Home delivery in unhygienic environments is common among Afghan women; only one third of births are delivered at health facilities. Institutional delivery is central to reducing maternal mortality. The factors associated with place of delivery among women in Afghanistan were examined using the Afghanistan Mortality Survey 2010 (AMS 2010), which was open to researchers. The AMS 2010 data were collected through an interviewer-led questionnaire from 18,250 women. Odds ratio (OR) and 95% confidence interval (CI) of non-institutional delivery were estimated by logistic regression analysis. When age at survey, education, parity, residency, antenatal care frequency, remoteness, wealth and regions were adjusted, the OR of non-institutional delivery was 8.37 (95% CI, 7.47-9.39) for no antenatal care relative to four or more antenatal care visits, 4.07 (95% CI, 3.45-4.80) for poorest household relative to women from richest household, 2.02 (95% CI, 1.43-2.84) for no education relative to higher education, 1.78 (95% CI, 1.52-2.09) for six or more deliveries relative to one delivery, and 1.50 (95% CI, 1.36-1.67) for rural relative to urban residency. Since antenatal care was strongly associated with non-institutional delivery after adjustment of the other factors, antenatal care service may promote institutional deliveries, which can reduce maternal mortality ratio in Afghanistan.