Although several studies have found links between spousal violence and birth outcomes, little is known about these associations in Sub-Saharan Africa. To help address this gap, researchers analyzed data from the 2004 Cameroon Demographic and Health Survey, a nationally representative study that, unlike other large African surveys, differentiated between voluntary fetal loss (abortion) and involuntary fetal loss (miscarriage, stillbirth). The analysis focused on the 2,562 female respondents aged 15–49 who completed a special survey module concerning spousal violence. Respondents reported whether they had ever experienced violence from their spouse, including physical violence (ranging from pushing and shoving to attacks with a weapon), emotional violence (verbal or physical public humiliation, or verbal threats to the woman or her family) and sexual violence (forced sex or sexual acts). Associations between violence and birth outcomes—categorized as early fetal loss (miscarriage), late fetal loss (stillbirth), or any fetal loss—were assessed using logistic regression models that controlled for maternal and spousal age, parity, marriage type (polygamous or monogamous), socioeconomic status, residence (urban or rural), maternal and spousal education, and religion.

On average, respondents were 30 years old; about a third had a secondary or higher education, and one in five were in a polygamous marriage. About half (51%) reported having experienced at least one type of spousal violence, most often physical (39%) or emotional (31%) violence; sexual violence was less common (15%).

Twenty-five percent of women had had at least one miscarriage or stillbirth; these fetal losses were more prevalent among women who had been exposed to spousal violence than among those who had not (29% vs. 21%), and more common among women who had experienced multiple types of violence than among those who had experienced just one (32% vs. 26%). In multivariate analyses, the odds of fetal loss were higher among women who had experienced spousal violence than among those who had not (odds ratio, 1.5); the association held for physical (1.5), emotional (1.6) and sexual (1.7) violence and was slightly higher among women who had experienced multiple types (1.7) than among those who had experienced only one type (1.4). Odds ratios for early fetal loss and for late fetal loss were generally similar to the odds for fetal losses overall.

The researchers estimated that for women who had experienced spousal violence, 33% of fetal losses would be averted if spousal violence were completely eliminated. Reducing spousal violence by 50% would prevent 17% of fetal losses. Among women who had experienced sexual violence or multiple forms of violence, the proportions of fetal losses prevented would be even greater: forty-seven percent if violence were eliminated, and 24% if violence were reduced by half.

The findings for recurrent fetal loss generally mirrored those for all episodes of fetal loss. Eight percent of women had had more than one episode of fetal loss; the odds of recurrent loss were elevated among women who had experienced any type of violence, physical violence or sexual violence (odds ratio, 1.5 for each), and were highest among women who had experienced emotional violence (1.7) or two or more types (1.6). Among women who had experienced some form of spousal violence, eliminating violence entirely would prevent about a third of recurrent fetal losses (33%), and reducing violence by half would prevent a sixth of the losses (17%). Among women experiencing sexual violence, interventions that eliminated all such abuse would prevent 47% of recurrent fetal losses, and reducing sexual violence by half would prevent 24% of losses.

The researchers acknowledge that it is not certain that spousal violence or its aftereffects actually caused fetal losses, as the study data were cross-sectional and did not establish whether the violence women reported preceded their miscarriages and stillbirths. Moreover, both violence and fetal losses may have been reported inaccurately by respondents. Nonetheless, the findings underscore the importance of reducing spousal violence in Cameroon and elsewhere in Sub-Saharan Africa. Such violence, the authors note, not only may affect birth outcomes through physical trauma, but also through psychological effects: Abused women may withdraw "from participation in public life" and thus not obtain adequate prenatal care. Overall, they contend, the results provide "a strong argument for routine prenatal screening for spousal abuse in Cameroon," especially given that Sub-Saharan Africa has "a high prevalence of spousal violence and the highest fetal death rate in the world."—P. Doskoch

REFERENCE

1. Alio AP, Nana PN and Salihu HM, Spousal violence and potentially preventable single and recurrent spontaneous fetal loss in an African setting; cross-sectional study, Lancet, 2009, 373(9660):318–324.