Why we're wasting money on medical tests and how behavioural insights can help

In 2013 and 2014, more than 314,000 CT scans of the lower back were ordered in Australia, most of which showed no abnormalities. In routine cases of low back pain, X-rays and CT scans provide no meaningful information to guide treatment, exposing patients to unnecessary radiation.

A number of factors have contributed to this, including increased consumer expectations, an ageing population, financial incentives (where doctors have a stake in imaging services) and “defensive medicine”, which is doctors protecting themselves against possible litigation arising from missing a diagnosis.

This is one of numerous areas of wasted health-care expenditure around the world. Studies in the US have reported that 20 to 25% of all healthcare delivered is either not needed, or harmful. The situation in Australia appears much the same. A conservative estimate of avoidable costs in Australia’s public hospital system is A$928 million.

We can reduce some of this waste by looking at why doctors continue to order these tests and use behavioural techniques to change the situation.

Why so much waste?

One of the drivers of this waste is increasing consumer demand for medical tests. New technologies and increased public awareness have led to increases in mass screening for breast, bowel and cervical cancer.

Popular media further fuels demand; publicity of Angelina Jolie’s preventative mastectomy in 2013 led to the “Angelina Jolie effect” – a two-fold increase in consultations for breast cancer genetic testing and risk-reduction surgery. While there is evidence to support screening in these cases, it has empowered consumers to request tests for a variety of other ailments, including X-rays and CT scans for routine low back pain.

Reducing healthcare waste relating to unnecessary tests has been a major priority for researchers, governments and health services for decades. Ironically, much of this effort has itself been wasted. Historical approaches to improving healthcare quality have revolved around the assumption that providing knowledge will solve the problem; if doctors are told X-rays and CT scans are not recommended in routine cases of low back pain, they will stop ordering them.

But the idea that knowledge leads to action is a flawed assumption. We know we should eat more vegetables and exercise more, but it doesn’t mean we do. In medicine, as in everyday life, there is a gap between what we know and what we do.

How to use behavioural insights to help change doctors’ behaviour

If it’s not just knowledge that drives human behaviour, how can we find out what does? The answer is deceptively simple: ask people why they do what they do.

Behavioural researchers have identified 14 domains that influence our behaviour. In addition to knowledge, some of these influences include social influences, the environmental context, our professional identity and our beliefs about our capabilities.

When a team of researchers applied this psychological framework to the problem of overuse of X-rays in routine low back pain, they uncovered new insights into this behaviour. Some GPs reported they lacked skills in communicating to patients these investigations are of little or no value.

This was addressed through role play: using a prepared script to simulate a patient demanding an X-ray and giving doctors a response script suggesting alternative approaches, such as advice about appropriate activities and pain management strategies.

An example of such a script is:

X-rays don’t really provide useful information that would change how we manage routine cases of back pain. They also expose you to radiation. Right now the best thing I can give you is some advice on how to manage your back pain. We can revisit the need for an X-ray or CT scan if more serious symptoms develop.

Studies have demonstrated positive impacts of such techniques in changing low back pain health-care practices. But behaviour change should not stop at doctors. It’s also important to create more widespread public awareness that some tests are unnecessary and potentially harmful. NPS MedicineWise, an independent, federal government-funded health organisation, developed a consumer resource outlining five questions to ask your doctor about tests.

Studying the ‘why’

The X-ray example shows that rather than continually producing and passively disseminating guidelines telling doctors what to do, it’s more worthwhile analysing why they do what they do.

Surprisingly, linking psychological theory to health-care improvement only began in earnest at the turn of the century. Alarmingly, over 15 years later, less than 10% of published quality-improvement studies explicitly report the use of such theory.

But this approach has demonstrated potential. For example, a recent review of 29 studies aiming to reduce overuse of antibiotics found education alone was not as effective as interventions that employed additional behavioural techniques such as “enablement” (making it easier to do the right thing) and “restriction” (using rules such as restricting prescriptions to prevent doing the wrong thing).

However, much more research is needed in this area. An estimated 75 trials testing new cures for diseases and injuries are published per day, equating to 319,000 since the year 2000. In comparison, roughly 7,000 studies have evaluated the effectiveness of the use of behavioural insights to make sure these cures are put into practice.

In other words, for every 45 trials designed to discover new cures, there is only one trial designed to test the use of behaviour change techniques to ensure these cures are applied to patients.

Without re-balancing this equation, there’s a risk of compounding the problem of waste in health care. Knowledge of what to do isn’t enough. We need to explore why doctors, patients and health-care professionals behave the way they do, and how we can influence their behaviour for the better. Only this can harness the full potential of medical research breakthroughs.

Peter Bragge receives funding from a variety of government and research granting organisations to conduct healthcare quality improvement research, all of which is paid to his employer, Monash University. He played no role in any of the research outlined in this article.