Table 4. Maternal health-care voucher schemes and the use of maternity care, Cambodia, 2005–2010.

Women offered vouchers	Estimated percentage point change in probabilitya of outcome attributable to the voucher scheme
	Three or more antenatal care visits			Delivery in a public health-care facility			Postnatal care
	Mean	SEb		Mean	SEb		Mean	SEb
All voucher schemes
All	3.2	2.3		10.1**	4.4		5.3**	2.4
Poorc	4.8	3.6		15.6**	4.6		4.6	4.5
Non-poor	2.1	2.7		5.3	4.7		6.8**	1.8
Targeted voucher schemesd
All	−1.6	2.8		7.5	5.2		6.4*	3.9
Poorc	−3.4	4.0		11.3**	5.4		7.4	6.2
Non-poor	1.6	5.2		2.3	5.8		5.6***	2.1
Universal voucher schemes
All	5.4**	2.4		11.8**	5.8		4.7	2.9
Poorc	10.1**	3.9		17.8***	5.0		2.4	5.5
Non-poor	2.9	2.5		7.0	5.6		6.0***	1.8
No. of observations	4869	NA		7221	NA		5656	NA

NA, not applicable; SE, standard error; *, P < 0.10; **, P < 0.05; ***, P < 0.01 (Z test of no effect).

^a The table shows the partial effect of the voucher schemes on the probability of each outcome, in percentage points, estimated using logit models that included the covariates listed in Table 3 plus the birth period (i.e. month and year) and district fixed effects. The effect of a voucher scheme was averaged over all births in intervention districts when the voucher scheme was in operation.

^b SEs were adjusted for clustering at the operational district level.

^c Women in the poorest 40% of households, as determined using a wealth index, were regarded as poor. Non-poor women were from the other 60% of households.

^d Schemes targeted poor women.

Note: Data were obtained from the 2010 Cambodian Demographic and Health Survey.