Sex selection of foetuses in India has led to 7.1 million fewer girls than boys up to age six, a gender gap that has widened by more than a million in a decade, according to a study released Tuesday.
In Indian families in which the first child has been a girl, more and more parents with access to prenatal ultrasound testing are aborting a second female in the hope that a subsequent pregnancy will yield a boy, said the study, published in The Lancet.
The increasingly lopsided ratio of girls to boys is larger in wealthy households than poorer ones, the researchers reported.
Between 1980 and 2010, they estimate, four to 12 million girls were aborted because of their sex.
"Selective abortion of female foetuses, usually after a firstborn girl, has increased in India over the past few decades, and has contributed to a widening imbalance in the child sex ratio," they conclude.
The female shortfall for the zero-to-six age bracket was 6.0 million in 2001, and 4.2 million in 1991.
"Increases in selective abortion of girls are probably because of persistent son preference combined with decreases in fertility," the authors say.
The mean number of children per Indian woman fell from 3.8 in 1990 to 2.6 in 2008.
Selective abortion of female foetuses accounts for two to four percent of female pregnancies in India, roughly 300,000 to 600,000 per year out of 13.3 to 13.7 million carrying a girl in 2010, the study found.
From 2001 to 2011, the practice increased at a rate of 170 percent, slowing from 260 percent over the previous decade.
In the study, researchers led by Prabhat Jha of the Centre for Global Health at the University of Toronto, analysed census data from 2011 and earlier.
The also examined over 250,000 births from national surveys to calculate the difference in the girl-boy ratio for second births in families in which the first-born child had been a girl.
They found that this ratio fell from 906 girls per 1,000 boys in 1990 to 836 girls per 1,000 boys in 2005, an annual decline of a half of a percent.
Declines were much greater in mothers who had gone to school for at least ten years than in mothers with no education at all. The same trend held true for wealthier households compared to poorer ones.
If the first child was a boy, however, there was no drop in the girl-boy ratio for the second child, showing that families -- especially those better off and more educated -- are far more likely to abort girls if the firstborn is also female.
By contrast, "we did not yet see any clear evidence of selective abortion of firstborn female foetuses," the researchers said.
Unlike Beijing, New Delhi does not enforce the kind of "one-child policy" that led to the selective abortion of firstborn females in China.
The practice has left the country with 32 million more boys than girls, creating an imbalance that will endure for decades. In China, 94 percent of unmarried people aged 28 to 49 are male.
But first-pregnancy female abortions might increase in India too if fertility continues to drop, particularly in urban areas, the study warned.
A 1996 government regulation designed to prevent the use of ultrasound for prenatal sex determination is widely flouted, the researchers say, pointing out that few health providers have been charged or convicted.
"The financial incentive for physicians to undertake this illegal activity seems to be far greater than the penalties associated with breaking the law," S. V. Subramanian of the Harvard School of Public Health said in a commentary, also in The Lancet.