Baby killing in South Africa – uncovering the unthinkable

Organisation: Position: Deadline Date: Location:

South Africa has one of the world’s highest reported homicide rates involving babies. A first national study, published this week, finds the annual killing of hundreds of newborns and infants to be a “serious social and public health problem”. Abandoning babies indicates a failure of health services – there is a critical need for intervention to assist vulnerable mothers.

Researchers led by Naeemah Abrahams of the gender and health research unit at the South African Medical Research Council in Cape Town estimate the South African rates for neonaticide (killing a child within the first 28 days of life) to be 19.6 per 100,000 live births, and for infanticide (killing a child under one year) to be 27.7 per 100,000 live births.

These rates are surpassed only by an estimate for the Tanzanian city Dar es Salaam of 27.7 per 100,000 live births “and are much higher than those reported in developed settings”, the researchers write in a paper published on 26 April in the peer-reviewed international open access journal PLOS Medicine.

A review of 40 years of research on both infanticide and neonaticide showed the incidence in developed countries – such as the United States, United Kingdom and New Zealand – to range from 2.1 to 6.9 per 100,000 live births.

Mothers were the perpetrators in all the neonaticides and were the most common perpetrators overall, according to “Gender Differences in Homicide of Neonates, Infants, and Children under Five Years in South Africa: Results from the cross-sectional 2009 National Child Homicide Study”.

The other authors are Shanaaz Mathews from the Children’s Institute at the University of Cape Town, Lorna J Martin of the University of Cape Town and the Forensic Pathology Services of the Western Cape Government, and Carl Lombard, Nadine Nannan and Rachel Jewkes of the Medical Research Council in Cape Town


Homicide of children is a global problem. Many countries do not often study homicides of children, even though infant mortality is a key indicator of the development and health status of a country and the effectiveness of its child protection and health services.

Child homicide is often not considered a priority against competing public health challenges. In South Africa, for instance, homicide accounts for 4.5% of death among children under five years. HIV-Aids, neonatal causes and diarrheal disease are among the leading causes of death.

A 2014 report on violence against children by UNICEF concluded that the under-five-year age group is the second largest homicide group after 15-19 year olds. The pattern was also uncovered in this first South African National Child Homicide Study, with 40% of all child homicides among the under-five-year group.

“Studies mainly from high-income settings have shown that children under one year old have a four times greater risk of being killed than children in all other age groups, with the highest risk in the first day of life,” the researchers report.

Decreasing child mortality was a key Millennium Development Goal and is a priority in the new Sustainable Development Goals.

“But South Africa has been struggling to attain its goal; child mortality has remained above the global rate for the last 15 years,” says the study. “In 2009, the neonatal mortality rate was 14 per 1,000 live births, the infant mortality rate was 39 per 1,000 live births, and the under-five-year mortality rate was 56 per 1,000 live births.”

Despite all this, child homicide has received little research attention. A problem for research is poor identification, reporting and recording of these deaths, especially in poor settings. The homicide figures from the study are certainly underestimates. The only other South African child homicide study was a secondary analysis of children up to 14 years from four cities.

Some findings

Drawing on funding from the Open Society Foundation and the Medical Research Council, Abrahams and colleagues studied medical and legal data from a random sample of 38 medico-legal laboratories in urban and rural settings across South Africa for 2009. Child and perpetrator data were collected from police interviews.

They estimate that 454 children under the age of five years old were killed that year. Most deaths (53%) were among newborns aged zero to six days and 74% were infants under one year old. The great majority (85%) of newborns died after being abandoned.

Overall, there was no gender bias among child homicides, but 56% of homicides of boys took place in an urban setting compared to 44% of homicides of girls. The researchers believe this may reflect lesser propensity to kill boys in traditional areas in which son preference is strong.

Evidence of sexual assault was found in both sexes – but only in older children (one to four years) – “with a higher proportion of sexual violence found for girl homicides than for boy homicides,” the study says.

