Table 3. Change in maternal health indicators associated with the intervention, by analytical approach, 2011–2012, Zambia.

Indicator	Unadjusted			Adjusted for covariates			Propensity score matching
							Calliper			Radius			Nearest neighbour
	Difference-in-difference estimatea	tb		Difference-in-difference estimatea	tb		Difference-in-difference estimatea	tb		Difference-in-difference estimatea	tb		Difference-in-difference estimatea	tb
Women who used emergency transport	14.0	14.26*		14.6	14*		12.4	2.81*		13.1	2.75*		18.7	3.47*
Women who knew three obstetric danger signs	10.1	3.85*		10.5	3.79*		14.2	4.25*		14.9	4.44*		10.3	3.79*
Women who knew they should receive antenatal care in the first trimester	18.1	7.14*		16.6	6.18*		15.5	4.97*		14.5	4.43*		15.7	6.54*
Women who received four or more antenatal care visits from a skilled birth attendant, with the first in the first trimester	5.5	2.08*		4.7	1.70		5.0	1.52		6.1	1.68		4.0	1.4
All deliveries involving a skilled birth attendant	21.4	8.32*		18.4	6.8*		24.0	7.51*		22.0	6.48*		22.0	8.56*
Deliveries involving a skilled birth attendant at a health-care facility	19.3	6.02*		16.3	6.02*		21.0	6.23*		18.7	5.2*		16.4	5.9*
Women who received postnatal care within 6 days	2.3	0.88		0.6	0.23		2.5	0.75		5.3	1.50		0.2	0.08
Women who used modern contraception immediately after giving birth	6.8	2.57*		7.1	2.55*		4.0	1.29		6.0	1.66		6.0	2.09*

* P < 0.05.

^a The difference-in-difference estimate is a measure of the difference between the change from before to after treatment in intervention districts and the change from before to after treatment in control districts. It is given in percentage points.

^b The t-statistic is the ratio of the change in percentage to the standard error.