Background
Many women in Sub-Saharan African countries do not receive key recommended interventions during routine antenatal care (ANC) including information on pregnancy, related complications, and importance of skilled delivery attendance. We undertook a process evaluation of a successful cluster randomized trial testing the effectiveness of birth plans in increasing utilization of skilled delivery and postnatal care in Ngorongoro district, rural Tanzania, to document the time spent by health care providers on providing the recommended components of ANC.

Methods
The study was conducted in 16 health units (eight units in each arm of the trial). We observed, timed, and audio-recorded ANC consultations to assess the total time providers spent with each woman and the time spent for the delivery of each component of care. T-test statistics were used to compare the total time and time spent for the various components of ANC in the two arms of the trial. We also identified the topics discussed during the counselling and health education sessions, and examined the quality of the provider-woman interaction.

Results
The mean total duration for initial ANC consultations was 40.1 minutes (range 33-47) in the intervention arm versus 19.9 (range 12-32) in the control arm p < 0.0001. Except for drug administration, which was the same in both arms of the trial, the time spent on each component of care was also greater in the intervention health units. Similar trends were observed for subsequent ANC consultations. Birth plans were always discussed in the intervention health units. Counselling on HIV/AIDS was also prioritized, especially in the control health units. Most other recommended topics (e.g. danger signs during pregnancy) were rarely discussed.

Conclusion
Although the implementation of birth plans in the intervention health units improved provider-women dialogue on skilled delivery attendance, most recommended topics critical to improving maternal and newborn survival were rarely covered.