Mothers were identified or suspected as perpetrators in nearly two-thirds of homicides (71%) and in all of the neonaticides. “Few children were killed by strangers (2.5%), and one in five of the children were killed in the context of a family murder.” The child’s home and public spaces were the most common places where bodies were found.

The perpetrator was convicted in only 17% of homicides in children under five years. There were high levels of joblessness among perpetrators. Their age was known in less than half of cases, with the mean age being 28 years – or 25 years if the perpetrator was a mother and 23.5 years if the mother abandoned her newborn.

There is not information on motives for abandoning newborns, and so it is not known whether babies are left to die – or in the hope that they will be found alive. “The abandoned neonates were commonly discovered in garbage and refuse dumps and public spaces.”

Nearly two-thirds of neonaticides that involved abandonment (74%) occurred in Gauteng (43%) and KwaZulu-Natal (31%) where 40% of live births occur.

Some implications

“Our study shows that the first six days of life are the time point of highest risk for being killed in South Africa among children under five years, with the risk declining thereafter,” write Abrahams and colleagues.

“This suggests that there is a particularly high rate of unwanted pregnancy going to term, which is remarkable in a country that has one of the most liberal abortion laws in the world and reasonably good contraception services.”

“It points primarily to a failure of maternal and reproductive health services. Research among women who were denied abortions in Cape Town showed that despite the law, there are numerous barriers that women encounter when seeking a legal abortion in the public sector.” These include advanced gestational age and lack of trained staff for second trimester abortion.

Further: “Two reviews showed that mothers’ risk of perpetration was associated with economic stress, unemployment, younger age, limited education, social isolation, mental illness, substance abuse, and being victims of intimate partner violence.” Qualitative studies have found that women who conceal pregnancies and abandon babies just do not want them.


“The killing of children is the extreme part of a continuum of violence against children in South Africa,” says the study, adding that it is a serious social and public health problem and suggests failures of state services.

“Intervening is critical at all levels, including reproductive services as well as child protection services,” Abrahams and colleagues conclude. But while prevention should be the priority, the evidence base on prevention is limited.

“One of the ways to identify risks and patterns and develop early intervention systems is to employ a child death review system similar to those used for prevention of maternal mortality and perinatal deaths. Such reviews could assist in identifying failures in the child protection system and assist in identifying cause of death in suspected abuse cases,” the authors write.

On a positive note, a pilot child death review project has been initiated in response to the study. “The process of developing review teams of all those involved in the investigation of child homicides has seen services come together for the first time.”

Full article on the PLOS Medicine site

For more information contact:

Naeemah Abrahams

Senior Researcher, Gender & Health Research Unit

Medical Research Council of South Africa

+27 21 9380445


Gender Differences in Homicide of Neonates, Infants, and Children under Five Years in South Africa: Results from the cross-sectional 2009 National Child Homicide Study



Homicide of children is a global problem. The under-5-y age group is the second largest homicide age group after 15-19 y olds, but has received little research attention. Under- standing age and gender patterns is important for assisting with developing prevention inter- ventions. Here we present an age and gender analysis of homicides among children under 5 y in South Africa from a national study that included a focus on neonaticide and infanticide.

Methods and Findings

A retrospective national cross-sectional study was conducted using a random sample of 38 medico-legal laboratories operating in 2009 to identify homicides of children under 5 y. Child data were abstracted from the mortuary files and autopsy reports, and both child and perpetrator data data were collected from police interviews. We erred towards applying a conservative definition of homicide and excluded sudden infant death syndrome cases.

We estimated that 454 (95% CI 366, 541) children under the age of 5 y were killed in South Africa in 2009. More than half (53.2%; 95 CI 46.7%, 59.5%) were neonates (0╨28 d), and 74.4% (95% CI 69.3%, 78.9%) were infants (under 1 y), giving a neonaticide rate of 19.6 per 100,000 live births and an infanticide rate of 28.4 per 100,000 live births. The majority of the neonates died in the early neonatal period (0-6 d), and abandonment accounted for 84.9% (95% CI 81.5%, 87.8%) of all the neonates killed. Distinct age and gender patterns were found, with significantly fewer boy children killed in rural settings compared to urban settings (odds ratio 0.6; 95% CI 0.4, 0.9; p = 0.015). Abuse-related killings and evidence of sexual assault were more common among older girls than in all other age and gender groups.

