Teen pregnancies are dramatically dropping. But researchers aren’t sure why

When public health researcher Peter Barron and his team sat down to update their 2022 analysis, which had found birth rates among teenagers climbing year on year in every province, they expected a straightforward exercise.

Instead, the researchers were left scratching their heads. The rates of births by girls aged 10 to 19 declined every year, in every province, reversing the worrying trend their 2022 work showed. Dramatically.

The study, published in the South African Medical Journal earlier this month, found an overall decline of 16% between April 2021 and March 2025, with births by the youngest girls, aged 10 to 14, dropping by nearly 40%.

It should have been straight-up good news for the researchers. But Barron was baffled.

“In public health, when you start getting 5 to 10% drops, that makes your crap detector go on red alert,” he says. “You start thinking: Is the data correct? So we looked at it from a number of different angles and everything suggested it was. 

“We were trying to find out the reason — delving into the information, trying to see if there were changes in institutions or provinces and districts. But it was across the board.”

Barron went to some of the country’s leading demographic modellers to ask what could be causing it. But they couldn’t explain it either.

“It’s the elephant in the room,” says Barron, who says he nearly scrapped the study because there was nothing solid the researchers could pin to the dramatic decrease. 

But the numbers tell an important story on their own. One that affects everything from the number of children the education department needs to budget for in schools to socio-economic indicators, such as sexual and gender-based violence.

The researchers found that in 2024/25 one in every 24 girls aged 15 to 19 gave birth. Urban versus rural also showed interesting trends, with steeper declines in the more rural provinces compared with the more urban.

Termination of pregnancy rates were higher in more urban provinces. For example, in 2025, the rate for girls aged 15 to 19 in the Northern Cape was about 61 per 1 000, compared with about 38 per 1 000 in Gauteng.

But even with the declines, the numbers remain far too high, says Barron. World Bank data from 2024 showed South Africa’s adolescent pregnancy rate is more than five times the global high-income country average. 

While the rates are among the lowest in Africa, they are higher when compared with other middle-income countries, such as India, Vietnam and Argentina.

Tanya Pampalone spoke with Barron about what the numbers mean, why they matter and why the data is difficult to gather.

Why does the adolescent pregnancy rate matter so much as an indicator of what’s happening in a country?

The World Health Organisation uses it as one of the Sustainable Development Goal indicators, because it’s a good measure of a number of things — how empowered young girls are, rates of sexual violence, literacy, access to contraception, life skills education, parental support. A whole range of things society needs to get right.

Girls who become pregnant tend to drop out of school early, so their life choices become constrained; they haven’t got an education and they’re often not well-equipped to look after children. Then you get a generational carry-over. Children of adolescent girls also tend to do worse than children of women in their 20s.

The finding about the youngest girls — those aged 10 to 14 — dropping by nearly 40% was pleasing because by definition, almost all the pregnancies would constitute statutory rape.

What are the broader implications of a falling birth rate for South Africa?

It has both negative and positive implications for society — that’s why countries have censuses. You need to be able to project into the future. For South Africa, a lower birth rate is a good thing in many respects; the population has grown exponentially over the past 50 years and the greater the population, the more resources are required to sustain it. 

The backlog in housing gets decreased. In five years, when fewer children hit school age, you might find there are too many schools — but if you kept the budgets the same, you’d be able to decrease class sizes. There are real benefits.

The longer-term economic implications are more complex. Developed countries face serious problems when the economic base shrinks and there are fewer breadwinners to sustain an ageing population — unless you get migration from outside. That’s a very different question for South Africa than it is for Europe or the US.

Did the numbers align with other trends?

The overall birth rate for all women has also decreased over this period. And for women over 20, the rate of decrease has been faster than the decrease among adolescents. 

From a peak of just over a million deliveries around 2020 or 2021, the numbers drop year on year. Overall, it’s fallen by about 20%. I’ve got no plausible reason to explain this. 

Something like Covid could have flicked a switch — but we don’t know whether people were having less sex or spending more time outdoors. This is just speculation.

What about the differences between urban versus rural provinces?

In the rural provinces, the birth rate has dropped faster than in the urban provinces and that’s also unexplained.

Termination of pregnancy is a different story — that’s where access is everything. The rates in rural provinces have been variable and that’s much more easily explained. 

If you’re in Limpopo or Mpumalanga and you’ve only got three or four institutions doing terminations and one of those closes for whatever reason — sometimes nurses or doctors say they don’t want to perform them — then access reduces and people can’t get the terminations they need. They haven’t got the choice that’s available in urban areas.

But the birth rate decline is the big driver behind the drop in adolescent pregnancies, even accounting for the increase in terminations, which is a much smaller factor going in the opposite direction.

The data itself is complicated, with multiple databases and different time periods. Is that part of the problem?

The department of health reports from April 1 to March 31, to fit with the fiscal year. Then there’s data from the department of home affairs, which is largely on a calendar basis. 

Over time, these two things should equal out but at any one point, with different time periods, you’re going to get slightly different numbers.

If, on an annual basis, the departments of health, home affairs and education got together and had a quick meeting to talk about the numbers and publish them in a transparent, easy-to-understand manner, it would solve many of the problems. 

It wouldn’t solve the problem of adolescent pregnancy but it would at least give a coherent basis from which people can start talking about it.

How might US funding cuts to HIV and sexual health services affect the numbers?

The US wasn’t funding much in the way of routine family planning services. But there has been a meaningful decrease in condom supply. In rural areas, where the public sector is the main distributor, that matters enormously.

There’s something else worth flagging: lenacapavir — the new injectable HIV prevention drug being rolled out — will do nothing to reduce pregnancy. It prevents HIV infection but it’s not a condom, which prevents pregnancy, HIV and all the other STIs like chlamydia and gonorrhoea that are rising in South Africa. 

Condoms have been shown over and over again to be one of the most cost-effective methods of HIV and pregnancy prevention. But they’re not popular.

The hope is that when someone comes in to get their lenacapavir injection, someone in the health system asks them about contraception at the same time. That kind of integration would make a real difference.

Could we see the rate climb again?

It’s possible. The total number of births is decreasing but the rate for women aged 20 to 39 is decreasing faster than for adolescents. That means the adolescent share of all births has increased. 

What we need now is the 2026 data from the department of health, which should be coming out in the next month or two. That will tell us a great deal about where we’re headed.

This story was produced by the Bhekisisa Centre for Health Journalism. Sign up for the newsletter.