Has there been a downward trend in Africa too or is the BBC.com headline in fact correct?
Estimate: 71% of pregnancies resulted in live births
Distinguishing between pregnancies and live births is an important first step to understanding pregnancy figures for adolescents (girls between 15 and 19 years old).
Because of stillbirths, miscarriages and abortions 71% of adolescent pregnancies in sub-Saharan Africa result in live births. This is according to a 2007 estimate of the US research organisation the Guttmacher Institute.
Nationally representative Demographic and Health Surveys (DHS) do gather country-specific adolescent pregnancy data. These household surveys record pregnancy with a first child at the time of the survey as well as prior childbearing, regardless of the outcome.
However, the “complex” analysis of the data that would be required to determine if there has been an increase throughout Africa has not been done to date. This Erica R. Nybro, senior adviser for communication at the DHS Program – funded by the U.S. Agency for International Development (USAID) – told Africa Check.
Pregnancy data ‘cannot be completely trusted’
Pregnancy surveys present particular challenges, among them underreporting.
Pregnancy data “cannot be completely trusted” or verified, said Dr Venkatraman Chandra-Mouli, a scientist in the department of reproductive health and research at the World Health Organization(WHO). He explained one of the challenges: “Most women find it very hard to speak about their abortions. In many places, it is both socially stigmatising and illegal.”
Another difficulty is the ability to gather data on miscarriages. Dr Kristin Bietsch, a demographer at the Population Reference Bureau whose research focus is fertility in Africa, told Africa Check that “cultural situations in some societies limit the discussion of pregnancy before the birth has taken place”.
She gave the example of a survey in East Africa in which only 50% of women who were seven months pregnant told enumerators this was the case when asked if anyone in the household was pregnant.
In sub-Saharan Africa, few studies have attempted to estimate adolescent pregnancies on the basis of data for live births, foetal losses and abortions, according to Adebayo Fayoyin, regional communications adviser for the United Nations Population Fund (UNFPA). Such attempts have “largely been based on limited data from small-scale country studies” and are not suited to tracking regional trends over time.
Birth rate declined from 139 per 1,000 teens to 109
Conversely, birth data is more readily available and appropriate for comparing trends over time on a regional level.
As a result, it is common for researchers to focus on birth data instead.
This data shows that sub-Saharan Africa’s adolescent birth rate has decreased: from an estimated 139 births per 1,000 adolescents between 1990 and 1995, to 109 per 1,000 for the period 2010 to 2015.
The adolescent birth rate for the whole of Africa is lower than the rate for sub-Saharan Africa – and it has also been on the decline: from 122 births per 1,000 between 1990 and 1995 to 98 for 2010 to 2015.
This is according to the United Nations economic and social affairs department’s World Fertility Patterns 2015, which bases rates on data from “censuses, surveys, vital and population registers, analytical reports and other sources for a given country”.
(Note: The report listed three African countries with a slight increase in their teen birth rates: Somalia, Zimbabwe and Lesotho. In a 2013 review of adolescent pregnancy by UNFPA, researchers noted that a country’s national birth rate may conceal increases in some areas, such as certain provinces or rural settings.)
In cases where empirical demographic information is limited or the available data is unreliable, “models and indirect measures of fertility estimation have also been used,” said Dr Petra Nahmias, population affairs officer in the UN economic and social affairs department.
She added: “Although you cannot make a statistical inference regarding the decline in [the] adolescent birth rate, we expect that this decline is true as it is consistent with changes in other related determinants such as age at first marriage and girls’ schooling.”
The increased enrolment of girls in schools counts among the reasons for the decline in the adolescent birth rate in the region, said Nancy Williamson, adjunct associate professor at the Gillings School of Global Public Health, University of North Carolina. “And there is some increased knowledge and use of contraception.”
However, at the rate of decline since 1995, one demographer estimated that it would take almost 50 years (from 2013) for the region to reach the current levels of births to teenagers in Europe.
Conclusion: Claim of rise in teen pregnancies in Africa not supported by available data
The headline of a recent BBC report on teenage motherhood in South Africa misleadingly implied that teenage pregnancy in Africa is on the rise. The report covered only one country and did not compare teenage pregnancy statistics over time.
That said, pregnancy data in Africa is hard to come by because of the inadequate reporting of miscarriages and abortions, among other reasons.
In the absence of comparable teen pregnancy data for sub-Saharan Africa, researchers opt for the use of birth data to determine trends over time.
The best available birth data for the region shows that there has been a modest decline: from an estimated 139 births per 1,000 adolescent girls between 1990 and 1995, to 109 per 1,000 for the period 2010 to 2015.
This fact-check was written by Liesl Pretorius, researcher for Africa Check, a non-partisan fact-checking organisation. View the original piece on their website here.