Linguistic validation of the Greek M.D. Anderson Symptom Inventory – Head and Neck Module

Abstract

Background

Our goal is to linguistically validate the Greek translation of the M.D. Anderson Symptom Inventory Index – Head and Neck Module.

Patients & Methods

Following forward and backward translation of the previously validated head and neck cancer specific items of the English MDASI-HN into Greek (G-MDASI-HN), it was administered along with a cognitive debriefing to head and neck cancer patients able to read and understand Greek. Individual and group responses are presented using descriptive statistics.

Results

From 02/2009 through 06/2009 30 subjects with head and neck cancer completed the G-MDASI-HN followed by completion of the accompanying cognitive debriefing. Ninety-eight percent of the individual G-MDASI-HN items were completed. “Voice” item was not completed by 5 patients. Average time to complete the G-MDASI-HN was 13.3 minutes. Average ease of completion was rated at 1.21 on a 0 to 10 scale with “0” being “very easy” and “10” being “very hard”. Only 10% of patients reported trouble completing any item, namely “distress” and “numbness”.

Conclusions

The Greek-MDASI-HN is linguistically valid and a patient-reported instrument that can be used both in outcomes research and as a clinical tool.

Introduction

Patients with cancer may be experiencing substantial tumor- or treatment-related symptoms, which can have great impact on their overall comfort and function. Optimal symptom control requires adequate and ongoing symptom assessment and should be guided by patient report, rather than by physician rating alone. Use of patient symptom reports for clinical decision making and effectiveness research may be preferred over quality of life measures, as patient symptoms are felt to more closely reflect the disease and treatment process (1). The M.D. Anderson Symptom Inventory (MDASI) is a brief, reliable, and validated patient-reported questionnaire designed to capture and quantify general cancer- and treatment-related symptoms, which can help guide patient-specific and programmatic evaluations and interventions. The MDASI contains 13 core items representing important symptoms common across all cancer types and 6 items of how these symptoms interfere with major activities of daily life (2). A Greek version of the MDASI has been previously validated in terms of content, construct, reliability, and known group validity (3).

The MDASI was designed so that modules for specific tumor and treatment sites could be developed. For a given anatomic location and depending on local tumor extent, patients with head and neck cancer can be subjected to a number of unique and serious symptoms. Furthermore, patients with head and neck cancer are commonly treated with a combined modality approach (combinations of chemotherapy, surgery, and/or radiation), known to be associated with significant acute and long term toxicity. The MDASI-head and neck module (MDASI-HN) is a validated disease site specific instrument, inclusive of the same 13 core and 6 interference items, plus an additional 9 tumor- and treatment-related symptoms important in head and neck cancer patients (4).

In order to ensure inclusion of Greek speaking head and neck cancer patients in future symptom prevention and intervention research studies that use the MDASI-HN as a primary endpoint measure and to allow integration of the MDASI-HN as a clinical assessment tool in primary Greek-speaking regions, our goal is to linguistically validate the Greek version of the MDASI-HN (G-MDASI-HN).

Patients & Methods

The MDASI had previously been translated into a Greek language version (G-MDASI) (3). In order to develop the G-MDASI-HN, the 9 head and neck cancer specific items of the MDASI-HN were subsequently translated into Greek using standard forward and backward translation methods, procedures that we have been following as necessary first steps when psychometrically validating foreign language versions of the MDASI (5–9) and were recommended by an international task force (10).

Consecutive adult patients with malignancy of the head and neck region, able read and understand Greek, were recruited in the Department of Radiation Oncology at the Democritus University of Thrace in Alexandroupolis, Greece. The G-MDASI-HN was self-administered by the participating patients and was completed via pencil and paper. All G-MDASI-HN symptom items are rated on 0 to 10 numeric scales from “not present” to “as bad as you can imagine”, and the G-MDASI-HN interference items are rated on 0 to 10 numeric scales from “did not interfere” to “interfered completely." Time taken by each participant to complete the G-MDASI-HN was recorded by nursing staff. Since the purpose of this study was purely linguistic validation, patient demographic, tumor, and treatment details were not recorded.

To ensure ease of completion, relevance, and comprehensibleness of this translated version, and in keeping with recent recommendations, subjects also completed a cognitive debriefing of the G-MDASI-HN (10; 11). The cognitive debriefing was completed with the assistance of nursing staff, who were both Greek and English speaking. Subjects were asked to rate overall ease of completion of the G-MDASI-HN. Subjects were queried if they were comfortable answering each specific item, if any item was unclear, and if they had any suggestion on how to make any item better. Subjects were also asked if any item was redundant, if any item should be deleted, or if any item should be added. Here we present the G-MDASI-HN item response rate and cognitive debriefing results using descriptive statistics.

Results

From 02/2009 through 06/2009 30 subjects participated and completed the G-MDASI-HN and the accompanying cognitive debriefing. Overall, 822 of the possible 840 (98%) individual G-MDASI-HN items were completed by the subjects. The most and second most likely items to be left blank by subjects were problems with “voice” and “constipation”, which were not completed by 5/30 and 3/30 subjects respectively.

Average time to complete the G-MDASI-HN was 13.3 minutes (range 5–30 minutes). Average G-MDASI-HN ease of completion was rated at 1.21 (range 0 to 7) on a 0 to 10 scale with “0” being “very easy” and “10” being “very hard”. The majority of participants (19/30) thought that all G-MDASI-HN items, question, phrases, and words were easy to understand. Of the remaining 11, 6 subjects reported difficulty understanding “distress” or “numbness” items (3 for each item). Other individual items rated with some difficulty in understanding were “drowsy” (1 subject), “pain” (1 subject), “sad” (1 subject), “voice” (2 subjects), and “relate” (1 subject). All (30/30) subjects reported feeling “comfortable” answering each item. No subject thought any specific item should be deleted, and one subject suggested adding an alopecia-related item.

Discussion

Here we report the linguistic validation results of the G-MDASI-HN. These cognitive debriefing results suggest overall ease of completion, relevance, and comprehensibleness of this translated patient-reported instrument in this Greek patient population. However a few points require some further discussion. Problems with “voice” was left blank by 5/30 subjects. While we don’t have patient, tumor, or previous treatment details available, during cognitive debriefing two of these 5 queried whether this question pertained to “before or after laryngectomy”. However, the MDASI-HN asks patients to rate the severity of their symptoms on a 0–10 scale over the past 24 hours. Therefore, we hypothesize that this question was left blank by these two subjects not because of trouble understanding the “voice” item, but rather failure to rate this item over the last 24 hour period and/or how to respond on a 0–10 scale if they had no speech (i.e. the patients many have had a laryngectomy without ability for speech). Ten percent of the subjects reported difficulty understanding the individual item related to “distress”. Ten percent of subjects also reported difficulty understanding the “numbness” item. However, upon cognitive debriefing all three subjects asked for clarification of location of numbness, suggesting that they understood “numbness” but preferred to further characterize this symptom by describing location, rather than strictly assigning a severity rating. Since these items are part of the core 13 core items from the MDASI, which has been previously validated in larger study of Greek speaking subjects (3), we continue to include these particular items in the G-MDASI-HN.

In conclusion, the G-MDASI-HN is a linguistically valid disease site specific version of the G-MDASI and can be a useful instrument in patient-reported outcomes research and a clinical tool to allow rapid identification of head and neck cancer patient specific symptoms in need of intervention.