Flu Shot: Yes or No? The Behavioural Science Behind the Flu Shot Nudge 

New research shows a small, deliberate change to how a decision is presented to physicians can meaningfully increase how many patients get a flu shot. At Penn Medicine primary care clinics, researchers built an electronic health record (EHR) prompt that required a clinician to make an active choice – ’yes’ or ’no’ – on placing a flu vaccine order for every eligible patient, rather than allowing the question to be silently skipped. 

Key takeaways 

• A forced ’yes/no’ EHR prompt raised flu vaccine ordering from 14.5% to 36.3% at Penn Medicine clinics – a 37.3% relative increase. 

• Active choice and opt-out defaults are different mechanisms: one removes the ability to skip a decision, the other changes what happens if no one acts. 

• The same logic works in Asia-Pacific, though the effect size in a Singapore trial was more modest than the US result. 

• The lever that matters most is rarely more information – it’s who has to act, and what happens by default when no one does. 

The Flu Shot Nudge: How One EHR Prompt Changed Physician Behaviour 

The results were substantial. Vaccine ordering rates at the intervention clinic rose from 14.5% to 36.3% over the study period – a 6.6 percentage-point adjusted increase compared with control clinics, and a 37.3% relative increase versus the same period the previous year⁴. Notably, more than 99.9% of vaccine orders placed through the prompt actually resulted in the patient being vaccinated – this wasn’t a nudge that changed intentions without changing behaviour, it changed what physicians and patients actually did. 

Active Choice vs Opt-Out Defaults: Two Different Nudges, Two Different Mechanisms 

It’s worth being precise about which behavioural mechanism is doing the work here, because ’nudge’ covers several genuinely different tools. The EHR prompt above is an example of active choice: removing the ability to passively skip a decision by forcing an explicit yes or no. Behavioural scientists Keller, Harlam, Loewenstein and Volpp found that this kind of enhanced active choice can be strikingly effective on its own – in one of their studies, 75% of participants chose a flu shot under an enhanced active-choice prompt, compared with 42% under a standard opt-in framing³. 

A related but distinct lever is the opt-out default. In a randomised study of university and hospital employees, Chapman and colleagues found that pre-scheduling a flu shot appointment by default – requiring employees to actively cancel if they didn’t want it – lifted vaccination rates to 45%, compared with 33% among employees who had to actively schedule their own appointment, roughly a 36% relative increase¹. The mechanism here isn’t a forced choice, it’s a changed default: almost everyone in the opt-out group ended up with an appointment on the books, and having an appointment – whether chosen or defaulted into – was what predicted actually getting vaccinated. 

Both approaches work, but for different reasons, and confusing them can lead to the wrong intervention being designed for the wrong problem: active choice interrupts an information gap or a habit of avoidance, while default scheduling removes a logistical or scheduling barrier altogether. 

Evidence From Asia-Pacific: A Modest but Real Effect in Singapore

The same logic has been tested closer to home. A cluster-randomised crossover trial across 22 private GP clinics in Singapore tested point-of-care informational nudges – prompts to physicians and patients about flu and pneumococcal vaccine eligibility for seniors – across more than 8,800 patient visits. Influenza vaccination rates rose from 4.8% to 5.9%, and pneumococcal vaccination rates rose from 3.7% to 5.7%². The effect is real but considerably more modest than the EHR active-choice results above – a reminder that a nudge’s size depends heavily on the specific decision point, patient population, and existing barriers it’s designed around, not just the general behavioural principle behind it. 

Designing a vaccine or adherence nudge for your own market? DeltaMV’s Healthcare Behavioural Strategy team can help you choose the mechanism that fits the actual barrier. 

What This Means for Choice Architecture Beyond the Clinic 

The underlying question is broader than flu shots. Every clinical or purchasing decision already has a choice architecture, whether anyone designed it deliberately or not – a form that can be skipped, a question that’s easy to leave blank, a default that quietly favours inaction. Hypothetically, choice architecture should have no effect on a perfectly rational decision-maker; behavioural science consistently shows that real patients, and real physicians, are anything but perfectly rational. 

