Validity and Reliability of the Eating Assessment Tool (EAT-10)

Transcription

1 Annals of Otology. Rhinology A Laryngology ll7(l):9l Annals Publishing Company. All rights reserved. Validity and Reliability of the Eating Assessment ool (EA-) Peter C. Belafsky, MD, PhD; Debbie A. Mouadeb, MD; Catherine J. Rees, MD; Jan C. Pryor, MA; Gregory N. Postma, MD; Jacqueline Allen, MBChB, FRACS; Rebecca J. Leonard, PhD Objectives: he Eating Assessment ool is a self-administered, symptom-specific outcome instrument for dysphagia. he purpose of this study was to assess the validity and reliability of the lo-item Eating Assessment ool (EA-). Methods: he investigation consisted of phases: ) line-item generation, ) line-item reduction and reliability, ) normative data generation, and ) validity analysis. All data were collected prospectively. Internal consistency was assessed with the Cronbach alpha. est-retest reliability was evaluated with the Pearson product moment correlation coefficient. Normative data were obtained by administering the instrument to a community cohort of healthy volunteers. Validity was assessed by administering the instrument before and after dysphagia treatment and by evaluating survey differences between normal persons and those with known diagnoses. Results: A total of 69 surveys were administered to 8 patients. he internal consistency {Cronbach alpha) of the final instrument was.96. he test-retest intra-item correlation coefficients ranged from.7 to.9. he mean {±SD) EA- score of the normal cohort was. ±.. he mean EA-IO score was.58 ±.8 for patients with esophageal dysphagia,. ±. for those with oropharyngeal dysphagia, 9.9 ±.6 for those with voice disorders,. ±.6 for those with head and neck cancer, and.7 ±9.6 for those with reflux. he patients with oropharyngeal and esophageal dysphagia and a history of head and neck cancer had a significantly higher EA- score than did those with reflux or voice disorders {p <.). he mean EA-IOscoreof the patients with dysphagia improved from 9.87 ±.5 to 5. ± 7. after treatment (p <.). Conclusions: he EA- has displayed excellent internal consistency, test-retest reproducibility, and criterion-based validity. he normative data suggest that an EA- score of or higher is abnormal. he instrument may be utilized to document the initial dysphagia severity and monitor the treatment response in persons with a wide array of swallowing disorders. Key Words: data collection, dysphagia, outcome assessment, quality of life, questionnaire, survey, swallowing, SWAL- QOL. INRODUCION Dysphagia, or difficulty swallowing, is associated with significant morbidity and mortality. Swallowing disorders have profound social, emotional, and socioeconomic implications.'""^ Persons with dysphagia are more likely to be anxious and depressed.'* More than % of persons with dysphagia experience panic or anxiety during meals, causing them to eat in isolation.^ he disorder is underrecognized by most clinicians and is frequently underreported by patients."^ In a survey of 97 adults in family medicine office waiting rooms in Georgia,.6% reported problems swallowing that occurred at least several times per month, and only 6.% of the respondents with dysphagia had discussed the problem with their doctor.** In stroke patients, the prevalence of dysphagia exceeds 5%, and it may be as high as 8% in patients with Parkinson's disease.^-'^ Other causes of dysphagia include head and neck cancer, neuromuscular diseases, gastroesophageal reflux disease, and primary esophageal abnormalities {including motility disorders and Zenker's diverticulum [ZD]). he majority of dysphagia research has focused on disease pathophysiology, diagnosis, and therapy.^' A paucity of the current literature is dedicated to outcome assessment and quality of life in those persons afflicted with swallowing problems. here is need for a patient-centered outcome measure that incorporates not only the physiologic implications From the Center for Voice and Swallowing, Department of Otolaryngology-Head and Neck Surgery, University of California-Davis, Sacramento, California (Belafsky. Mouadeb, Rees. Pryor. Allen, Leonard), and the Center for Voice and Swallowing Disorders, Department of Otolaryngology-Head and Neck Surgery, Medical College of Georgia, Augusta, Georgia (Postma). Presented at the meeting of the Atnerican Broncho-Esophagological Association, Orlando, Florida, May -, 8. Correspondence: Peter C. Belafsky. MD, PhD, Center for Voice and Swallowing, Dept of Otolaryngology, University of Califomia- Davis. 5 Stockton Blvd. Suite 7, Sacramento. CA

