The validity of the Shona version of the EQ-5D, of life measure

Abstract

Introduction: The EQ-5D, a generic health related quality of life measure developed in Europe, has been recently translated into Shona, the language of the majority of Zimbabweans. Although the reliability of the Shona version of the EQ-5D has been established within a community setting, the reliability and validity of the measure within a group of Shona speaking people with disabilities has not been examined.

Aim: The aim of the study was to examine the reliability and concurrent validity of the Shona version of the EQ-5D, within the context of a house-to-house survey of disability in a high-density area of Harare, Zimbabwe.

Methods: As part of a house-to-house survey of disability in a high-density area in Zimbabwe, 588 Shona speaking subjects with disability/morbidity or their proxies were asked to respond to the Shona version of the EQ-5D questionnaire. Those who were able to understand the concept filled in the visual analogue scale.  A testretest was done to determine the reliability of the EQ-5D.  Timed walking was used to investigate the validity of the
domain of mobility and the International Classification of Impairment, Disability and Handicap Beta Draft  (ICIDH2) was used as the gold standard for usual activities and self care. The concurrent validity of the anxiety/depression domain was determined against the Shona Symptom Questionnaire, which is a validated screen for depression in the Zimbabwe population. No measure of pain could be found which had been validated in Zimbabwe.

Data analysis: The Intraclass-correlation  (ICC) and Pearson’s correlation co-efficient were used to determine the test re-test reliability of the descriptor section and visual analogue scale of the EQ-5D respectively. The t-test, ANOVA, and post-hoc Scheffe test were used to compare the EQ-5D with the measures of function.

Results:
Each domain of the Shona EQ-5D demonstrated reliability on the test re-test (ICC ranging from .58 for self care to .85 for mobility, p<.01).  The first and second scores on the visual analogue scale were significantly correlated  
( Pearson’s r=.79, p<.001).  Those who reported no problems with mobility walked significantly faster than those who reported some problems with mobility (t=-6.2, p<.001). The mean number of activity limitations using the International Classification of Functioning was significantly different between  those who reported no, some or severe limitations in
usual activities  (F=39.9 p<.001).  Of those reporting no, some and severe problems with self-care .6%, 13.6% and 62.5% respectively were found to have functional limitation in dressing on the ICIDH2  (the numbers were too small to apply statistical analysis). There was a significant difference between the mean number of affirmative answers in the Shona Symptom Questionnaire in respondents who reported no, moderate and severe problems with anxiety/depression  (F=70.7, p<.001).

Discussion and conclusion: It is concluded that the EQ-5D is a robust indicator of health related quality of life across different cultures. It is suggested that the Shona version of the EQ-5D can be used with confidence in a sample of Shona speaking subjects. Physiotherapists in the region are encouraged to translate and validate questionnaires to ensure that research with non-English speaking members of the Southern African population is performed with appropriate instrumentation