Ensuring Daughter Survival in Tamil Nadu, India

Abstract

The south Indian state of Tamil Nadu is a relatively recent addition to the list of Indian states to experience the phenomenon of 'missing girls'. Nonetheless, the government and NGOs in the state have been active for some time in collecting data to track gender differences in survival and in introducing interventions to prevent daughter elimination. Against this background, this article has two aims. First, it provides a temporal and spatial analysis of patterns of daughter deficits in Tamil Nadu over the period 1996 to 2003. Second, it undertakes an examination of the modus operandi, underlying assumptions, strengths and weaknesses of various interventions and assesses their effect on daughter elimination.

A scoping review of methods for assessment of sex differentials in early childhood mortality

“…The studies were published in a period between 1929 and 2020. Up to the 1980s, there were only thirteen publications [ 15 , 27 , 51 , 54 , 56 , 57 , 61 , 63 – 68 ], and the number increased to 21 in the 1980s [ 8 , 23 , 28 , 34 , 58 , 59 , 69 – 83 ], 46 in the 1990s [ 3 , 4 , 6 , 12 , 22 , 24 , 33 , 39 , 43 , 44 , 46 , 49 , 52 , 53 , 84 – 115 ], decreased to 30 in the 2000s [ 7 , 11 , 16 , 17 , 21 , 25 , 26 , 36 , 41 , 47 , 50 , 60 , 62 , 116 – 132 ], and reached 44 in the period between 2010 and 2020 [ 1 , 2 , 5 , 13 , 14 , 18 – 20 , 29 – 32 , 35 , 37 , 38 , 40 , 42 , 45 , 48 , 55 , 133 – 156 ].…”

“…Some authors reported on female-to-male sex ratios for mortality, but most presented results as male-to-female ratios, either as a simple ratio or multiplied by 100, with values of 1 and 100, respectively, indicating equality between boys and girls. Also, in some of them, the excess mortality of girls was estimated [ 2 , 14 , 18 – 20 , 42 , 93 , 102 , 109 , 132 , 153 , 154 ].…”

“…This heterogeneous set of studies relied upon a variety of data sources and analytical methods to define the expected differentials in male and female deaths. The first study was published in the late 1980s [ 77 ], five dated from the early 1990s [ 12 , 93 , 95 , 100 , 105 ], and the other eight from the 2010s [ 2 , 18 – 20 , 42 , 136 , 138 , 153 ]. A summary of the characteristics is presented in Table 1 .…”

“…Regarding the age groups under assessment, six studies reported on more than one category, usually IMR, CMR, and U5MR [ 2 , 12 , 95 , 100 , 105 , 138 ], five studies only on IMR [ 18 – 20 , 42 , 93 ], and three solely on U5MR [ 77 , 136 , 153 ].…”

“…Several reference populations were used as the standard against which authors assessed results from one or more countries. The reference populations include: life table models from multiple high-income geographies [ 12 , 95 , 105 , 138 ]; survey data from LMICs [ 18 , 77 , 136 ], official statistics gathered by the United Nations [ 93 , 153 ], estimates published by other authors [ 42 ], and multiple sources including census, surveys, vital registration and surveillance systems [ 2 ]. Three studies from India used internal comparison parameters, as the authors chose the Kerala state or a set of countries as standards of low gender bias [ 18 – 20 ].…”