2 9 Belafsky et ai Eating Assessment ool of dysphagia, but also the emotional, social, and psychological consequences. he SWAL-QOL is a symptom-specific outcome instrument that was developed to assess the severity of oropharyngeal dysphagia."''^ he instrument consists of items and can be cumbersome to complete. Because of this limitation, the instrument has not been widely accepted into clinical practice. In a survey of speech pathologists in California who treat dysphagia, none of the respondents indicated that they used the SWAL-QOL regularly (Belafsky. unpublished data). Reasons cited included the length of time required for the patient to complete the survey and for the clinician to score the survey. Other attempts to develop quality-of-iife measures and/or symptom surveys have either focused on a certain subset of dysphagic patients (such as the M. D. Anderson Dysphagia Inventory IMDADI] for head and neck cancer patients) or are too cumbersome for clinicians to readily score and utilize expeditiously in the clinic.^-^ here remains a clinical need for a rapidly administered and easily scored dysphagia instrument that can be administered on each patient visit in order to assess symptom severity, quality of life, and treatment efficacy. he sur\'ey should be applicable to a broad range of dysphagic patients, including those with both oropharyngeal and esophageal phase dysphagia. he purpose of this study was to assess the validity and reliability of a comprehensive dysphagia outcome instrument designed to meet these clinical needs: the Eating Assessment ool (EA-). MAERIALS AND MEHODS Approval to conduct this investigation was granted by the Institutional Review Board at the University of California-Davis. All data were collected prospectively and coded and recorded into SPSS. for Macintosh (SPSS Inc. Chicago. Illinois). he study consisted of separate phases. Phase J: Line Item Generation. A multidisciplinary group of dysphagia experts was assembled to construct the original survey instrument. he group consisted of gastroenterologists. otolaryngologists, speech-language pathologists, and nutritionists. Each clinician was asked to review the dysphagia literature, draw from his or her own clinical experience, examine other dysphagia questionnaires, and contribute questions he or she deemed to have excellent face validity to the original survey. All items were arranged in a 5-point Likert scale. he group of clinicians then met in committee to review all of the generated questions. Redundant line items were removed. A list of 5 questions was then shortened by open ballot voting to produce the original -item Eating Assessment ool (EA-). Phase : Line Item Reduction and Reliahility Analysis. o assess single line item test-retest reproducibility. we administered the EA- on separate occasions to normal controls and to persons with previously diagnosed voice or swallowing disorders. One hundred healthy persons with no past medical history of voice, swallowing, reilux, airway, neurologic, rheumatologic. hématologie, or neoplastic disorders were administered the survey on separate occasions separated by at least hours. he survey was also given to a large cohon of persons with voice and swallowing disorders (n = 5) on separate occasions separated by at least hours. Internal consistency was assessed with the Cronbach alpha. est-retest reliability was evaluated with the Pearson product moment correlation coefficient. Inter-item correlations were utilized to evaluate line item redundancy. he most redundant and poorly reliable items were removed from the survey instrument, and the lo-item Eating Assessment ool (EA-) was derived. Phase : Normative Data Generation. Normative data were obtained by giving the instrument to a community cohort of healthy volunteers with no medical history of voice, swallowing, reflux, airway, neurologic, rheumatologic, hématologie, or neoplastic disorders. he mean total symptom survey score plus SD of