The effect of traditional birth attendant training on maternal and neonatal care.

OBJECTIVE To determine whether traditional birth attendants (TBAs) trained via the "SMART Dai" method were superior to untrained TBAs in knowledge and practice regarding maternal and newborn care. METHODS In a cluster-randomized trial in the Dera Ghazi Khan District of Punjab, Pakistan, 120 rural communities each with a population of approximately 5000 were randomly assigned to a community-based intervention (CBI) or a health systems intervention (HSI). In the CBI communities, 288 TBAs underwent an innovative 8-day training course on maternal and newborn care, initially evaluated by pre- and post-tests. After an average of 19 months post-training, 277 TBAs, together with 257 comparably chosen untrained TBAs from the HSI communities, were tested and interviewed. Patients from both referred and non-referred deliveries were also interviewed. RESULTS Characteristics of TBAs in the two groups were similar. The TBAs were evaluated according to various measures of knowledge, skill, and practice (including referral), with patient reports on practice compared with TBA reports. By most measures, trained TBAs outperformed untrained ones, often to significant degrees. CONCLUSION SMART Dai training seemed to be an important factor in the significant reduction in perinatal mortality in the CBI areas. Properly trained TBAs can substantially contribute to improved delivery outcomes.