The lesson from the flu shot research isn’t ’add a pop-up’ – it’s that changing who has to act, and what happens by default when no one acts, can matter more than the underlying clinical facts or messaging content. For any drug, device, or health behaviour that depends on a physician or patient making an active decision, the question worth asking is the same one raised by these studies: what is the current default, is anyone actually being asked to choose, and would making that choice explicit – or changing the default – close the gap between intention and action? 

How to Apply This: 3 Steps 

1. Map the current default. Identify what happens today when no one actively decides – is the question skipped, or is there already a default in place? 

2. Choose active choice or an opt-out default, deliberately. Active choice fits a genuine information gap or habit of avoidance; an opt-out default fits a logistical or scheduling barrier – the two aren’t interchangeable. DeltaMV’s piece on opt-in vs opt-out defaults looks at the same choice in a very different context. 

3. Build the decision into the existing workflow. The Penn Medicine result worked because the prompt sat inside the EHR system clinicians already used – not as a separate step they had to remember. 

What Are the Limitations of These Nudges? 

• Effect sizes vary considerably by setting: the Penn Medicine active-choice result was far larger than the Singapore informational-nudge result. 

• An opt-out default needs an existing scheduling or booking system to default people into – it doesn’t work as a standalone message. 

• None of these nudges change the underlying clinical facts; they change who has to act and what happens by default, which only helps where hesitancy, not genuine refusal, is the barrier. 

Frequently Asked Questions 

What is a flu shot nudge? 

A small, deliberate change to how the flu vaccination decision is presented – such as a forced active choice or an opt-out default – rather than a change to the clinical facts or messaging content. 

Does message tone matter as much as the nudge mechanism? 

The nudge mechanism tends to matter more for action at the point of decision, though tone still shapes how a message is received more broadly – see our piece on ’feel good’ messages vs scare tactics for how framing affects health campaigns beyond a single clinical decision. 

Which works better: active choice or an opt-out default? 

Neither is universally better – they solve different problems. Active choice works best against an information gap or avoidance habit; an opt-out default works best against a scheduling or logistical barrier. 

Designing the Right Nudge for Your Brand 

Getting this right requires more than borrowing a tactic that worked somewhere else – active choice, opt-out defaults, and informational prompts each solve a different problem, and applying the wrong one can waste a genuine opportunity to shift behaviour. 

DeltaMV’s Healthcare Behavioural Strategy practice helps healthcare and pharmaceutical brands diagnose exactly where a decision is being lost, and design the specific nudge, default, or choice architecture that fits the actual barrier. Get in touch with DeltaMV’s Healthcare Behavioural Strategy team. 

Sources 

1. Chapman, G. B., Li, M., Colby, H., & Yoon, H. (2010). Opting in vs opting out of influenza vaccination. JAMA, 304(1), 43–44. https://jamanetwork.com/journals/jama/fullarticle/186162 

2. Ho, H. J., Tan, Y. R., Cook, A. R., Koh, G., Tham, T. Y., Anwar, E., Chiang, G. S. H., Lwin, M. O., & Chen, M. I. (2019). Increasing influenza and pneumococcal vaccination uptake in seniors using point-of-care informational interventions in primary care in Singapore: A pragmatic, cluster-randomized crossover trial. American Journal of Public Health, 109(12), 1776–1783. https://pubmed.ncbi.nlm.nih.gov/31622142/ 

3. Keller, P. A., Harlam, B., Loewenstein, G., & Volpp, K. G. (2011). Enhanced active choice: A new method to motivate behavior change. Journal of Consumer Psychology, 21(4), 376–383. https://www.cmu.edu/dietrich/sds/docs/loewenstein/EnhancedActiveChoice.pdf 

4. Patel, M. S., Volpp, K. G., Small, D. S., Wynne, C., Zhu, J., Yang, L., Honeywell, S., Jr., & Day, S. C. (2017). Using active choice within the electronic health record to increase influenza vaccination rates. Journal of General Internal Medicine, 32(7), 790–795. https://link.springer.com/article/10.1007/s11606-017-4046-6