Center for the Study of Regional Development, School of Social Science, Jawaharlal Nehru University, New Delhi, India.
Centre for Global Health Research, Unity Health Toronto and Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada.
Lancet Glob Health. 2021 Jun;9(6):e813-e821. doi: 10.1016/S2214-109X(21)00094-2. Epub 2021 Apr 8.
Half of the world's missing female births occur in India, due to sex-selective abortion. It is unknown whether selective abortion of female fetuses has changed in recent years across different birth orders. We sought to document the trends in missing female births, particularly among second and third children, at national and state levels.
We examined birth histories from five nationally representative household surveys (National Family Health Surveys 1-4 and District Level Household Survey 2) to compute the conditional sex ratio (defined as the number of girls born per 1000 boys depending on previous birth sex) in India during 1981-2016. We estimated decadal variation in conditional sex ratio for 1987-96, 1997-2006, and 2007-16, and quantified trends in the numbers of missing female births for the states constituting >95% of India's population, as well as in 5-year intervals for each survey round. We used multivariate logistic regression to calculate the odds ratio of a second (or third) girl depending on the sex of the earlier child (or children), adjusting for education, wealth, religion, caste, and place of residence.
We assessed 2·1 million birth histories across the five surveys. Applying the conditional sex ratios from the surveys to national births, we found that 13·5 million female births were missing during the three decades of observation (1987-2016), on the basis of a natural sex ratio of 950 girls per 1000 boys. Missing female births increased from 3·5 million in 1987-96 to 5·5 million in 2007-16. Contrasting the conditional sex ratio from the first decade of observation (1987-96) to the last (2007-16) showed worsening for the whole of India and almost all states, among both birth orders. Punjab, Haryana, Gujarat, and Rajasthan had the most skewed sex ratios, comprising nearly a third of the national totals of missing second-born and third-born females at birth. From about 1986, the conditional sex ratio for second-order or third-order births after an earlier daughter or daughters diverged notably from that after an earlier son or sons. From 1981 to 2016, the sex ratio for second-born children after an earlier daughter decreased from 930 (99% CI 869-990) to 885 (859-912), and that for third-born children after two earlier daughters decreased from 968 (866-1069) to 788 (746-830). The probability of missing girls was mostly determined by earlier daughters, even after considering wealth quintile and education levels. The conditional sex ratio among the richest and most educated mothers was most distorted compared with lower wealth and education groups, and generally decreased with time, until a modest improvement in 2007-16.
In contrast to the substantial improvements in female child mortality in India, missing female births, driven by selective abortion of female fetuses, continues to increase across the states. Inclusion of a question on sex composition of births in the forthcoming census would provide local information on sex-selective abortion in each village and urban area of the country.
None.
For the Hindi translation of the abstract see Supplementary Materials section.
全球一半的女性出生人数缺失发生在印度，这是由于性别选择性堕胎造成的。目前尚不清楚近年来不同出生顺序的女性胎儿选择性堕胎是否发生了变化。我们旨在记录女性出生人数缺失的趋势，特别是在二胎和三胎中，在国家和邦两个层面上。
我们通过五项全国代表性家庭调查（国家家庭健康调查 1-4 项和地区家庭调查 2 项）来研究出生史，以计算印度在 1981 年至 2016 年期间的条件性别比（定义为每 1000 名男婴中出生的女婴数量，取决于前一次出生的性别）。我们估计了 1987-96 年、1997-2006 年和 2007-16 年期间的十年间条件性别比的变化，并量化了构成印度 95%以上人口的各邦的女性出生人数缺失的趋势，以及每轮调查的 5 年间隔。我们使用多元逻辑回归来计算第二（或第三）个女孩的几率，这取决于前一个孩子（或多个孩子）的性别，同时调整了教育、财富、宗教、种姓和居住地。
我们评估了五项调查中的 210 万次出生史。根据调查中的条件性别比应用于全国出生人数，我们发现，在 30 年的观察期间（1987-2016 年），有 1350 万女性出生人数缺失，基于自然性别比为每 1000 名男婴中有 950 名女婴。1987-96 年至 2007-16 年期间，女性出生人数缺失从 350 万增加到 550 万。对比第一个十年（1987-96 年）和最后一个十年（2007-16 年）的条件性别比，在两个出生顺序中，印度和几乎所有邦都出现了恶化。旁遮普邦、哈里亚纳邦、古吉拉特邦和拉贾斯坦邦的性别比例最为扭曲，占全国第二胎和第三胎出生的女性缺失人数的近三分之一。自 1986 年左右以来，第二胎或第三胎出生的条件性别比，在前一个女儿或女儿之后明显与前一个儿子或儿子之后不同。从 1981 年到 2016 年，在有一个女儿之后出生的第二个孩子的性别比例从 930（99%CI 869-990）下降到 885（859-912），而在有两个女儿之后出生的第三个孩子的性别比例从 968（866-1069）下降到 788（746-830）。即使考虑到财富五分位数和教育水平，女孩缺失的概率主要由前一个女儿决定。与较低的财富和教育群体相比，最富有和受教育程度最高的母亲的条件性别比最为扭曲，而且总体上随着时间的推移而下降，直到 2007-16 年略有改善。
与印度女性儿童死亡率的大幅提高形成对比的是，在各邦，由于女性胎儿的选择性堕胎，女性出生人数缺失仍在继续增加。在即将进行的人口普查中纳入关于出生性别构成的问题，将为该国每个村庄和城市地区提供关于性别选择性堕胎的当地信息。
在印度，尽管女性儿童死亡率有了显著改善，但由于性别选择性堕胎，女性出生人数缺失仍在继续增加，这种情况在各邦都存在。
