How using Behavioural Science helps reduce doctors over-testing

Excessive medical tests carry financial, emotional and personal costs for patients, and they push up healthcare costs. Behavioural science helps reduce over-testing by doctors through peer comparison: when doctors see how their ordering compares with similar colleagues, they change their behaviour – the Hawthorne Effect at work. This article explains why doctors over-test, how peer comparison works, what studies from Australia and Japan show, and four steps to apply it.

Key takeaways

• Over-testing is partly a training habit that doctors carry into practice.
• Peer comparison works through professional identity, loss aversion and specific, easy-to-act-on feedback.
• Personalised letters cut imaging referrals by about 10% among Australia’s highest-volume GP referrers.

What Is Over-Testing, and Why Does It Matter?

Over-testing means ordering more medical tests than a patient needs – and more information isn’t always best for the patient.¹ Order enough tests and a doctor will inevitably find some meaningless abnormality, yet feel compelled to treat it, often leaving the patient worse off. Campaigns such as Choosing Wisely tackle the same problem with lists of recommendations.

Why Do Doctors Over-Test?

Medical educators have hypothesised that the problem lies in how doctors are trained.¹ There’s an assumption that the more procedures a doctor performs, the more confident they’ll be in independent practice, and a 2014 study in JAMA found that doctors trained in areas with more aggressive testing went on to practise the same way. There are now efforts to challenge this by focusing on the human cost of treatment.

How Does Peer Comparison Reduce Over-Testing?

Peer comparison taps into the Hawthorne Effect: people behave differently when they feel they’re being watched.

Professors at Johns Hopkins Hospital in Baltimore have been leveraging doctors’ competitive spirit with personalised reports comparing each doctor’s testing and treatment rates with their peers’. The results are still being studied.¹ The same pressure works on qualified doctors too: Dr Adam Cifu of the Pritzker School of Medicine in Chicago reined in his antibiotic prescribing simply by learning how he compared with colleagues.¹ Similar efforts have taken place in Australia – see our post on reducing GP over-prescription of antibiotics.

Three behavioural drivers explain why it works so reliably on doctors:

• Professional identity: most doctors see themselves as careful, competent clinicians, so a report showing they order more tests than peers is far more motivating than a generic reminder about guidelines.
• Loss aversion: people are more motivated to avoid falling behind peers than to become top performers, so ‘you are ordering more than 80% of similar doctors’ outperforms an aspirational benchmark. Learn more about loss aversion.
• Specificity and ease: feedback works when it is personal, timely and easy to act on – the doctor’s own number, how it compares, and a plausible next step.

Does Peer Comparison Work? Evidence From Australia and Japan

Yes. In both studies below, clinical guidelines didn’t change – what changed is that doctors could see themselves the way their peers saw them.

Case 1: Imaging referrals in Australia

• Context: in 2022, Australia’s health department targeted GPs in the top 20% nationally for musculoskeletal imaging referrals.
• Intervention: personalised letters comparing each doctor’s ordering with similar GPs in their region.
• Result: over 18 months, recipients reduced imaging referrals by roughly 10% relative to a control group, in a study of more than 3,800 physicians.²
• Lesson for practice: target the highest-volume orderers and compare them with genuinely similar peers.

Case 2: Antimicrobial prescribing in Japan

• Context: Kobe Children’s Primary Emergency Medical Center. This case concerns prescribing, but uses the same peer-comparison mechanism.
• Intervention: a monthly newsletter comparing each physician’s prescribing with a target and with colleagues – no penalties, just visibility.
• Result: unnecessary prescriptions of a commonly over-used antibiotic class fell by 67% within a year.³
• Lesson for practice: even a simple, regular feedback tool can shift clinical habits.

How to Reduce Over-Testing With Behavioural Science: 4 Steps

1. Identify the highest-volume orderers using ordering data, as Australia did.
2. Compare doctors with similar peers in the same setting or region.
3. Frame the comparison around falling behind to draw on loss aversion and professional identity.
4. Make it timely and regular, with a clear next step – like Kobe’s monthly newsletter.

What Are the Limitations of Peer Comparison?

• The effect can fade once nobody is looking, unless feedback is personal and repeated.
• Effects are moderate: around 10% in Australia – meaningful at scale, but not a cure-all.

Frequently Asked Questions

• Why do doctors order too many tests?

Partly because of training: doctors trained in aggressive-testing areas tend to keep practising that way.

• What is the Hawthorne Effect in healthcare?

The tendency of people, including doctors, to behave differently when they feel they’re being watched.

• Can the same approach work outside healthcare?

Yes – ‘the feeling of being watched’ can also drive behaviours like recycling or step challenges. See how social proof encourages environmentally friendly behaviours.

Next Steps: Turn Behavioural Insight Into Action

Visibility changes behaviour more reliably than more data does. DeltaMV’s Describe, Understand, Shape approach maps where a physician’s behaviour breaks down, then designs the low-friction nudge – making the better choice easy, social and emotionally resonant – that moves it. Whether the goal is reducing unnecessary testing, faster adoption of a new treatment, or shifting prescribing habits, the lever is rarely more information; it’s a better-designed moment of comparison.

Want to explore what a behavioural approach could do for your next quality-improvement initiative or brand plan? Get in touch with DeltaMV’s Behavioural Strategy team.

Sources

1. Gordon, M. (2019, April 19). Why do doctors overtreat? For many, it’s what they’re trained to do. NPR. https://www.npr.org/sections/health-shots/2019/04/19/715113208/why-do-doctors-overtreat-for-many-its-what-they-re-trained-to-do
2. O’Connor, D. A., Buchbinder, R., Glasziou, P., Maher, C. G., McCaffery, K. J., Schram, D., Maguire, B., Ma, R., Billot, L., Gorelik, A., Traeger, A. C., Albarqouni, L., Checketts, J., Vyas, P., & Clark, B. (2022). Effect of an individualized audit and feedback intervention on rates of musculoskeletal diagnostic imaging requests by Australian general practitioners: A randomized clinical trial. JAMA, 328(9), 850–860. https://doi.org/10.1001/jama.2022.14587
3. Shishido, A., Otake, S., Kimura, M., Tsuzuki, S., Fukuda, A., Ishida, A., Kasai, M., & Kusama, Y. (2021). Effects of a nudge-based antimicrobial stewardship program in a pediatric primary emergency medical center. European Journal of Pediatrics, 180(6), 1933–1940. https://doi.org/10.1007/s00431-021-03979-3