Influenza is a significant cause of morbidity and mortality around the world. Nudges are small changes to the environment or choice architecture that can be designed to significantly increase influenza vaccination rates.
Influenza is a significant cause of illness, hospitalization and mortality around the world. Vaccination is a low-cost, effective way to reduce the burden of influenza. Yet, improving vaccination rates has proven challenging. For example, despite numerous marketing campaigns and other initiatives, influenza vaccination rates in the United States have remained mostly unchanged for nearly a decade, with 60% of the population remaining unvaccinated each year1. In a more recent study, successful approaches have often been personnel intensive and expensive, hindering broader-scale deployment2.
Nudges are subtle changes to the design of the environment or choice architecture that are meant to influence behaviour in a predictable way, but without restricting choice3. They are often meant to remind, guide or motivate decisions. Nudges have been effective for improving outcomes in a variety of healthcare settings and can be delivered to patients, clinicians or both4. They are often easy to implement and low cost, making them a scalable approach to change behaviour.
Writing in Nature Human Behaviour, Yokum et al. report the findings of a large-scale field experiment using mailed letters to nudge elderly patients to obtain influenza vaccination5. They randomly assigned 228,000 Medicare beneficiaries to a control arm or to receive one of four interventions with different letters aimed at encouraging vaccination. These letters varied the sender, between the Director of the National Vaccine Program Office or the US Surgeon General, and whether or not the letter included one of two types of implementation prompts to select a date and time to get the vaccine. While they found no differences in vaccination between the different types of letters, they did find that sending a letter of any kind increased vaccination rates compared with the control. Effect sizes were modest ranging from 0.4 to 0.9 percentage points in the main analysis, and 0.8 to 1.4 percentage points in the analysis of patients with complete data. The interventions were not associated with any changes in hospitalization rates.
The authors should be commended for reviewing past evidence and using it to inform the design of a large-scale randomized trial. While effect sizes were small, if applied to the entire US Medicare population it could lead to an increase of 500,000 more vaccinations per year. Often policy initiatives implement interventions with little attention to prior evidence or proper evaluation. A more systematic approach such as the one by Yokum et al. allows us to both learn the results of the intervention and understand how to build on it for future work.
The findings from this trial expand our current understanding of how to use nudges to improve influenza vaccination rates. Figure 1 depicts a ladder of nudge interventions for patients. Nudges on the bottom of the ladder are more passive, serving mostly as reminders. Nudges in the middle of the ladder use information to motivate behaviour either through social influences or by prompting planning action. At the top of the ladder, nudges are more aggressive and focus on guiding choice by reducing friction such as increasing options (enabling) or changing the path of least resistance (changing default setting). These types of interventions can also be deployed to clinicians.
The interventions in this trial tested several nudge strategies directed to patients. First, lower on the ladder, the letters served as educational reminders by informing individuals of the risks of influenza and the benefits of vaccination. Prior systematic reviews have found mixed results with approaches that focus on reminding patients to obtain vaccination2, with lower impact among the elderly6. This may be due to decreased salience of the intervention in that many patients simply disregard or do not read the letters. Indeed, evidence indicates that patient outreach is often more effective when it is personal, such as verbal communication by telephone or during primary care visits2.
Second, the trial tested ways to frame information by changing the sender in hopes of altering social influences. However, it may have been the case that neither of the two figures displayed were recognizable to most individuals. While not feasible in this study, a picture of the patient’s primary care physician could have had a larger impact because of the closer social connection to the patient.
Third, the trial tested the use of implementation intentions by prompting individuals to select a time and date for vaccination, but did not specify a location. A previous study found implementation intentions increased vaccination rates by 4.2 percentage points, but that study was done with younger employees and vaccination clinics were offered in the worksite, which enabled choice7. Prompts to plan future actions are more likely to be effective if locations for vaccination are identified and more easily accessible.
Notably, the trial did not use nudges at the top of the ladder, which, while more aggressive, are also often more impactful8,9. A previous study tested ways to frame default options among employees at a university10. Those who had an appointment automatically scheduled for them to receive influenza vaccination (opt out) had a 12 percentage point increase in vaccination compared with employees that were reminded to make an appointment on their own (opt in). While appointment scheduling may not be feasible through a large-scale letter-based intervention, there are ways to simply frame information to present vaccination as the default with the ability to opt out that have been demonstrated to change behaviour in other contexts11.
Influenza vaccination rates have remained mostly unchanged for nearly a decade1. This study by Yokum et al. is one of the largest experiments of its kind and demonstrates how nudges can be deployed at scale to change patient behaviour. In future research, there are opportunities to deliver nudges to both patients and clinicians. As we continue testing the best way to design these approaches, we should also consider moving higher up on the nudge intervention ladder.
Influenza vaccination coverage. Centers for Disease Control and Prevention https://www.cdc.gov/flu/fluvaxview/index.htm (accessed 1 August 2018).
Lau, D. et al. Ann. Fam. Med. 10, 538–546 (2012).
Thaler, R. H. & Sunstein, C. R. Nudge: Improving Decisions about Health, Wealth, and Happiness (Penguin Books, New York, 2009).
Patel, M. S., Volpp, K. G. & Asch, D. A. N. Engl. J. Med. 378, 214–216 (2018).
Yokum, D., Lauffenburger, J. C., Ghazinouri, R. & Choudhry, N. K. Nat. Hum. Behav. https://doi.org/10.1038/s41562-018-0432-2 (2018).
Thomas, R. E., Russell, M. L. & Lorenzetti, D. L. Vaccine 28, 1684–1701 (2010).
Milkman, K. L., Beshears, J., Choi, J. J., Laibson, D. & Madrian, B. C. Proc. Natl Acad. Sci. USA 108, 10415–10420 (2011).
Halpern, S. D., Ubel, P. A. & Asch, D. A. N. Engl. J. Med. 357, 1340–1344 (2007).
Patel, M. S. et al. JAMA Intern. Med. 176, 847–848 (2016).
Chapman, G. B., Li, M., Colby, H. & Yoon, H. JAMA 304, 43–44 (2010).
Aysola, J. et al. Am. J. Health Promot. 32, 745–752 (2018).
Public Health: Ethical Issues Ch. 3 (Nuffield Council on Bioethics, London, 2007); http://nuffieldbioethics.org/report/public-health-2/policy-process-practice
M.S.P. is founder of Catalyst Health, a technology and behaviour change consulting firm, and is on the medical advisory board for Healthmine Services, Life.io and Holistic Industries.
About this article
Cite this article
Patel, M.S. Nudges for influenza vaccination. Nat Hum Behav 2, 720–721 (2018). https://doi.org/10.1038/s41562-018-0445-x