Background
Afghanistan has one of the world’s highest fertility rates and, related to this, an infant mortality rate far higher than its South Asian neighbors. Contraception enhances family spacing, improves women’s safety in child birth and, as a result, reduces infant and child mortality. Until recently, there has been a paucity of information on the comparative rates of contraceptive practices in the country and socioeconomic correlates of uptake. We aimed to elucidate the factors influencing the use of contraception in Afghanistan using recent, robust national data.

Methods
Using Afghanistan Mortality Survey (AMS) 2010 data, the distribution of Contraceptive Prevalence Rate (CPR) and correlates of contraceptive use among currently married women aged 15–49 years were explored. We initially summarised descriptive data on 25,743 married women and then derived predictors of the use of any form of contraception using a multiple logistic regression model.

Results
The prevalence of self-reported current use of any contraceptive method was 21.8% (95% CI: 20.4-23.4) at the national level though there was a wide variation in practice between provinces. Herat province in the West region had a highest contraceptive prevalence rate of 49.4% while Paktika in the Southeast region had the lowest CPR of 2%. Multiple logistic regression analysis showed that a family size of greater than 6 living children strongly predicted contraceptive use (AOR 7.4 (95% CI:6.1-9.0)). Other independent predictors included: secondary or high level of education (AOR 2.1 (95% CI: 1.8-2.5)) and being in the wealthiest stratum (OR 2.1 (95% CI 1.5-3.0)). Rural residence predicted a lower use of contraception (AOR, 0.72; 95% CI: 0.56-0.92).

Conclusion
Contraceptive uptake rate was low overall with wide inter provincial variation. Strengthening female education, targeting married women in rural area and women with no education may enhance the effectiveness of National Family planning program in Afghanistan.