this normal cohort was considered the upper limit of normal for the EA-. Phase : Validity Analysis. he EA- was administered to a cohort of 6 individuals before and after treatment. Criterion-related validity was assessed by comparing pretreatment and posttreatmenl EA- survey results with the paired-samples /-test. EA- scores were also compared between diseased and normal populations with the independent samples /-test. A Bonferroni correction (ct/n) was used to adjust for the evaluation of multiple comparisons. RESULS Phase J: Line tern Generation. he original EA- devised by the multidisciplinury panel is displayed in able. AH items were deemed to have excellent face validity on committee open ballot voting. he internal consistency of the original EA- was excellent (Cronbach alpha =.97). Phase : Line item Reduction and Reliability Analysis. he single intra-item test-retest correlations ranged from.8 to.9 for all questions on the EA- (able ). he inter-item correlations

3 Belafsky ei al. Fating Assessment oiil 9 ABLE I. EAING ASSESSMEN OOL (EA-) Circle the appropriate response. o what extent are the following scenarios problematic for you?. My swallowing problem has caused me io lose weight.. My swallowing problem interferes with my ability to go out for meals.. My swallowing problem interferes with my work or oihcr activities.. Swallowing liquids takes extra effort. 5. Swallowing solids takes extra effort. 6. Swallowing pills lakes extru effort. 7. I have altered my diet because of my swallowing problem. 8. Swallowing is painful. 9. ^c pleasure of eating is affected by my su allowing.. When swallow f(x>d sticks in my tlintat.. When I swallow food sticks in mychesl.. I L'oujzh when I cat.. I am afraid to eat because of my swallowing problem.. My swallowing problem is a burden to my family. 5. I get tired when eat. 6. I avoid eating in front of people. 7. I am afraid of choking in my sleep. 8. I become short of breath when I eat. 9. People perceive me as sick because of niv SWLIHOU ing problem.. Swallowing is stressful. = : No problem. I ~\ -} -> -> (ïtal EA- - Severe problem ranged from.6 to.87. he least reliable and cients of the EA- indicated that the instrument is most redundant items were removed from the instru- highly reproducible; they ranged from.7 to.9. ment to devise the EA- (able ). he line item reduction improved the internal consistency (Cron- Phase : Normative Data Generation. he mean bach alpha) of the survey instrument from.97 to (±SD) age of the normal population (n= ) was he test-retest Intra-item correlation coeffi- ± 6 years. Fifty-three percent of the subjects were ABLE. INDIVIDUALES-REES LINE IEM MEASURES FOR EA- EA- Survey Question Correlation Coefficient. My swallowing problem has caused me to lose weight..88. My swallowing problem interferes wiih my ability io go oui for meals.,9. My swallowing problem interferes wiih my work or other activities Swallowing liquids takes extra effort Swallowing solids takes extra effort Swallowing pills takes extra effort I have altered my diet because of my swallowing problem Swallowing is painful he pleasure (if eating is affected hy my swallowing,.86. When swallow food sticks in mv throat When I swallow food sticks in my chest..8. I cough when I eat..7. I am afraid to eat because of m\ sw;ilu>\\ing problem My swallowing problem is a burden to my family.,69 5. I get tired when I eat avoid eating in front of people I am afraid of choking in my sleep I become short of breath when I eat People perceive me as sick becuuse of my swallowing problem..65. Swallowing is stressful. l>.799 otal EA- Ü.958

4 9 Belafsky et al. Eating Assessment ool ABLE. EAING ASSESSMEN OOL (EA-) Circle the appropriate response. o what extent are the following scenarios problematic for you?. My swallowing problem has caused me to lose weight.. My swallowing problem interferes with my ability to go out for meals.. Swallowing liquids takes extra effort.. Swallowing solids takes extra effort. 5. Swallowing pills takes extra effort. 6. Swallowing is painful. 7. he pleasure of eating is affected by my swallowing. 8. When I swallow food sticks in my throat. 