Public service to get the medical evaluation treatment

Randomised control trials are set to be rolled out in parts of the public service to evaluate the effectiveness of programs and policies, similar to how drugs and other medical treatments are evaluated.

The idea is the brainchild of Fenner MP Dr Andrew Leigh, who has been a proponent of applying the trials to public policy since prior to his election to parliament.

The idea has been floating around government policy proposals since the Albanese government came to office, however Dr Leigh, who is also Assistant Minister for Employment, is now proposing that the trials be applied to the delivery of employment services, a highly contentious area with diametrically opposed positions on what works.

If applied successfully, the trials could upend how public programs are evaluated and delivered, with perhaps some surprising results.

The first recorded randomised trial was conducted by James Lind in 1747 to identify a treatment for scurvy. Twelve sailors in pairs were given different treatments at random, with the Scottish physician able to prove that oranges and lemons were a cure for scurvy.

In the centuries since, this method of testing and evaluation has been widely adopted in medicine, but only recently has the idea been applied to public policy.

Last year, the government established the Australian Centre for Evaluation to run these trials across the public service and reduce the reliance on external reviews.

Currently, the Centre is in the midst of a trial of its own, applying the randomised control trial to the delivery of online job services. Dr Leigh said the trial is still underway, but could provide a model for how these would operate in other areas.

"This will provide us with a stronger evidence base in an area where often people have been making decisions based on expert wisdom rather than strong evidence," he said.

In a speech to a gathering of not-for-profit employment providers this week, Dr Leigh said these trials could be conducted in a range of social policy areas.

"They will provide insights on what works from employment and skills to health and social policy."

In particular, given the audience, Dr Leigh honed in on employment services and the evaluation of proposed delivery models.

"Too often we invest in ad-hoc pilot programs without evaluating whether they truly deliver for people and communities."

The government is planning on overhauling the delivery of some employment services, with a greater role for the Australian Public Service. To evaluate whether this will be effective, it is running a pilot trial in the Playford local government area, north of Adelaide.

Dr Leigh said that pilots often lacked a rigorous comparison group, meaning what would have happened to a similar group of people who were not part of the pilot.

"We want a good comparison group that would be a counterfactual, to know what would have happened otherwise."

There are other concerns with running randomised controlled trials outside of the lab.

Beyond the practical concerns about separating out the mess of human behaviour into controlled groups, there are ethical concerns where one group is given access to services that are not provided to others.

Dr Leigh said any trials would undergo an official ethics review process, and said there was another ethical obligation to ensure that governments were delivering the best services possible.

"If you know for sure that a government program works, then sure it's unethical to take people out of it, but if you don't know whether a program works on not, some people argue that it's unethical not to rigorously test it," he said.

"We owe that to participants and we owe it to the taxpayers who are paying the bills."

The move to insource some employment services already has its supporters and critics. On the one hand, the Community and Public Sector Union, which has long argued for insourcing, has vociferously supported the proposal, which is also being trialled in Broome.

The peak body for employment services organisations on the other hand said the proposal would be exorbitantly expensive and there was no evidence governments could do a better job.

Dr Leigh said the trials, when conducted, could sometimes produce surprising results.

"One of my favourite examples is Scared Straight, a program which took delinquent youths and put them behind bars for a day. It was thought that you would scare delinquent youth into a path of honesty and righteousness for the rest of their lives," he said.

"It turned out that the day in jail actually increased the chance delinquent youth will end up committing offences.

"Ideology only takes you so far."