita R, Goh KL. 26 Rabeneck L, Wristers K, Goldstein JL, Eisen G, Dedhiya SD, Validation study of the Leeds Dyspepsia Questionnaire in a Burke TA. Reliability, validity, and responsiveness of severity multi-ethnic Asian population. J Gastroenterol Hepatol 2011; of dyspepsia assessment (SODA) in a randomized clinical 26: 1669–76. trial of a COX-2-specific inhibitor and traditional NSAID therapy. Am J Gastroenterol 2002; 97: 32–9. 15 Department of Statistics Malaysia. Population distribution and basic demographic characteristics. The 2000 Population 27 Drossman DA. The functional gastrointestinal disorders and Housing Census of Malaysia. Putrajaya: Department of and the Rome III process. Gastroenterology 2006; 130: Statistics 2000. 1377–90. 16 Fayers PM, Machin D. Quality of Life: Assessment, Analysis 28 Manabe N, Haruma K, Hata J et al. Clinical characteristics of and Interpretation. Chichester: John Wiley & Sons, 2000. Japanese dyspeptic patients: is the Rome III classification applicable? Scand J Gastroenterol 2010; 45: 567–72. 17 Miwa H, Ghoshal UC, Fock KM et al. Asian consensus report on functional dyspepsia. J Gastroenterol Hepatol 2012; 29 van Kerkhoven LA, Laheij RJ, Meineche-Schmidt V, 27: 626–41. Veldhuyzen-van Zanten SJ, de Wit NJ, Jansen JB. Functional dyspepsia: not all roads seem to lead to Rome. 18 Mahadeva S, Raman MC, Ford AC et al. Gastro-oesophageal J Clin Gastroenterol 2009; 43: 118–22. reflux is more prevalent in Western dyspeptics: a prospective comparison of British and South-East Asian patients with dyspepsia. Aliment Pharmacol Ther 2005; 21: 1483–90. © 2014 Chinese Medical Association Shanghai Branch, Chinese Society of Gastroenterology, Renji Hospital Affiliated to Shanghai Jiaotong University School of Medicine and Wiley Publishing Asia Pty Ltd

Copyright of Journal of Digestive Diseases is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.