9. I cough when I eat.. Swallowing is stressful. = No problem I = Severe problem otal EA- male. he mean EA- score for the normal cohort was. ±.. he mean plus SD yielded an upper limit of normal for the EA- score of.. he normative data suggest that an EA- score of or higher (mean + SD) is abnormal. Phase : Validity Analysis. he instrument was administered to a arge cohort of 5 individuals with voice and swallowing disorders. he mean age of this cohort was 6 ± years. Fifty-four percent were male. Sixty-six patients (8%) had reflux disease, 5 (%) had a voice disorder, 5 (%) had a known source of oropharyngeal dysphagia, (8%) had a history of head and neck cancer, and 6 (%) had a known cause of esophageal phase dysphagia. he oropharyngea! dysphagia group consisted largely of persons with a history of stroke or progressive neurologic disease such as Parkinson's disease, amyotrophic lateral sclerosis, or pseudobulbar palsy. he group with esophageal phase dysphagia included those individuals with esophageal motility disorders, neoplasia, webs, strictures, or rings. he mean EA- score was.7 ±9.6 for those with reflux disease, 9.9 ±.6 for those with a voice disorder,. ±. for those with oropharyngeal dysphagia,. ±.6 for those with a history of head and neck cancer, and.58 ±.8 for those with esophageal phase dysphagia (Fig ). he EA- Fig. Mean Eating Assessment ool (EA-) scores stratified by diagnostic category. score was similar in individuals with oropharyngeal and esophageal dysphagia and in those with a history of head and neck cancer (p >.5). Persons with oropharyngeal and esophageal dysphagia and a history of head and neck cancer had signiflcantly higher EA- scores than those with reflux or voice disorders (p <. ). here was no difference in the EA- score between those with a voice disorder and those with reflux disease (p >.5). he EA- scores of all of these diagnostic categories were significantly higher than the normative data in the healthy community cohort (p <.). he EA- was given to 6 persons who were undergoing treatment for dysphagia. he mean age of this cohort was 65 ± 6 years. Fifty-flve percent were female. wenty-three patients (5%) had a diagnosis of ZD, (%) had a diagnosis of ph- or endoscopy-proven reflux disease, and 9 (%) had a diagnosis of esophageal stricture. All persons witb ZD underwent endoscopie diverticulotomy. Persons with reflux disease underwent treatment with once- or twice-daily proton pump inhibitors, and those with an esophageal stricture underwent esophageal dilation with a balloon or bougie. he mean pretreatment EA- score was 9.87 ± io.5. his improved signiflcantly, to 5. ± 7., after treatment (p <.). All variables on the EA- showed a highly significant improvement with intervention, indicating excellent criterion-based validity (Fig ). Patients with reflux disease had a mean improvement in EA- score of 9.6 ±., those with ZD had a mean improvement of 6. ± 7., and those with an esophageal stricture had a mean improvement of 8.6 ± 6. (p <. ). AU of these improvements remained signiflcant after adjusting for the number of tests performed (adjusted Bonferroni a =.5). Patients treated for a stricture and ZD displayed greater improvement on the EA- than did those treated for reflux disease (p <.5). his difference was not signiflcant with the adjusted a.

5 Belafsky et al, Eating Assessment ool Before Ireatmenl After treatment L n - Reflux Stricture Zenker s Fig. Mean reduction in EA- scores stratified by clinical diagnosis. DISCUSSION Dysphagia is simply difficulty swallowing. It is a symptom, not a disease. Patients with the symptom of dysphagia can have a vast array of clinical diagnoses, ranging from mild acid reflux to terminal esophageal cancer. Because dysphagia is a symptom, it is essential that the clinician be able to document the severity of a patient's self-perception of the disability caused by the swallowing problem. he EA- is a self-administered survey instrument for the subjective assessment of dysphagia. Normative data suggest that an EA- score of or greater is abnormal. he instrument has displayed excellent internal consistency, test-retest reproducibility, and criterion-based validity. he EA- has been utilized in our center to document initial dysphagia severity and to monitor response to treatment in persons with a wide range of swallowing disorders. Other dysphagia inventories were developed and validated before the EA-. None of these instruments are in widespread clinical use. he SWAL- QOL was developed to assess quality of life and outcomes in patients with oropharyngeal dysphagia.""'