Just last week, in The Bachelor Australia, Richie Strahan deployed the screaming dolls to test the parenting prowess of his bevy of beauties.

But a baby simulator program aimed at preventing teenage pregnancies may in fact do just the opposite, an Australian study shows.

The gurgling dolls may have inadvertently made teen motherhood too appealing, with many students doting on their electronic progeny and enjoying the attention that came with it, the researchers said.

Schoolgirls asked to care for the baby bots were more likely to become pregnant in their teens, found the first randomised controlled trial that tracked whether participants became pregnant.

The 13- to 15-year-old girls enrolled in the Virtual Infant Parenting Program (VIP) were twice as likely to give birth in their teens than girls who had no intervention (8 per cent versus 4 per cent), found the trial involving 1567 students from 57 schools in Western Australia.

When it came to pregnancies, 17 per cent of the schoolgirls given the dolls got pregnant versus 11 per cent among controls, according to the research paper published in the international journal The Lancet on Friday.

Girls who were given the baby bots were also more likely to have an abortion (9 per cent) in their teens compared with their counterparts (6 per cent), found the study that followed participants until the age of 20. The researchers linked the data collection to abortion clinics and hospital records.

Schoolgirls were given the baby simulators to care for over a weekend as part of the VIP program designed by the Telethon Kids Institute researchers.

The baby simulators – which cost roughly $1200 each – replicate the sleeping and feeding patterns of a baby. They cry when they need to be fed, burped, rocked or changed. They detect and report any mishandling, the length of crying time, the number of changes and general care.

Girls enrolled in the VIP program also attend education sessions delivered by nurses that cover pregnancy, good nutrition, the financial costs of having a baby, sexual health, contraception and respectful relationships.

They also watch a documentary featuring teenage mothers talking about their experiences.

"The risk of pregnancy is actually increased compared to girls who didn't take part in the intervention," said lead author Dr Sally Brinkman, a senior research fellow at the Telethon Kids Institute.

"We don't know what the actual mechanism is as to why the program failed. There are a few different theories out there," Dr Brinkman said.

The problem may have been that the programs were just too appealing to the schoolgirls.

"Anecdotally, a lot of the students really enjoyed the program … there was a lot of positivity around the program, so it didn't really work in putting the kids off," Dr Brinkman said.

The students got quite a lot of positive attention from family and even strangers on the street and they went about their weekends toting the electronic dolls.

"It was quite an intense experience for the students and perhaps that's why they really liked it," Dr Brinkman said

But the program did not present teen pregnancy in a positive light.

"What we tried to do was to present the reality of what it would be like to be a teen mother," Dr Brinkman said.

"We definitely were not saying you can't become a teenage mother. We didn't want to demonise that, but the intention was clearly behind the program to increase contraceptive use and if you were going to have a baby to do it in a healthy way, and part of doing it in a healthy way was to delay," she said.

This season's The Bachelor Australia pitted contestants against each other in a baby simulator competition last week. The women were forced to prove what good mothers they could be to the plastic automaton as they vied for Strahan's affection.

But the baby bots were designed for a very different reason.

Australia ranks sixth highest in teenage pregnancy rates among OECD countries.

It is understood between 1800 and 2000 Australian schools had the simulators, though they are mainly used for early childhood education programs.

The program was adapted from the US program "Baby Think It Over" by Realityworks. Similar programs are used in 89 countries, and are expanding into low- and middle-income countries.

Despite their popularity, there is little evidence of their effectiveness, the authors said.

"Evidence now suggests they do not have the desired long-term effect of reducing teenage pregnancy. These interventions are likely to be an ineffective use of public resources for pregnancy prevention," Dr Brinkman said.

She questioned whether such an individualised approach was needed considering the gradual decline in teenage pregnancies in Australia overall.

Realityworks chief executive Timmothy Boettcher said the baby simulators were extremely effective in parenting education programs and to deter teen pregnancy when used in conjunction with the company's curriculum.

"I am flattered that people are looking at new ways to use parts of our product, with things like different curricula for different purposes and measuring the outcomes, again flattering," Mr Boettcher said. "It, however, is not the same as using the product we sell in the way we provide it."

The simulators were currently used by more than 40,000 institutions worldwide, he said.

Dr Julie Quinlivan at the University of Notre Dame said one reason the intervention failed was that parenthood takes "two to tango".

"The intervention was directed at teenage girls and neglects the fathers," she wrote in an accompanying editorial in the Lancet.

Teenage girls idealised parenthood, and interventions needed to start in primary school, especially for children who faced "adversity in early childhood", she said.

"By the time a child reaches secondary education, the traumatised brain might have already evolved towards a desire for early childbearing to address subconscious evolutionary fears," she said.

Successful programs needed to invest early in vulnerable children from disadvantaged backgrounds.

"We cannot afford the quick fix, especially when it doesn't work," she said.

The study authors conceded that the control cohort tended to have higher education and higher socio-economic status, but after adjusting for these factors they found no difference in their findings.

Another potential limitation of the study was the low participation rate, with only 45 per cent of eligible students enrolled in the control group and 58 per cent in the intervention group. The study did not measure miscarriage rates.