Directly observed and reported respectful maternity care received during childbirth in public health facilities, Ibadan Metropolis, Nigeria.

Respectful maternity care (RMC) is believed to improve women's childbirth experience and increase health facility delivery. Unfortunately, few women in low- and middle-income countries experience RMC. Patient surveys and independent observations have been used to evaluate RMC, though seldom together. In this study, we assessed RMC received by women using two methodologies and evaluated the associated factors of RMC received. This was a cross-sectional study conducted in nine public health facilities in Ibadan, a large metropolis in Nigeria. We selected 269 pregnant women by cluster sampling. External clinical observers observed them during childbirth using the 29-item Maternal and Child Health Integrated Program RMC observational checklist. The same women were interviewed postpartum using the 15-item RMC scale for self-reported RMC. We analysed total RMC scores and RMC sub-category scores for each tool. All scores were converted to a percentage of the maximum possible to facilitate comparison. Correlation and agreement between the observed and reported RMC scores were determined using Pearson's correlation and Bland-Altman analysis respectively. Multiple linear regression was used to identify factors associated with observed RMC. No woman received 100% of the observed RMC items. Self-reported RMC scores were much higher than those observed. The two measures were weakly positively correlated (rho = 0.164, 95%CI: 0.045-0.278, p = 0.007), but had poor agreement. The lowest scoring sub-categories of observed RMC were information and consent (14.0%), then privacy (28.0%). Twenty-eight percent of women (95%CI: 23.0% -33.0%) were observed to be hit during labour and only 8.2% (95%CI: 4.0%-18.0%) received pain relief. Equitable care was the highest sub-category for both observed and reported RMC. Being employed and having completed post-secondary education were significantly associated with higher observed RMC scores. There were also significant facility differences in observed RMC. In conclusion, the women reported higher levels of RMC than were observed indicating that these two methodologies to evaluate RMC give very different results. More consensus and standardisation are required in determining the cut-offs to quantify the proportion of women receiving RMC. The low levels of RMC observed in the study require attention, and it is important to ensure that women are treated equitably, irrespective of personal characteristics or facility context.