In Tanzania, Women Can Correctly Assess Whether They Can Use the Pill

Susan London

First published online:

Tanzanian women of reproductive age can accurately determine whether they are ineligible to use combination oral contraceptives because of medical reasons such as blood clots or diabetes, according to a cross-sectional study conducted in rural and peri-urban regions.1 Overall, the women’s own assessment of their eligibility or ineligibility, as aided by a poster depicting medically valid contraindications, agreed with the assessment of trained nurses in four out of five cases. Only about one in nine women who said they were eligible were found to be ineligible by nurses.

The study was conducted in 2010 in Tanzanian drug shops that were accredited to dispense combination oral contraceptives to women after assessing their eligibility. The investigators trained nurses to use the 2008 update of the World Health Organization’s Medical Eligibility Criteria for Contraceptive Use to screen women for medical contraindications. The nurses approached women who were visiting drug shops (regardless of the reason for their visit) in Tanzania’s Ruvuma and Morogoro regions and asked them to assess their eligibility to use oral contraceptives with the help of a poster. The poster had text and images depicting the World Health Organization contraindications; pregnancy was also included as a contraindication, given that it obviates the need for contraception. The nurses then assessed the women’s eligibility using a checklist of the same contraindications, and measured their blood pressure. Both women and nurses provided the reasons for their eligibility decisions. The investigators calculated the accuracy of the self-assessments, using the nurses’ assessments as the gold standard.

Of the 2,395 women approached by the nurses, 1,776 met the study’s eligibility criteria (they were aged 18–39 and literate) and 1,651 agreed to participate. On average, participating women were 28 years old and had two children. Most were married or cohabiting (71%) and had no more than a primary school education (72%). Fifty-eight percent had used oral contraceptives at some time.

From the self-assessments, 29% of the women concluded that they were ineligible to use oral contraceptives, a proportion similar to the 27% who were ineligible according to the nurse assessments. The overall agreement between self-assessments and nurse assessments was 81%. Only 8% of women said that they were eligible when in fact they were not, and 11% said that they were not eligible when in fact they were.

Women who were ineligible to use oral contraceptives correctly classified themselves 70% of the time (corresponding to the sensitivity of self-assessment), and women who were eligible to use the method correctly classified themselves 85% of the time (corresponding to the specificity of self-assessment); those who said they were eligible were more likely to be correct than those who said they were ineligible. In bivariate analyses, women had an elevated likelihood of correctly assessing their eligibility if they had at least a secondary education (odds ratio, 1.5), had ever been pregnant (1.5), had previously used oral contraceptives (1.4) or were currently using any method (1.4).

The most common medically valid reasons given for women’s ineligibility to use oral contraceptives were current or possible pregnancy, current breast-feeding of an infant younger than six months, severe headaches and hypertension. Each was cited as the reason for ineligibility in 1–5% of self-assessments and 3–7% of nurse assessments.

However, 14% of women and 3% of nurses gave medically invalid reasons for ineligibility, such as fear of side effects and partner disapproval of family planning. In an analysis restricted to women who reported only medically valid reasons for ineligibility, self-assessment had a sensitivity of 63% and a specificity of 97%. Moreover, some participants—notably those who were pregnant or breast-feeding—would not have been screening themselves for oral contraceptive use under real-world circumstances; when these women were excluded from the analysis, along with those classified as ineligible for medically invalid reasons, the proportion of women who were ineligible to use oral contraceptives was 13% according to self-assessment and 9% according to nurse assessment.

Nurses deemed only 3% of women to be ineligible for oral contraceptive use because of hypertension, despite the fact that 11% of women had blood pressure at or above the threshold for ineligibility proposed by the World Health Organization (140/90 mm Hg). However, the authors note that the health risks posed by pregnancy may exceed the risk of oral contraceptive use among women with hypertension.

Taken together, the study’s findings show that "poster-based self-screening is a good test," according to the investigators. They propose that women’s ability to accurately self-assess their contraindications to combination oral contraceptives may also apply to progestin-only contraceptives, such as injectables, which have fewer contraindications and potentially could be administered by trained staff in drug shops. "The present results support the case for over-the-counter sales of [combined oral contraceptives] with self-screening for contraindications by women in Tanzania," they conclude.—S. London


1. Chin-Quee D et al., Women’s ability to self-screen for contraindications to combined oral contraceptive pills in Tanzanian drug shops, International Journal of Gynecology and Obstetrics, 2013, 123(1):37–41.