host picture

  | Bookmark and Share

Disappearing Daughters - An expert responds

Professor Prabhat Jha
Professor Prabhat Jha is one of Canada's most authoritative voices on the issue of "disappearing daughters" or "missing girls" in his home country of India. In 2011, he co-authored an internationally acclaimed study published in the Lancet, showing that selective abortion of girls in India has been increasing, and may account for up to 12 million missing girls over the last three decades.

Professor Jha is the founding director of the Centre for Global Health Research at St. Michael's Hospital in Toronto.

Here is his response to the recent Sex Ratio study published in the Canadian Medical Association Journal:

"For accurate public reporting, it is important not to overstate the problem. IF (and that remains a big IF) selective abortion is happening in Ontario, then the numbers from the Joel Ray study would suggest it's less than 35 a year or less than 1% of all births among Indian-born mothers.

The reaction of some hospitals to ban information on ultrasound results might well do more harm than good, and might well be an overreaction.

By contrast, the scope of proven selective abortion in India is larger (in absolute terms perhaps 0.6 million/year), and the consequences much greater on society. There is already an appropriate debate for the global community (including the Indian diaspora) on the same (Canada's role in the global selective abortion debate, Toronto Star, Tue Jan 31 2012).

A few points:

The Joel Ray study provides circumstantial evidence that among women of Indian origin giving birth in Ontario, more boys than girls are born among women during their 3 or 4th delivery. There is no suggestion that more boys than girls are born in the first, second births, which is consistent with our research in India (Jha et al, Lancet 2006 and Lancet 2011).

There were about 30,000 (31,963 to be exact, based on the sum of the numbers in figure 1) births amongst Indian born women from 2002-7. Only 4000 of these (4098 to be exact) were higher order births (ie parity 2 or higher).

If sex ratios were "normal" at a ratio of 1.05, then 1999 girls would be expected to be born at parity 2 or higher. The actual number was 1754 (1385 +369). So, the absolute number of missing girls was about 245, which is less than 1% of 30,000 births (to be exact 245/31963=0.77%).

This implies:

  1. If selective abortion is responsible for the boy-girl imbalance for higher order births, then it is very small (less than 1% of Indian deliveries, or in absolute terms perhaps 35/year (ie 245/7).
  2. Important but subtle biases, such as higher migration of women who are about to give birth to a son might well explain this finding and suggest that selective abortion is NOT the explanation.
  3. The study, most crucially, does not have information on the gender of the previous birth and our studies in India have said that earlier girl births are the key factor to explain if subsequent pregnancies might undergo selective abortion. This "conditional" sex ratio is the most robust way to estimate missing girls.
  4. The role of chance to explain these findings cannot be excluded- the study is quite small (ie just 35 events a year), and the boy-girl gap seen in Korean born for 2nd children, but not for 3rd or higher children suggests that the play of chance could operate.
  5. We have an ongoing large national study covering Canada, the US, South Africa where more reliable estimation based on conditional sex ratios, which should be out shortly.
  6. Any policy or program implication of possible selective abortion should be commensurate with the size and consequence of the problem, which this study does not address. Dr. Rajendra Kale's editorial banning ultrasound results may well do more harm than good."

Professor Prabhat Jha, MD, DPhil
University of Toronto Chair in Disease Control
Director, Centre for Global Health Research, LKSKI, St. Michael's Hospital
Dalla Lana School of Public Health, University of Toronto