First published online:


Approximately 13 million infants worldwide were born before 37 weeks' gestation in 2005, according to the first estimation of the global incidence of preterm birth.1 The analysis, which was based on maternal morbidity and mortality data from 92 countries between 1997 and 2002 and national preterm birthrates between 2003 and 2007, estimated that 12.9 million preterm births occurred in 2005. Some 85% of all preterm births occurred in Africa (4.1 million) and Asia (6.9 million). Globally, 10% of all live births were preterm. Africa and North America (excluding Mexico) had the highest proportions of preterm births (12% and 11%, respectively) and Europe the lowest (6%). The authors speculate that in Africa, preterm birth is most often caused by intrauterine infection and lack of availability of medication, whereas in North America, it most frequently results from the older age of mothers and the elevated incidence of multiple pregnancies. The authors comment that "the use of modern technology allows survival of many preterm neonates in developed countries, but such care is not widely available in developing countries. …Thus, the development of strategies for improving access to effective care in developing countries must remain a top research and operational priority."

1. Beck S et al., The worldwide incidence of preterm birth: a systematic review of maternal mortality and morbidity, Bulletin of the World Health Organization, 2010, 88(1):31–38.


Expedited approval and use of generic antiretroviral medications within the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) has led to increased access to such medications in PEPFAR-funded countries and cost savings of more than $300 million between 2005 and 2008.1 According to an analysis of annual survey data from 16 countries funded by PEPFAR, annual spending on antiretroviral drugs almost doubled between 2005 and 2008, from $117 million to $202 million, with spending on generic medications increasing from 9% ($11 million) to 76% ($155 million) of the total drug expenditure per year. Procurement also increased substantially, from six million one-month antiretroviral packs to 22 million between 2005 and 2008. The proportion of the antiretroviral packs procured that were generic rose from 15% in 2005 to 89% in 2008. In 15 of the 16 PEPFAR-funded countries studied, more than 80% of the monthly packs procured in 2008 were generic, ranging from 81% in Uganda to 100% in Zimbabwe; the exception was South Africa, where only 25% of the packs procured that year were generic. The estimated yearly cost savings generated by using generic rather than proprietary antiretroviral drugs increased from $8.1 million in 2005 to $214.6 million in 2008, for an overall savings of more than $323.3 million during the four-year-period.

1. Holmes CB et al., Use of genetic antiretroviral agents and cost savings in PEPFAR treatment programs, Journal of the American Medical Association, 2010, 304(3):313–320.


A woman's height is inversely associated with her child's risk of dying before the age of five and with indicators of substandard growth and weight gain, according to an analysis of data from 109 Demographic and Health Surveys conducted between 1991 and 2008 in 54 low- to middle-income countries.1 Of the pooled sample of 2.7 million children aged 0–59 months, 312,553 (12%) died by the age of five; country-specific child mortality rates ranged from 3% in Jordan to 25% in Niger. In adjusted analyses, every one- centimeter increase in maternal height was associated with a decreased risk of under-five mortality (relative risk, 0.99). This association was stronger among the youngest children than among the oldest: The relative risk of death associated with each one-centimeter increase in maternal height was 0.98 for those younger than one month versus 0.99 for those 12–59 months old. Compared with the risk of child mortality among the tallest mothers (those whose height was at least 160 cm), the risk among other women increased as their height decreased, from 1.06 for those 155–159.9 cm in height to 1.40 for those shorter than 145 cm. In additional analyses, each one-centimeter increase in women's height was associated with a decreased risk of their children being underweight, stunted or wasted (0.97–0.99); compared with risk among the tallest mothers, other women's risk of their children being underweight or stunted increased for each shorter maternal height group (1.24–2.15). The authors comment that their findings suggest "the presence of an intergenerational transmission from mother's own nutrition, disease, and socioeconomic circumstances during her childhood to her offspring's health and mortality in their infancy and childhood."

1. Özaltin E, Hill K and Subramanian SV, Association of maternal stature with offspring mortality, underweight and stunting in low- to middle-income countries, Journal of the American Medical Association, 2010, 303(15):1507–1516.


Use of oral contraceptives is related to women's experience of changes in the intensity, frequency and timing of migraine headaches, according to a study of 493 sexually active 18–40-year-old current pill users conducted in Sao Paulo, Brazil, between August 2008 and January 2009.1 Of the 480 women who reported ever having experienced a headache, 17% had had at least five headaches classified by the researchers as a migraine (i.e., lasting 4–72 hours with two or more of the following characteristics: unilateral pain, rhythmic throbbing, greater pain with movement and moderate to severe intensity). Women in the migraine group were more likely than those in the nonmigraine group to report that their headache intensity or frequency worsened with their pill use (odds ratio, 3.0); however, women in the migraine group were also more likely to report that their headache intensity or frequency improved with pill use (3.9). In addition, the odds of experiencing headaches mostly around or during the contraceptive hormone–free interval were greater for migraine sufferers than for other women (2.1). In an analysis of specific pill formulations among women in the migraine group, the only significant association found was between decreased headache intensity or frequency and the combination of ethinylestradiol and drospirenone (34.1).

