Table 3. Female probability of dying (₃₅q₁₅) and the maternal mortality ratio (and their coefficients of variation) obtained from sibling histories data reported by women only and pooled data reported by women and men for the 1–5 calendar years preceding the survey¹.

Female mortality
Country	Year of survey	Based on reports of women				Based on reports of women & men				% Decrease in CI width
		35q15	95% confidence interval		Coefficient of variation	35q15	95% confidence interval		Coefficient of variation
Cameroon	2004	0.212	0.190	0.234	5.3	0.216	0.197	0.236	4.6	12
Congo	2005	0.252	0.216	0.288	7.3	0.242	0.211	0.273	6.5	15
Indonesia	2007	0.079	0.066	0.092	8.2	0.079	0.068	0.090	7.3	11
Malawi	1992	0.221	0.180	0.263	9.5	0.213	0.176	0.251	8.9	9
Nigeria	2008	0.167	0.153	0.180	4.1	0.169	0.157	0.181	3.5	14
Tanzania	1996	0.178	0.147	0.210	9.0	0.179	0.151	0.207	8.0	11
Uganda	1995	0.279	0.242	0.315	6.7	0.282	0.252	0.313	5.5	16
Zambia	2007	0.454	0.406	0.502	5.4	0.437	0.404	0.471	4.0	29
Zimbabwe	1994	0.145	0.123	0.168	7.9	0.155	0.133	0.178	7.5	2
Zimbabwe	2005–06	0.465	0.438	0.492	3.0	0.477	0.453	0.502	2.5	8
Maternal Mortality Ratio
Country	Year of survey	Based on reports of women				Based on reports of women & men 2				% Decrease in CI width
		MMRatio	95% confidence interval		Coefficient of variation	MMRatio	95% confidence interval		Coefficient of variation
Cameroon	2004	736	567	904	11.7	743	579	906	11.2	3
Congo	2005	944	542	1346	21.7	899	522	1275	21.4	6
Indonesia	2007	219	146	293	17.2	217	145	289	16.9	3
Malawi	1992	747	413	1081	22.8	713	398	1027	22.5	6
Nigeria	2008	709	598	819	8.0	715	608	822	7.6	3
Tanzania	1996	620	444	797	14.5	616	439	794	14.7	0
Uganda	1995	661	463	859	15.3	669	475	863	14.8	2
Zambia	2007	564	384	745	16.3	542	372	711	16.0	6
Zimbabwe	1994	435	275	594	18.7	419	266	572	18.7	4
Zimbabwe	2005–06	749	535	963	14.6	780	567	994	14.4	0

¹In order to pool data , the data obtained from women were weighted in a similar manner to data obtained from men to adjust for selection bias; however, women’s data had to be further adjusted for zero survival bias. The deaths and exposure reported by women were weighted by the inverse of the number of surviving siblings of the respondent to adjust for selection bias and further weighted by the inverse of the probability of being reported at all to adjust for zero survival bias (men provide information on women sibships with no surviving members).

²The PMDF (obtained from women) is multiplied by female mortality (obtained from pooled data) to generate the MMRatio.