|
Neurology India, Vol. 54, No. 3, July-September, 2006, pp. 316-317 Letter To Editor Authors' reply Pandian JeyarajDurai, Kalra G, Jaison A, Deepak SS, Shamsher S, Singh Y Department of Neurology, Royal Brisbane and Women's Hospital, Herston Road, Brisbane, Queensland 4029, Australia
Code Number: ni06105 Sir, I appreciate the curiosity shown and the comments received for our original article published recently in Neurology India.[1] A great deal of attention has been focussed on the survey instrument that was used in the study. As stated in the paper, we adapted and modified the questionnaire from our previous survey among general public in Northwest India.[2] Before using the questionnaire we did translate and back translate the contents of the questionnaire from English to the relevant local vernacular languages. A separate copy of the questionnaire was used for this particular study. I fully agree with the concern that was raised regarding the inter-rater reliability and validity of the questionnaire. Using existing questionnaires or instruments save time and effort. It also avoids the need to establish the reliability and validity of the instrument.[3] If an existing instrument meets 80% or more of the needs of a researcher, it can be used with minor modifications after pre-testing.[3] Pre-testing is designed to address the content validity of the questionnaire.[4] If the question is ambiguous or misunderstood, the answers obtained in the study will not be useful. Pre-testing will typically pick up most of the serious problems in comprehension of an instrument.[4] The questionnaire in the present study was pre-tested in a sample of 25 patients and relevant changes were made to various terms that are used for "stroke" in the local languages Punjabi and Hindi.[1] Other Researchers have also followed similar methodology of pre-testing the questionnaire both in stroke and epilepsy without focussing on reliability and validity.[5],[6],[7],[8] The demographic data is clearly shown in [Table - 1] (Interviewees). We did not adapt the educational status from the western culture. The educational categories that we had used are relevant to the study cohort. For the purpose of statistical analysis we had put together illiterates and the primary education group. Education was a significant factor in multivariable analysis, for knowing the organ affected in stroke and the knowledge of warning symptoms of stroke. The results are not going to change even if we re-group into illiterates vs. others. Only when the patient′s were unable to participate in the interview (aphasia, loss of consciousness etc.) the relative was interviewed. Moreover, in the sub-group analysis (patients vs. relatives) we did not find any difference in the knowledge of stroke between the two groups. It is almost impossible to carry out a perfect survey and every study has its own limitations. India is a vast country with diverse socio-cultural practices. Our findings can not be extrapolated to other parts of our country. Similar studies are needed from other states, in order to design population specific educational programs about stroke. References
Copyright 2006 - Neurology India |
|