Skip Navigation


Global Projects


National Maternal Mortality Study in Afghanistan (Reproductive Age Mortality Survey)

Kabul, Afghanistan


In 2002, a reproductive age mortality survey (RAMOS I) was conducted by the MoPH, the US Centers for Disease Control and Prevention and UNICEF in four districts of Afghanistan. The districts were selected to reflect an urban to rural/remote continuum: urban - Kabul City, Kabul province; semi-rural - Alisheng, Laghman province; rural - Maiwand, Kandahar; and rural remote - Ragh, Badakshan (1). The study found that all indicators of mortality risk were among the highest recorded globally, increasing substantially with remoteness. Estimates of under-five mortality ranged from 109 in Kabul City to 323 in Ragh, Badakshan. The maternal mortality ratio (MMR) was 418 (235–602) in Kabul, 774 (433–1115) in Alisheng, 2,182 (1451–2913) in Maiwand, and 6,507 (5026–7988) in Ragh (1). Maternal deaths exceeded every other cause of death among women of reproductive age except in Kabul. Of the women who died, 75% of their infants also died and none had accessed a skilled attendant in the two remote areas – Ragh and Maiwand (1).

These study findings initiated a strong response from the Afghan MoPH and the international humanitarian aid community. Enhancement of maternal and child health services became a major priority area and the focus of nearly a decade of health strategies and policies. In an effort to identify changes in mortality rates in the same geographic areas over time the MoPH and stakeholders approved a repeat of the RAMOS I study using the same two-stage survey methodology, as well as qualitative data collection to assess contextual influences affecting maternal mortality decline.

Data collection tools for RAMOS II built upon those used in 2002 to ensure comparability but included additional questions to explore equity, care seeking, and other domains of interest thought to affect health outcomes. Security considerations affected data collection and quality in Alisheng, Laghman and quantitative data collection was suspended before fully complete in the interest of safety. In total, quantitative data collection was only fully completed in Kabul City, Kabul and Ragh, Badakshan, and the findings from these two districts are the focus of this report.

The qualitative component of the study, conducted in all four RAMOS I study districts, included both interviews and focus groups. All qualitative interviews and discussions used a semi-structured interview guide that was specific to level (national/community), position, and in the case of focus groups, sex, of the participant. Interviews with health officials and service providers were designed to collect information about how the availability and quality of services has changed since the RAMOS I study was conducted, the capacity to address obstetric complications at district and provincial health facilities, and the challenges faced in providing maternal health services at each of these levels.

In Kabul city, RAMOS II covered 10 villages, 8,963 households, and a total study population of 63,329 – 30% of whom were women of reproductive age (18,850). In the northeast province of Badakshan, a total of 67,359 individuals in 8,845 households were sampled across 259 villages in district of Ragh. Of this total population, an estimated 25% were women of reproductive age. While the ethnic distribution of the Ragh study population was almost entirely Tajik, Kabul city had a more ethnically diverse population, with 66% of residents being Tajik, 20% Pashtun, 10% Hazara, and 1% Pashai.

All estimates of maternal mortality show evidence of decline in both Kabul and Badakshan, although only one measure, the maternal mortality rate, reached statistical significance in Kabul City. In Kabul, the MMRatio declined from 418 deaths per 100,000 live births in 2002 to 166 in 2011, a decline of approximately 60.2%. The maternal mortality rate (MMRate) declined from 0.7 deaths per 1000 women of reproductive age to 0.2 deaths, a statistically significant decline of approximately 75%. The adult lifetime risk of maternal death, which measures the proportion of women who reach age 12 who will die before age 49 from maternal causes, declined from 1 in 42 in 2002 to 1 in 149 in 2012. The proportion of deaths to women of reproductive age did not change significantly. In Badakshan, the maternal mortality ratio, maternal mortality rate, and the lifetime risk of maternal death all declined by statically significant margins. The MMR declined by 89%, from 6507 deaths per 100,000 live births in 2002 to 713 deaths in 2012 while the MMRate declined by 87%. In 2002, approximately 1 in 3 adult women in Badakshan were estimated to die of maternal causes; by 2011 this proportion decreased to 1 in 19. The proportion of deaths due to maternal causes in Badakshan fell from 65% in 2002 to 42% in 2011.


  • Start Date: 
  • End Date: