The effect of a training and clinical facilitation programme for registered midwives in primary maternity settings with respect to managing labour: a pragmatic cluster randomised trial

Abstract

Background: Intrapartum complications contribute to nearly half of all avoidable maternal and perinatal deaths nationally. Inadequate understanding of the labour process by midwives, poor documentation of labour monitoring and inadequate systemlevel support may lead to wrong diagnosis, incorrect management, and the potential for missed opportunities to prevent mortality. Aim: To evaluate the effect of an intervention package of training and clinical facilitation on the quality of clinical management in labour by registered midwives in primary level public sector health facilities in rural South Africa. Methods: Research design : Pragmatic cluster randomised trial with 12 month follow-up. Setting and participants : Seventeen clusters stratified by geo-political region and size of service; 1020 labour records (60 per cluster / site; systematic random sample); and 154 registered midwives employed in the study sites during the study period. Participants were not blinded. Intervention : A package of clinical faclitation training for selected experienced midwife clinicians / managers, and an intrapartum educational update for midwives. Intervention and control sites continued receiving routine communication, all clinical guidelines and scheduled outreach activities. Main outcome measures : Primary outcome - clinical practice measuring partograph utilisation, using a modified partograph checklist, the testing of which is described in this study. Secondary outcome - midwives’ knowledge and skills, measured by written and clinical tests. Outcomes were analysed at the individual level using regression methods that allowed for clustering. The evaluator was blinded to the study allocation. Findings: The mean scores for the total partograph were not statistically significantly different between arms; the mean difference was 1.55 points out of a possible score 47 (95% CI: -1.18 to 4.28) p= 0.27. At a score of 27 the estimated absolute difference was 13.6% (95% CI : 0.16 to 0.25) p = 0.026. The total score for midwives’ knowledge and skills was 7 points (out of a possible 119) higher in the intervention arm (95% CI : 2.1 to 12.3), p=0.006. Conclusions: Although there was no difference in the quality of the overall completion of the partograph, there was a statistically significant difference in those of better quality completions in the intervention arm. Midwives’ knowledge and skills were higher in the intervention arm and those in the control arm deteriorated over time. This difference was statistically significant. Recommendations and implications for practice: This indicates a critical need to provide continuing professional education to midwives and to arrange midwifery staffing that optimises clinical practice in settings where intrapartum care is offered. In addition to regular, sustainable programmes to enhance partograph utilisation and midwife knowledge and skills, barriers to the utilisation of the partograph need to be investigated and addressed.