Mothers were identified as the perpetrators in all of the neonaticides and were the most common perpetrators overall (71.0%; 95% CI 63.9%, 77.2%). Abandoned neonates were mainly term babies, with a mean gestational age of 38 wk. We did not have information on abandonment motives for all newborns and did not know if babies were abandoned with the intention that they would die or with the hope that they would be found alive. We therefore considered all abandoned babies as homicides.


Homicide of children is an extreme form or consequence of violence against children. This national study provides one of the first analyses of neonaticide and infanticide by age and gender and shows the failure of reproductive and mental health and social services to iden- tify and help vulnerable mothers. Multi-sectoral prevention strategies are needed.

Full article on the PLOS Medicine site



In an accompanying Perspective in PLOS Medicine, authors independent of the research team – Delan Devakumar and David Osrin of University College London in the United Kingdom – note that Abrahams and colleagues used the best methods available to them and produced estimates of child homicide that are “substantial”.

Devakumar and Osrin note that protecting vulnerable children is a priority and prevention should be strengthened by work with adolescent women to provide advice and support on sexual health, contraception and childbirth.

Child Homicide: A global public health concern

Delan Devakumar and David Osrin

Despite recent increases in child survival, some 5.9 million children still die each year, and reducing global childhood mortality remains a public health priority [1]. The greatest numbers of deaths are due to infections, intrapartum events, and preterm births, but reductions in all causes of mortality are needed to reach the Sustainable Development Goal target of 25 deaths for every 1,000 live-born infants.

In a research article in PLOS Medicine, Naeemah Abrahams and colleagues shed some light on the occurrence of child homicide [2]. Tragedy radiates from such events, backward in time to the pressure and emotional burden that might lead people, willfully or not, to end the life of a child, and forward to the effects on parents and families. Children who survive attempted homicide may go on to suffer long-term traumatic consequences [3].

Abrahams and colleagues examined data on homicides of children aged under five years in South Africa. They began by assessing records of unnatural deaths over a single year, 2009, in postmortem reports from a sample of medico-legal laboratories of different sizes and in urban and rural settings. They contacted investigating police officers and interviewed them to gather more information. Most of the deaths (74.4%) were of infants less than one year old (of whom 53.2% were neonates in their first 28 days of life). Mortality rates among neonates, infants, and children aged 1–4 years were 19.6, 28.4, and 1.0 per 100,000 live births, respectively. These figures are substantial: assuming an overall neonatal mortality rate of 12–14 per 1,000, the neonatal mortality estimate from homicide in the present study corresponds to ~1.5% of all neonatal deaths in South Africa in 2009.

Were the high rates of child homicide found in the study a product of South Africa’s high aggregate homicide rate (31 per 100,000), or were they unexpected [4]? Global homicide rates average 6.2 per 100,000, but there is large variation. Young people under the age of 20 years make up around a quarter of homicide victims [4]: 95,000 children were killed in 2012, a rate of eight per 100,000. Rates are particularly high in Latin America (12 per 100,000) and east and central Africa (10 per 100,000). The highest estimate is for El Salvador, at 27 per 100,000 [3]. In some countries, such as Venezuela, improvements in child health have been vitiated by child homicides [3].

We should view global estimates with caution as data are often incomplete, analysis requires assumptions, and misclassification is particularly likely with regard to homicide of young infants [3]. In general, homicides are likely to be underreported, particularly for neonates in countries that do not have complete coverage of birth registration and where deliveries take place outside institutions. Categorisation is always going to be elusive, but Abrahams and colleagues used the best methods they could, with an emphasis on conservative estimates. The researchers excluded cases for which no information was available and deaths ascribed to sudden infant death syndrome (of which up to 10% might represent homicide) [5], which could have produced an underestimate of the burden of child homicide. Conversely, they classified all cases of abandonment and subsequent death as homicide, which might have produced an overestimate.

