Introduction: Though all pregnant women got registered, prenatal depression was not looked for and treated. For the Government to undertake screening, locally validated screening scale was not available. Hence, this study was undertaken with primary objective of validating EPDS scale Tamil translation. Secondary objectives were to estimate prenatal depression prevalence and to arrive at shorter scale.
Methods: Cross sectional study was conducted among rural Tamil speaking pregnant women. Mini International Neuro Psychiatric Interview was used to confirm diagnosis. ROC curve was constructed. Test retest reliability and internal consistency were calculated. Item based regression analysis was done to predict items to constitute short scale. ROC curve was constructed for short scale. Prevalence of prenatal depression was calculated.
Results : Best cut off for full scale was 9 or more. Sensitivity, specificity, positive & negative predictive values, positive and negative likelihood ratios were 92%, 77%, 34%, 99%, 4 and 0.1 respectively. Intra class correlation coefficient 0.86; Cronbach’s alpha 0.77. Items 1, 6, 7 and 8 emerged as predictors for short scale; For best cut off of >=3, sensitivity, specificity, positive & negative predictive value, Positive and negative likelihood ratio were 87%, 80%, 35%, 98%, 4.37 and 0.16 respectively. Prevalence of prenatal depression was 11.1 % (8.1 to 15%).
Conclusions : EPDS Tamil translation could be a valid and reliable tool to screen for prenatal depression. Short scale could be of similar value. Prenatal depression was of considerable magnitude.