Using Respondent Driven Sampling to Estimate Lifetime Prevalence of Informal Sector Abortion Among Women of Reproductive Age in Soweto, South Africa

Caitlin Gerdts , Ibis Reproductive Health
Ruvani Jayaweera
Isabel Munoz, University of California, San Francisco
Relebohile Motana, Ibis Reproductive Health
Tshego Bessenaar, Ibis Reproductive Health
Paul Wesson, University of California, San Francisco

Existing data on the prevalence of informal sector abortion are fraught with bias; indirect methods rely on difficult-to-test assumptions, and direct methods suffer from underreporting. Respondent Driven Sampling (RDS) is a methodology that has been used to estimate prevalence of sensitive and illegal behaviors among hidden populations, but has never been employed for generating estimates for abortion. In April 2018, we launched a RDS study in Soweto, South Africa. Nine “seeds” completed an interview-administered questionnaire on their experience with abortion. Seeds recruited up to three eligible participants (women aged 15 - 49 living in Soweto) from their social network; recruits presented to study sites, completed the questionnaire, and recruited up to three participants. Data collection will continue through November 2018; as of August 2018, 439 women have participated. In preliminary estimates, 15.3% of participants reported ever terminating a pregnancy; 4.6% reported ever terminating a pregnancy in the informal sector.

See extended abstract

 Presented in Session 35. Abortion