The most common antecedent to death was abandonment of young infants, but there was little information on cause of death beyond this. Concealment of pregnancy is relatively common worldwide [6], and other studies have shown that suffocation and drowning are frequent methods of infanticide [7]. Causal inference is difficult if the child’s body is found in a partially decomposed state, and misclassification of stillbirths is possible. An assessment of abandoned fetuses and newborn infants in South Africa, by du Toit-Prinsloo and co-workers, found that 35% were decomposed. Amongst infants of greater than 26 weeks gestation, 28% (n = 31) were thought to have been born alive, but differentiation was not possible in 31% [8].

Abrahams and colleagues found no difference in child homicide rates by sex, although there was a decreased likelihood of male deaths in rural settings compared to urban settings. They rightly draw comparisons with south Asia and China, where both feticide and infanticide of girls have been a serious concern [9,10], but we should be cautious, given the modest size of Abrahams and colleagues’ study. In a related paper, Mathews and co-workers describe the epidemiology of child homicide in South Africa [11]. The pattern of homicide is similar for boys and girls, but changes with age. The homicide rate amongst males aged 15–17 years was five times that for females of the same age.

Abrahams and colleagues found—as have others—that mothers were the perpetrators in two-thirds of cases (maternal filicide). Studies from high-income countries suggest that the characteristics of mothers implicated in infanticide at birth and homicides of older infants are different. Women who commit neonaticide—the bulk of deaths—are more often young, unemployed or in school, and unmarried. Women implicated in the homicide of older infants tend to be older, and the homicide often occurs within a cycle of abuse [7]. The association of infanticide with maternal mental health is complex, and some studies support a link, while others do not [6]. Some women who commit infanticide are living with mental illness, including frank psychosis, but most infanticide does not seem to be associated with overt maternal mental illness [7].

What can the health community do? There are two general approaches: child protection and law enforcement, and primary prevention. Protecting vulnerable children is a priority, with an emphasis on supporting under-resourced and sometimes nonexistent child protection services, as is convicting perpetrators. In many cases, primary prevention of homicide through work with parents and families may be the best approach. Referring to deaths caused by parents, Resnick suggested a classification that included altruistic motives (to relieve suffering), acute psychosis, unwanted pregnancy, fatal consequences of child maltreatment, and revenge against another person, often a spouse [12].

Each of these categories has implications for the way we think about potential public health approaches. Some countries allow women to leave their infants anonymously in a safe place. The USA, for example, has introduced “safe havens” where infants can be abandoned legally. The effectiveness of such initiatives has yet to be determined, and it is not known whether mothers who might commit infanticide would call on them [13].

We agree with Abrahams and colleagues that more funds should go into maternity services, and also suggest that interventions need to be instigated before conception. As many births are unwanted, accessible and contextually appropriate family planning interventions are needed. Much work needs to be done with adolescent women to provide advice and support on sexual health, contraception, and childbirth. For women who present antenatally, a mental health assessment should be part of routine practice, with extra support for those in whom conditions are diagnosed or predicted [14]. Mortality data should be disaggregated and include homicide statistics, even if the numbers are small, so that we can move forward with a clearer picture of where interventions would yield the greatest benefit. We know a little, but not enough.

Child Homicide: A Global Public Health Concern

Receive Medical Brief's free weekly e-newsletter

Related Posts

Thank you for subscribing to MedicalBrief

MedicalBrief is Africa’s premier medical news and research weekly newsletter. MedicalBrief is published every Thursday and delivered free of charge by email to over 33 000 health professionals.

Please consider completing the form below. The information you supply is optional and will only be used to compile a demographic profile of our subscribers. Your personal details will never be shared with a third party.

Thank you for taking the time to complete the form.