Pregnancy intention status has little or no effect on medical supervision at delivery, child vaccination or adequacy of growth, once the impact of socioeconomic and demographic characteristics is accounted for; intendedness does appear to independently affect the odds of obtaining adequate prenatal care, however. According to an analysis of Demographic and Health Survey data from five developing countries, only in Peru was unwantedness at conception consistently associated with deficits in all four health indicators.1 Such associations were less consistent in the other four countries studied, where birth order appears to exert a stronger effect than intention status on maternal and child health outcomes.
The analysis is based on Demographic and Health Survey data from Bolivia (1998), Egypt (1995), Kenya (1998), Peru (1996) and the Philippines (1998). Women who had had a live birth in the previous five years (three years for Kenya) were asked whether, at the time they conceived, they had wanted to be pregnant, would have preferred to wait until later or did not want any more children; unintended conceptions ending in births are thus referred to as mistimed or unwanted births. The investigators used logistic regression to examine whether intendedness at conception influenced whether the woman received inadequate prenatal care (none before the sixth month of gestation) or gave birth outside of a medical institution or without professional supervision. They also investigated whether children whose conception was unintended suffered adverse health effects by assessing the odds of incomplete vaccination coverage (among children aged one or older) and inadequate growth (among all living children).
Unintended (unwanted plus mistimed) births were most common in Peru and least common in Egypt (58% and 29%, respectively, of all births). The proportion of all births that were unwanted ranged from 11% in Kenya to 37% in Peru, and the proportion that were mistimed varied from 10% in Egypt to 39% in Kenya. In every country, unwanted births became more frequent as birth order increased and mistimed births became less frequent as the interval between births lengthened.
In all five countries, both inadequate prenatal care and unsupervised deliveries were significantly more common for unwanted than for wanted births. Some 27-71% of women who had an unwanted birth had received inadequate prenatal care, compared with 21-65% of those with a wanted birth. In addition, 56-62% of unwanted births were unsupervised by a medical professional, compared with 40-55% of wanted births.
In all countries except Bolivia, children who had been unwanted at conception were significantly more likely not to have been vaccinated by age one. In the four other countries, 21-52% of children who had been unwanted at conception lacked full vaccination coverage, compared with 18-36% of wanted children. In contrast, inadequate growth was related to unwanted conceptions only in Bolivia and Peru, where 34% of children unwanted at conception were stunted, compared with 22-24% of wanted children.
In the analysis examining factors affecting the odds of receiving inadequate prenatal care, maternal education had a large, monotonic effect: The odds were 5-16 times as high among the least educated women as among the most educated women. Moreover, in all countries except Peru, birth order had a large independent and positive effect (i.e., the odds of inadequate care were significantly elevated for fifth- or higher-order births), and in Bolivia, Egypt and Peru, the odds of inadequate prenatal care were significantly higher in rural areas than in large cities (1.4-5.4). In all countries except Kenya, women living in the poorest households had significantly higher odds of inadequate prenatal care than those in the richest households (1.7-3.4).
Net of the effects of these variables, Peruvian and Philippino women whose pregnancy was unwanted had independently elevated odds of having received inadequate prenatal care (odds ratios, 1.4 and 1.2, respectively). Unwantedness had no significant effect on the odds of inadequate prenatal care in Bolivia or Kenya; however, it had a negative effect in Egypt (0.8). In addition, in Kenya, Peru and the Philippines, women with a mistimed birth had higher odds of inadequate prenatal care than those with a wanted birth (1.2-1.3).
Like inadequate prenatal care, unsupervised delivery was associated with higher birth order, lower maternal education, lower household wealth and rural residence. Once the effects of these factors were accounted for, Peruvian women whose pregnancy was unwanted had elevated odds of delivering without professional supervision (1.2); however, the odds for Egyptian women with an unwanted pregnancy were reduced (0.8).
The logistic regressions examining child health outcomes included additional controls for the age and sex of the child. The effects of specific socioeconomic characteristics on vaccination coverage and stunting varied widely by country. The effects of pregnancy intention status on vaccination were significant in Egypt, Kenya and Peru, however. For example, the adjusted odds of not having received the full set of vaccinations were significantly elevated among Kenyan and Peruvian children who had been unwanted at conception (1.6 and 1.2, respectively) and among Egyptian children whose conceptions had been mistimed (1.4). No independent association emerged between intendedness and vaccination coverage in Bolivia or the Philippines. Finally, the odds of stunting among children who were unwanted at conception were significantly elevated only in Peru (1.2). (No data were available on this outcome from the Philippines.)
Because unwantedness showed significant interactions with all explanatory variables, the authors stratified the women within each country by both educational level and area of residence. The results were broadly similar to those for the sample as whole.
The authors note that intendedness at conception is difficult to measure precisely. However, their findings of rising levels of unwantedness with birth order and of mistimed births with birth interval length "demonstrate that these concepts are understood by many women and ... merit serious analysis." According to the investigators, their key result is the lack of consistent associations between intendedness and three of the four outcomes studied (i.e., unsupervised delivery, incomplete vaccination and stunting). This inconsistency reflects the interrelatedness of intendedness and birth order, given that birth order appears to have an even "stronger and more pervasive influence" than intendedness.
The authors speculate that higher-order children in large families are at a health disadvantage compared with their first- and second-order siblings, and that large benefits for maternal and child health will accrue with the transition to smaller families. In the case of Peru in particular, where unwantedness was consistently associated with all four adverse health outcomes, improved contraceptive use "should lead to improvements in obstetric and child care, and in child growth."--L. Remez
1. Marston C and Cleland J, Do unintended pregnancies carried to term lead to adverse outcomes for mother and child? an assessment in five developing countries, Population Studies, 2003, 57(1):77-93.