Sex-selective Abortion in Rural Pakistan

Aims: The existence of sex-selective abortions has been documented in Asian countries including Pakistan. However, most of these studies have used indirect methods of estimation based on sex ratio at birth, which is known to give accurate estimates only in countries where complete registration of births is available, which is not the case in Pakistan. In the absence of birth registration, only direct estimation can provide the evidence and estimates of sex-selective abortion. Therefore, when a large survey on Gender-based violence was being planned, it was decided to add a component on sex-selective abortion also.

Study Design: Non- interventional Cross-sectional Study.

Place and Duration of Study: Six Districts of Pakistan: 2011- 2014.

Methodology: This study, was undertaken during 2011-14 in six rural districts; two from three provinces; Balochistan, Punjab and Sindh as part of a survey on domestic violence among 4,885 married women aged between 18-49 years. The fourth province, Khyber Pakhtunkhwa was not included because of political turbulence at that time.

The sample of 4,885 women was selected through multistage random sampling. Out of these 4,885 women, four thousands, six hundred and twenty (4620) were parous and the present analysis is based on responses of these 4,620 women.

Results: Out of 4,620 ever-pregnant women, 968 women (20.9%) reported ever having induced abortion. According to discussion with these 968 women, the reason for abortion in 338 women (34.9%) was presence of a female fetus. Some of them reported to have got ten the sex of the foetus confirmed by ultrasonography. The highest rate of sex-selective abortion was in Balochistan (62.5%). The corresponding figures were 19.6% for Sindh and 18.8% for Punjab. The prevalence of sex-selective abortion in Balochistan was significantly higher (P<0.0001) than the other two provinces.

Conclusion: Although the present study reveals a high prevalence of sex-selective abortions in Pakistan, yet larger studies based on both rural as well as urban areas encompassing quantitative as well as qualitative data are needed to further explore the phenomenon.