The effectiveness of antenatal birth plans in increasing skilled care at delivery and after delivery in rural Tanzania a cluster randomised trial

Objective: To determine the effectiveness of ANC birth plans (birth preparedness and complication readiness) in increasing skilled care at delivery and after delivery Rationale: Although birth plans are key elements of focused ANC in many developing countries including Tanzania that aim to increase skilled care utilization at delivery, after delivery and during emergency obstetric complications, robust empirical evidence is lacking on their effectiveness. Methods The study was a cluster randomized controlled trial (RCT) that was conducted in Ngorongoro district, rural northern Tanzania. The primary outcome was the proportion of women who delivered at the available health units and the secondary outcomes were the proportion of women who sought postnatal care within one month of delivery and women's and providers' satisfaction with ANC. The study was implemented in three phases: the formative qualitative study that aimed to understand contextual factors for the high ANC care coverage but low utilization of health facilities for delivery and how the intervention could be implemented in the study district, a RCT to determine the effectiveness of the intervention in increasing skilled attendance at delivery and postpartum, and a process evaluation of the intervention and control arms' ANC. The qualitative study involved 15 focus group discussions, 12 key informant interviews and participant observation of the ANC and delivery care at randomly selected health units. Eight health units were randomly assigned to the intervention (antenatal care with an emphasis on birth plans by care providers) and an equal number to the control group (care as provided currently). A total of 905 consenting pregnant women (404 in the intervention arm and 501 in the control) at 24 weeks of gestation and above were recruited and followed up to the initial postnatal care clinic attendance or during the postnatal interview at home depending on which occurred first. Results Both demand and supply sides factors prevented women from utilizing health units for delivery and immediate postnatal care, despite the high level of ANC uptake. Notably, women's lack of planning for accessing delivery care at health units, norms and traditions dictating that home delivery is equally safe and health system deficiencies (structural, process and outcome) were the key barriers identified. 2 Unpaired t-test statistic was used to assess the effectiveness of the intervention on the primary and secondary outcomes taking into account the clustering effect. Overall, 34.8% of women in the intervention arm and 20.3% in the control delivered in the health facilities (difference in proportion: 14.5% [-9.4-38.3] p=0.2138 for the crude analysis and 16.8% [2.6-31.0] p=0.0248 for the adjusted analysis). Postnatal care utilization was 62.1% in the intervention and 32.1% in the control group (difference in proportion for the crude and adjusted analysis 30.0% [11.3-48.7] p=0.0040 and 31.3% [15.4-47.2] p=0.0009 respectively). Altogether, 96.8% of women in the intervention and 84.7% in the control units were satisfied with the ANC they received (difference in proportion: 12.1% [-6.3-30.5] p=0.1668 for the crude and 12.6% [-5.4-30.5] p=0.1454 adjusted analysis). Similarly, 97.9% and 91.0% of providers in the intervention and control arms were satisfied with the ANC they provided (difference in proportion: crude analysis 6.9% [-3.2-17.1] p=0.1547 and adjusted analysis 7.8% [-0.7-16.3] p=0.0688). Overall, the intervention was implemented as per study protocol. The average time for initial ANC consultation in the intervention arm of the study was 40.1 minutes (range 33-47 minutes) compared to 19.9 minutes (range 12-32 minutes) in the control arm p<0.0001. The average time for consultation during follow-up ANC visits was 23.3 minutes (rangel5-31) for the intervention units versus 10.3 minutes (range 6-17)in the control p=0.0001. Likewise, providers in the intervention units spent more time for counselling/health education or promotion than in the control units at both initial ANC attendance and during subsequent visits (average time at initial attendance 24.5 minutes, range 19-32 in the intervention vs 10.5 minutes, range 5-18 in the control arm) p<0.0001 ). The respective time for follow-up visits was 13.8 minutes (range 6-17) vs 4.5 minutes (range 0-10) p=0.0001. Nevertheless, the improvement was largely on the discussion on birth plans and PMTCT, and not on the other topics in the national focused ANC guidelines. Conclusion and implication for practice A well-implemented antenatal birth plan intervention improved women's utilization of health units for delivery, and post delivery without substantially affecting the women's and providers' satisfaction with ANC. Implementation of birth plans in health care settings in low resource settings like Ngorongoro is feasible and should be promoted as an effective strategy to increase skilled delivery and postnatal uptake.