1. Machado RB et al., Epidemiological and clinical aspects of migraine in users of combined oral contraceptives, Contraception, 2010, 81(3):202–208.


The children of women who have experienced intimate partner violence are more likely than the children of other women to die within their first year of life, according to an analysis of National Family Health Survey data on 3,909 live births that occurred to 89,199 married 15–39-year-old women in four Indian states.1 The proportion of women who had ever experienced physical intimate partner violence ranged from 16% in Maharashtra to 31% in Tamil Nadu; the proportion who had experienced violence within the previous 12 months ranged from 9% in Maharashtra to 23% in Bihar. The infant mortality (death within 12 months of birth) rate varied by state, from 44 to 57 per 1,000 live births; the rate of perinatal mortality (death within one week of birth) was high (36–45) and contributed substantially to the infant mortality rate. In bivariate analyses, the infant mortality rate for women who had experienced more than one episode of violence in the previous year was greater than that for women who had experienced none (71 vs. 50); similar results were found for perinatal morality (55 vs. 40) and neonatal mortality (death within one month of birth; 53 vs. 34). There were no differences in mortality rates between women who had experienced one episode of violence and those who had experienced none. In multivariate analyses, all three mortality measures were positively associated with mothers' experience of more than one violent episode (hazard ratios, 1.6–1.9); no associations were found in analyses comparing women who had experienced one episode of violence and those who had experienced none. The authors suggest that "in addition to the health and human rights benefits for women themselves, the incorporation of violence prevention into current programmes may offer significant dividends in terms of improvements in child survival."

1. Koenig MA et al., Domestic violence and early childhood mortality in rural India: evidence from prospective data, International Journal of Epidemiology, 2010, 39(3):825–833.


Condom use at first and last sex and consistent condom use with steady and casual partners are more strongly associated with being motivated by habit rather than by conscious calculation or societal norms, according to the results of an online survey e-mailed to students of the University of Zagreb, Croatia, and posted on social-networking sites.1 Of the 1,145 sexually active individuals aged 18–65 who answered survey questions between April and May 2009 about their condom use history and their reasons for using condoms (measured by three motivation scales: calculative, habitual and norm-oriented), 27% had used condoms at both first and last sex; having used a condom at first sex was positively associated with condom use at last sex (odds ratio, 2.0). In multivariate analyses, a higher score on the habitual condom use scale (i.e., using condoms out of habit) was associated with greater odds of condom use at first and last sex (1.1), whereas a higher score on the calculative condom use scale (i.e., using condoms when there is a good reason) was associated with lower odds (0.9). Similarly, a higher use-out-of-habit score was positively associated with consistent condom use with casual partners in the last six months and with steady partners in the last month (1.2 and 1.3, respectively). Age, being in a relationship and having had more than the median number of lifetime partners were each negatively associated with habitual condom use. The authors comment that "To facilitate the routine use of condoms, educational and preventive programs need to focus on the development of habitual condom use."

1. Stulhofer A et al., Understanding the association between condom use at first and most recent sexual intercourse: An assessment of normative, calculative, and habitual explanations, Social Science & Medicine, 2010, 70(12):2080–2084.


Early initiation of antiretroviral therapy can reduce the risk of death among people living with HIV by 75%, according to a study of HIV-positive adults with CD4+ T-cell counts of 200–350/ mm3 conducted in Port au Prince, Haiti, from August 2005 to May 2009.1 The sample of 816 participants—half randomly assigned to the early treatment group, which received antiretroviral therapy within two weeks of study recruitment, and half to the standard treatment group, which received antiretroviral therapy when their CD4+ count fell below 200/mm3—were followed for an average of 21 months. During the study period, 29 participants died; 23 were in the standard group and six were in the early group (hazard ratio, 4.0). Seventeen of the deaths in the standard group were caused by infection, compared with only one in the early group. In addition, 36 cases of active tuberculosis occurred in the standard treatment group, compared with 18 in the early group (hazard ratio, 2.0). The authors comment that early initiation of antiretroviral therapy increases the estimated cost of treatment by $400 per person, and suggest that it would be cost-effective, given the 75% reduction in mortality risk and 50% reduction in the risk of active tuberculosis.

1. Severe P et al., Early versus standard antiretroviral therapy for HIV-infected adults in Haiti, New England Journal of Medicine, 2010, 363(3):257–265.