-^ he final version of the SWAL-QOL (and its partner instrument, the 5-item SWAL-CARE, which was designed to assess patient-perceived quality of care) was developed from an initial 85-item survey administered by mail to 6 dysphagia patients." his group was primarily male (76%), with a mean age of 67.9 years. he initial survey took an average of 56 minutes to complete. he survey length was reduced by 5% and then administered by mail to 86 patients with known oropharyngeal dysphagia, primarily from a US Veterans Administration hospital population. From this work, the final versions of the SWAL-QOL and SWAL-CARE were developed. Although this important work is to be commended, the -item SWAL-QOL is lengthy and time-consuming, so it is somewhat burdensome to use in the average clinical setting.!^ he SWAL-QOL uses a 5-point scale, with different instructions to the patient for different sections of the survey, and a lower score indicates a worse quality of life. he authors of the SWAL-QOL also compared it to bolus flow measurements as assessed retrospectively on videofluoroscopy in the same 86 patients used for the validation studies.''' he SWAL-QOL correlated modestly to most bolus flow measurements; oral transit time and total swallow duration had the best correlation. he authors concluded that the SWAL-QOL provides "independent but complementary" information about the dysphagic patient's experience. he MDADI was designed to assess dysphagiarelated quality of life in patients with head and neck cancer.' * With the recent focus on organ preservation in head and neck cancer treatment, there is a need for clinicians to better understand the various aspects of quality of life. ' ^ he MDADI incorporates domains (emotional, functional, and physical), as well as global question. his -item instrument is shorter than the SWAL-QOL. but scoring is still complicated, witb questions being scored contrary to the others. A lower score indicates a worse quality of life. his instrument was validated on patients with head and neck cancer, and test-retest reliability was assessed on 9 patients. Wallace et al^^ developed a 7-item survey instrument designed to assess symptom severity in patients with oropharyngeal dysphagia. his was validated on 5 patients with oropharyngeal dysphagia and patients with ZD. In the ZD group, the instrument was able to demonstrate an average of 7% improvement in scores after treatment. be maximum possible score on this inventory is,7, with worse symptom severity indicated by a higher score. his instrument has not been adopted into widespread use, primarily because it is based on a visual analog scale, which makes scoring difficult. he validity and reliability of the EA- has been demonstrated in a large cohort of patients with a wide variety of causes of dysphagia. be instrument is not limited to oropharyngeai dysphagia or to patients with bead and neck cancer. he development and validation methods are similar to those of other accepted patient-administered surveys, such as the Voice Handicap Index-.'^" j ^ g survey is rapidly administered and can be completed in less than minutes. he instrument is worded in a manner that makes it simple to total. here are no subscales to address, no visual analog scale to measure.

6 9 Belafsky et al, Eating Assessment ool and no formulas required to calculate a raw score. he clinician only needs to add up the numbers. An elevated EA- score indicates a higher self-perception of dysphagia. be omission of specific domains precludes the stratiflcation of subset disability into social, emotional, and functional categories. We think that this limitation is offset, however, by the test's simplicity, ease of administration and scoring, and utility in persons with a wide range of dysphagia causes. Future research is necessary to replicate these findings in a diverse sample of persons with a variety of causes of dysphagia. Evaluation of EA- data across age groups is necessary to ensure that the normative data are valid for both young and elderly populations. Normative data should also be confirmed across categories