'KPIs' have little relevance in managing our health system
Advertisement

'KPIs' have little relevance in managing our health system

There is an old saying that for every complex problem there is a solution that is simple, neat and wrong.

The modern delusion that the creation of simple ''key performance indicators'' can solve complex problems like the management of the health system, the performance of our education system or the effectiveness of our police is a case in point.

While it is inexcusable for a public servant to deliberately falsify records, it is inexplicable why politicians think that simplistic KPIs are a useful tool for guaranteeing service, especially after 20 years of evidence to the contrary.

Before looking at the absurdity of a policy that evaluates a health system through collecting data on emergency room waiting times, it's important to have a quick look at historical failure of public sector KPIs.

Perhaps the most serious breach of faith, if not the law, in recent years was the revelation that the Victorian Police Force had fiddled the crime statistics in the lead up to the previous state election.

Advertisement

Then of course we learned of a small number of teachers helping students to cheat NAPLAN tests, presumably to help students help their teacher to look good in the eyes of an indicator-obsessed employer.

A Victorian Auditor-General's report tabled in 2010 found that only 30 per cent of the 322 performance indicators that were reviewed were relevant to departmental objectives and provided appropriate information such as trend or comparative data.

There is no doubt that the vast majority of public servants, or private contractors, do not engage in fraud and deceit. But there is also no doubt that the greater the reward for meeting KPI targets the greater the likelihood of such behaviour. Indeed, the ultimate irony of our faith in KPIs is that the only reason we think such indicators are effective is the economist's assumption that people are selfish and will place personal interest ahead of the collective interest.

Economists have a polite term for greed and selfishness; we usually call it ''rational''. Indeed, as a profession we go so far as to describe generosity, self-sacrifice and community mindedness as ''irrational''.

Having built a model of human behaviour based on greed and self-interest it makes sense to assume that the only way to motivate people is to offer them financial reward for achieving specified goals. Indeed, it follows that the more clearly defined the goal and the larger the reward, the more motivated people will become.

It is true that people can behave selfishly when they make some decisions. We have all seen other people push in front of us in a traffic jam (queue-jumping is so un-Australian!), but the fact that some people are selfish in some instances doesn't mean that they always will be, or that we will get the best outcomes by assuming they will.

Would you rather your doctor be motivated by financial reward to treat you quickly or the intrinsic reward of doing their job well? Do you prefer your child's teacher to be motivated by a bonus for improving the average marks of the class or by the personal satisfaction of helping each child in the way that suits that child best? And finally, would you prefer the nurse that cares for your elderly parent to be focused on a bonus for whipping through each room quickly or by the desire to provide care and attention to each and every patient?

Of course, when it comes to selling your house, you probably have no problem paying your real estate agent a commission that rises with the sale price of your house. Most ''rational'' people understand that the agent will probably work harder for you if you give them a financial stake in your success.

So what's the difference?

For most humans, but for a smaller percentage of economists, there is a clear difference between real estate agents and aged-care nurses and, while they might struggle to articulate why, it probably revolves around the simplicity of measuring ''performance''.

When you sell your house the main thing you are probably concerned with is the price, and the higher the price you get, the better you probably think your agent did their job. You might also hope that they don't break the law and are polite to the potential buyers but a successful agent knows that in order to complete the sale they need to do those things anyway.

But when you think about medical care you probably think more about the importance of compassion and attention to detail than to patients seen per hour.

The federal government has created a system in which state and territory governments are paid for achieving the KPI of reducing emergency room waiting times. No doubt people would prefer to wait for less time than more, but when simple indicators are used to evaluate some systems things go wrong, even if the data isn't manipulated.

When we create rewards for reducing waiting times and offer no rewards for preventing future disease we make the health system less efficient not more. When hospitals prioritise treating patients quickly rather than comprehensively they increase the chance that patients will be re-admitted in worse health and at more cost in the future.

Australia is one of the richest countries in the world and we are in the middle of a mining boom of unprecedented proportions. But rather than invest in world class health, education, aged-care and other services our politicians, state and federal, have tried to narrow the definition of quality to the things that they like to measure rather than the things citizens like to experience.

High-quality services require secure funding, passionate staff and widespread community support. While performance data can help to ensure services are well managed, it can also be used to fundamentally distort the way services are delivered while distracting the public from what is really going on.

Dr Richard Denniss is executive director of the Australia Institute, a Canberra-based think tank. www.tai.org.au

Most Viewed in Politics

Loading
Advertisement