Skip to main content

Table 3 A comparison of dependent and independent variables for rural and urban respondents in the pre- and post-intervention groups

From: The spillover effect of midwife attrition from the Nigerian midwives service scheme

  Rural respondents Urban respondents
Characteristics Last childbirth occurred before 2010 Last childbirth occurred between 2010 - 2011 Difference T-test p-value Last childbirth occurred before 2010 Last childbirth occurred between 2010 - 2011 Difference T-test p-value
Mean age (years) 29.9 28.5 −1.4 < 0.001 30.9 29.7 −1.2 < 0.001
Average number of children ever born 4.4 4.2 − 0.2 < 0.001 3.9 3.6 −0.3 < 0.001
Mean number of years of formal education received 3.4 3.4 0.0 < 0.956 7.4 8.1 + 0.7 < 0.001
Percentage in the middle or a lower wealth quintile 83.3 84.0 0.7 0.128 28.4 27.9 −0.5 0.532
Other characteristics
 Any antenatal visit to public-sector clinics (%) 36.1 45.5 + 9.4 < 0.001 54.4 59.6 + 5.2 < 0.001
 Any antenatal visit to private clinics (%) 9.2 7.6 −1.6 < 0.001 26.6 26.6 0.0 0.982
 Antenatal visits to other sites/No antenatal care 52.8 46.3 −6.5 < 0.001 18.3 13.4 −4.9 < 0.001
 Deliveries in public-sector clinics/hospitals (%) 16.3 18.0 + 1.7 < 0.001 34.9 39.1 + 4.3% < 0.001
 Deliveries in private-sector clinics/hospitals (%) 7.9 6.3 −0.16 < 0.001 24.4 25.6 + 1.1 0.162
 Deliveries at home (%) 74.7 75.4 + 0.7 0.208 37.9 35.0 −2.9 0.001
Barriers to accessing primary healthcare services
 -those in need of permission (%) 16.4 14.7 −1.7 < 0.001 11.4 6.7 −4.7 < 0.001
 -those who couldn’t afford the cost of care (%) 63.3 52.1 −11.2 < 0.001 43.1 33.1 −10.0 < 0.001
 -those who thought the closest clinic/hospital was too far (%) 46.2 40.0 −6.3 < 0.001 20.5 15.7 −4.8 < 0.001
 -those in need of company (%) 21.2 18.1 −3.1 < 0.001 9.6 7.0 −2.6 < 0.001
 Number of observations 15,495 11,806 6224 5737
  1. The pre- and post-intervention groups made up of respondents whose most-recent birth occurred between 2003 and 2009 and 2010 – 2013 respectively. In the difference column, “+” indicates an increase, while “-” indicates a decrease. The significant reduction in barriers to accessing primary healthcare services over time may be a consequence of the MSS, or may mediate the effect of the scheme. However, we chose to leave them in the models to avoid omitted variable bias, and because our conclusions were robust to their exclusion
\