STUDY QUESTION
What is the contribution of the underuse of modern methods (MM) of contraception to the annual undesired pregnancies in 35 low- and middle-income countries?
SUMMARY ANSWER
Fifteen million out of 16.7 million undesired pregnancies occurring annually in 35 countries could have been prevented with the optimal use of MM of contraception.
WHAT IS KNOWN ALREADY
Every year, 87 million women worldwide become pregnant unintentionally because of the underuse of MM of contraception.
STUDY DESIGN, SIZE, DURATION
Demographic and health surveys (DHS) of 35 countries, conducted between 2005 and 2012, were analysed.
PARTICIPANTS/MATERIALS, SETTING, METHODS
Contraceptive use of 12 874 unintentionally pregnant women was compared with 111 301 sexually active women who were neither pregnant nor desiring pregnancy.
MAIN RESULTS AND THE ROLE OF CHANCE
An average of 96% of 15- to 49-year-old eligible women took part in the survey. When adjusted for covariates and compared with the use of MM of contraception, the use of traditional methods was associated with a 2.7 [95% confidence interval (CI): 2.3–3.4] times increase in odds of an undesired pregnancy, while non-use of any method was associated with a 14.3 (95% CI, 12.3–16.7) times increase. This corresponded to an estimated 16.7 million undesired pregnancies occurring annually in the 35 countries, of which 15.0 million could have been prevented with the optimal use of MM of contraception (13.5 million women did not use MM whilst 1.5 million women utilized MM incorrectly). Women with the lowest educational attainment and wealth quintile were 8.6 (95% CI: 8.2–9.1) and 2.6 (95% CI: 2.4–2.9) times less likely to use contraceptives compared with those with the highest level of each, respectively. Of the 14 893 women who neither desired pregnancy nor used contraception, 5559 (37.3%) cited fear of side effects and health concerns as the reason for non-use, 3331 (22.4%) cited they or their partner's opposition to contraception or religious prohibition and 2620 (17.6%) underestimated the risk of pregnancy.
LIMITATIONS, REASONS FOR CAUTION
Despite the fact that DHS are considered high-quality studies, we should not underestimate the role played by recall bias for past pregnancies. Few women report a current pregnancy in the first trimester and undesired pregnancies at that time are probably prone to under-reporting. Some terminated pregnancies may not be included in the current pregnancy group. Furthermore, covariates measured at the time of the survey may not have reflected the same covariates at the time the currently pregnant women became pregnant.
WIDER IMPLICATIONS OF THE FINDINGS
Underuse of MM of contraception burdens especially the poor and the less educated. National strategies should address unfounded health concerns, fear of side effects, opposition and underestimated risk of pregnancy, which are major contributors to undesired pregnancies.