. 2016 May 23;6:19. doi: 10.1186/s13561-016-0096-4

Table 5.

Impact of the FHCI on the utilisation of maternal health services

	Model 1 (M1)		Model 2 (M2)		LL test (M2 vs. M1)
	May’10	June’10^a	May’10	June’10
Antenatal care (ANC) visits in public facilities^b
Coefficients
Time trend			0.016	0.019^*
Policy	0.282^**	0.245^*	1.066^**	1.059^*
Policy * Time trend			−0.026^*	−0.029^*
Marginal effects (%)
Policy	1.0^*	0.8^*	4.2	4.1	3.38
Four ANC visits in public facilities (ANC4)^b
Coefficients
Time trend			0.004	0.003
Policy	0.118^*	0.125^**	0.260	0.396
Policy * Time trend			−0.005	−0.007
Marginal effects (%)
Policy	2.0^*	2.2^**	4.6	7.0	0.7
Delivery in public facility
Coefficients
Time trend			0.012^*	0.006
Policy	0.231^***	0.295^***	0.208^*	0.330^**
Policy * Time trend			−0.007	−0.005
Marginal effects (%)
Policy	4.8^***	6.1^***	4.3^*	6.8^**	9.34
Delivery in private facility^c
Coefficients
Time trend			0.002	0.009
Policy	0.081	0.081	−0.236	−0.410
Policy * Time trend			0.007	0.005
Delivery with trained health worker^d
Coefficients
Time trend			0.014^**	0.005
Policy	0.241^***	0.310^***	0.135	0.263^*
Policy * Time trend			−0.007	−0.002
Marginal effects (%)
Policy	4.6^***	6.0^***	2.6	5.1^*	16.87
Postnatal care within 48 h
Coefficients
Time trend			0.033^***	0.034^***
Policy	1.046^***	1.072^***	0.635^***	0.814^***
Policy * Time trend			−0.013	−0.017^**
Marginal effects (%)
Policy	14.1^***	15.2^***	9.1^***	11.7^***	89.8
Vitamin A given postnatally
Coefficients
Time trend			0.002	0.002
Policy	0.035	−0.081	0.782^***	0.924^***
Policy * trend			−0.022^*	−0.024^**
Marginal effects (%)
Policy	0.5	−1.2	13.2^***	15.6^***	116
Infant DPT full course^{e, f}
Coefficients
Time trend			0.026^***	0.026^***
Policy	0.670^***	0.671^***	0.536^***	0.727^***
Policy * Time trend			−0.014	−0.019^**
Marginal effects (%)
Policy	13.4^***	13.3^***	11.0^***	14.0^***	574

All models control for covariates listed in Table 4 & estimated with district and time (month) random effects

^* p < 0.1, ^** p < 0.05, ^*** p < 0.01

^aAlternative policy impact date suggested by the breakpoint function utilised

^bPolicy implementation assumed to be 9 months prior to delivery date

^cMarginal effects could not be estimated

^dIncludes doctor, midwife and MCH aide

^eSlightly lower but still significant result recorded where mother reported vaccinations also included

^fDeliveries excluded from November 2012 to allow for 1 year vaccination period