Translation and Validation of an Epilepsy Screening Instrument in Two Ghanaian Languages

Abstract

IntroductionThe prevalence of epilepsy in sub-Saharan Africa varies considerably, and the exact estimate in Ghana is unknown, with few data available from peri-urban areas. More community-based studies are required to understand the tangible burden of epilepsy in these areas and the difficulties in healthcare access.ObjectiveTo validate a household survey epilepsy-screening instrument in Shai-Osudoku and Ningo-Prampram District of Greater Accra Region, Ghana.MethodsWe developed a 17-item epilepsy screening instrument by modifying validated English language questionnaires. We included questions that could identify convulsive and non-convulsive seizures. Language experts translated and back-translated the survey instrument into the two languages in this region: Asante Twi and Dangme. Cases were people with epilepsy attending healthcare facilities where these languages are used. Controls were unaffected relatives of cases or people attending the healthcare centres for other medical conditions. We matched cases and controls for geographical location and ethnicity. An affirmative response to one of the seventeen questions was deemed as a positive screen.ResultsOne hundred and Forty Dangme Twi speakers (70 cases and 70 controls) and 100 Twi speakers (50 cases and 50 controls) were recruited. The sensitivity and specificity for Dangme were: Stage 1;100% (95% CI: 88.6, 94.9) and 80% (95% CI: 68.7, 88.6) and Stage 2, 98.6% (95% CI: 92.3, 100.0) and 85.7% (95% CI: 75.3, 92.9). The Dangme version reliably identified epilepsy with positive predictive values of 83.3% (95% CI: 73.6, 90.6) and 87.3% (95% CI: 78.6, 90.6) at stages 1 and 2. The questionnaire excluded epilepsy with negative predictive values of 100% (95% CI: 93.6, 100.0) and 98.4% (95% CI: 91.2, 100.0). For the Twi version, the sensitivity and specificity were: 98% (95% CI: 89.4, 99.9) and 92% (95% CI: 80.8, 97.8) at Stage 1, and for Stage 2, 96% (95% CI: 86.3, 99.5) and 94% (95% CI 83.5, 98.7). The Twi questionnaire reliably specified epilepsy with positive predictive values of 92.5% (95% CI: 81.8%, 97.9) and 94.1% (95% CI: 83.8, 98.8) at stages 1 and 2. It excluded epilepsy with negative predictive values of 97.9% (95% CI: 88.7, 99.9) and 95.9% (95% CI: 86.0, 99.5) for the two-stagesConclusionsOur questionnaire is valid for the two tested languages and is usable for community-based epilepsy surveys in Ghana. The questionnaire can be adapted for other resource-poor settings, although there will need to be translation and iterative in-country testing to ensure its validity is maintained.