of socioeconomic status. gender, and race. A comparison of EA- scores obtained before and after medical and surgical dysphagia therapy in larger cohorts will also help clarify the role of the EA- in the documentation and evaluation of treatment outcomes. CONCLUSIONS he BA- is a self-administered, symptom-speciflc outcome survey for dysphagia. he instrument has displayed excellent internal consistency, testretest reproducibility, and criterion-based validity. he normative data suggest that an EA- score of or more is abnormal. he EA- may be utilized as a clinical instrument to document the initial dysphagia severity and monitor the treatment response in persons with a wide array of swallowing disorders.. Smithard DG, Smeeton NC, Wolfe CD. Long-term outcome after stroke: does dysphagia matter? Age Ageing (X)7: 6:9-.. Smithard DG, O'Neill PA, Parks C, Morris J. Complications and outcome after acute stroke. Does dysphagia matter? Stroke 996:7:-. [Erratum in Stroke 998:9:8-.). Bulow M, Olsson R. Ekberg O. Do dysphagic patients with an absent pharyngeal swallow have a shorter survival than dysphagic patients with pharyngeal swallow? Prognostic importance of a therapeutic videoradiographic swallowing study (VSS). Acta Radiol 5:6:6-.. Eslick GD, alley NJ. Dysphagia: epidemiology, risk factors and impact on quality of life a population-based study. Aliment Pharmacol her 8:7: Bretan O, Henry MA, Kerr-Corrêa F. Dysphagia and emotional distress [in Portuguese]. Arq Gastroenterol 996:: Nguyen NP, Moltz CC, Frank C, et al. Long-term aspiration following treatment for head and neck cancer. Oncology 8:7: Ekberg O, Hamdy S, Woisard V, Wuttge-Hannig A, Ortega P. Social and psychological burden of dysphagia: its impact on diagnosis and treatment. Dysphagia :7: Wilkins, Gillies RA, homas AM, Wagner PJ. he prevalence of dysphagia in primary care patients: a HamesNet Research Network study. J Am Board Fam Med 7:: Mann G, Hankey GJ. Cameron D. Swallowing disorders following acute stroke: prevalence and diagnostic accuracy. Cerebrovasc Dis ::8-6.. Coates C, Bakheit AM. Dysphagia in Parkinson's disease. Eur Neurol 997:8:9-5.. McHomey CA, Bricker DE, Kramer AE, et al. he SWAL-QOL outcomes tool for oropharyngeal dysphagia in adults: I. Conceptual foundation and item development. Dysphagia :5:5-.. McHomey CA, Bricker DE, Robbins J, Kramer AE, Rosenbek JC. Chignell KA. he SWAL-QOL outcomes tool for oropharyngeal dysphagia in adults: H. Item reduction and preliminary scaling. Dysphagia :5:-. REFERENCES. McHomey CA, Robbins J, Lomax K, et al. he SWAL- QOL and SWAL-CARE outcomes tool for oropharyngeal dysphagia in adults: III. Documentation of reliability and validity. Dysphagia :7:97-.. Chen AY. Frankowski R. Bishop-Leone J, et al. he development and validation of a dysphagia-specific quality-of-hfe questionnaire for patients with head and neck cancer: the M. D. Anderson dysphagia inventory. Arch Otolaryngol Head Neck Surg :7: Wallace KL, Middleton S, Cook IJ. Development and validation of a self-report symptom inventory to assess the severity of oral-pharyngeal dysphagia. Gastroenterology :8: Langmore SE. An important tool for measuring quality of life. Dysphagia :5: McHomey CA, Martin-Harris B, Robbins J, Rosenbek J. Clinical validity of the SWAL-QOL and SWAL-CARE outcome tools with respect to bolus flow measures. Dysphagia 6:: Schwartz S, Patrick DL. Yueh B. Quality-of-life outcomes in the evaluation of head and neck cancer treatments. Arch Otolaryngol Head Neck Surg :7: Rosen CA, Lee AS, Osbome J, Zullo, Murry. Development and validation of the Voice Handicap Index-. Laryngoscope :: French C, Irwin RS, Fletcher KE, Adams M. Evaluation of a cough-specific quality-of-life questionnaire. Chest ::-.. Birring SS, Prudon B, Carr AJ, Singh SJ, Morgan MD, Pavord (D. Development of a symptom specific health status measure for patients with chronic cough: Leicester Cough Questionnaire (LCQ). horax :58:9-.. Atlas SJ, Gallagher PM, Wu YA, et al. Development and validation of a new health-related quality of life instrument for patients with sinusitis. Qual Life Res 5:: Alexander M. Berger W, Buchholz P, et al. he reliability, validity, and preliminary responsiveness of the Eye Allergy Patient Impact Questionnaire (EAPIQ). Health Qual Life Outcomes 6::67.

7