A great number of diseases result from unhealthy behaviors. Even individuals who are fully aware of the risks associated with certain behaviors and have the intention to make good choices struggle to do so. This can cause us to feel anxiety when making choices, and regret about not doing the right thing. The main reason is that many of the decisions we face on a daily basis are made automatically. Conversely, even deliberate decisions related to health behaviors are often sub-optimal. Thoughts such as “If today I’m feeling good, why continue taking the pills?” (which in behavioral slang should be called present bias), or “I remember the last time it hurt me a lot, so I won’t do the screening test again” (availability heuristic), bedevil our efforts to stay healthy. What is more, health care system sometimes creates complex environments for patients, in the sense that completing forms are very extensive; prescriptions don’t follow recommendations of medicine based on evidence, and even general practitioners (GP) can suffer from decision fatigue, decreasing the quality of service and attempting against patient’s health.
In the light of these realities, a new hope has emerged that might improve public health: applying choice architecture to medical centers. In this domain, where consumers have difficulty comprehending and doctors have difficulty communicating information for critical choices involving health, and where bad decisions can lead to enormous negative consequences, medical centers can serve a crucial role in enabling citizens to make better choices in health.
So, what exactly is choice architecture? Thaler, Sunstein and Balz, three well-known thinkers in behavioral science, define it as the “careful design of environments in which people make decisions.” In that regard, the key role for medical centers should be to simplify the decision processes of their patients or general practitioners in order to select the option that will make them better off.
Behavioral Science, Democratized
We make 35,000 decisions each day, often in environments that aren’t conducive to making sound choices.
At TDL, we work with organizations in the public and private sectors—from new startups, to governments, to established players like the Gates Foundation—to debias decision-making and create better outcomes for everyone.
Did I forget to take the pill? New solutions for problems of public health
Most people understand what constitutes a “good choice” in the context of health behaviors. To stay healthy, one should care for herself, follow her doctor’s recommendations, take treatments when prescribed, eat healthy, etc. The problem arises both from incomplete information as well as from our cognitive biases, which make adhering to these behaviors quite difficult. Moreover, we also suffer from bounded rationality. That is, even when we have all of the relevant information, we still fail to act optimally.
Fortunately, as we learn more about human behavior, these insights can be applied to the domain of healthcare. Employing simple nudges, medical centers can improve their patients’ outcomes by simplifying forms or sending text message reminders for treatments, among a number of similar, low-cost interventions. As these nudges improve individual health outcomes, they also improve social welfare as a whole. Thus, we hope that by using behavioral tools, patients will be nudged to make better health decisions and, therefore, increase their quality of life.
One of the first applications of behavioral insights to health policy was the case of the organ donation default. For those immersed in the behavioral literature, this example will be familiar. Essentially, the problem was that the rate of organ donation in many countries fell well short of the reported rate of desire to donate one’s organs: i.e., many people self-reported a desire to donate their organs, but did not enroll to be organ donors. The reason was that many forms on which people decide whether to donate were designed with an opt-in default (i.e., unless an action was taken to change it, one would not be an organ donor). By employing the insights of choice architecture, policymakers changed this to an opt-out default (i.e., you are an organ donor unless you actively decide not to be). This simple tweak in the design of a form had a tremendous effect on the number of organ donors — and as a result, not only saved lives, but also better aligned peoples’ actions with their own stated preferences.
Key takeaways for health centers
With that in mind, I present the following list of cost-effective innovations through which health centers can employ choice architecture to better enable patients to accomplish their health goals, and medical practitioners to deliver more efficient treatments:
- Reducing antibiotics using social norms. In 2014, Hallsworth and colleagues decided to run a trial to reduce over-prescription of antibiotics in England. To accomplish this, they wrote letters to general practitioners (GPs) in the top 20 percent of the prescribing-distribution, informing them of their rates of prescription relative to the lower 80 percent of GPs. They found an effect that was sustained for at least 6 months, substantially reducing antibiotic prescribing at a national scale. This low-cost program reduced unnecessary prescriptions of antibiotics, which contribute to antimicrobial resistance.
- In many cases, operating rooms implement critical processes of care by memory, increasing the likelihood of complications because of missing steps. That’s why, in 2009, Haynes and other researchers developed a nineteen-item checklist, applicable globally, to reduce the rate of preventable surgical complications. They tested the efficacy of this checklist in eight hospitals, and found that the rate of death for patients undergoing surgery fell from 1.6% to 0.8% by following the instruction of their checklist.
- Text message reminders on adherence to treatment. Low rates of adherence to artemisinin-based combination therapy (ACT) increase the risk of treatment failure and may lead to drug resistance. In 2011, Raifman and coauthors implemented a randomized controlled trial (RCT) that took place in Ghana, where the intervention consisted of text-messaging individuals receiving malaria treatment, with (i) a short reminder message “Please take your MALARIA drugs!” or (ii) a long reminder with encouragement component “Please take your MALARIA drugs! Even if you feel better you must take all the tablets to kill all the malaria”. The results showed that just sending the simple message significantly increased the odds of adherence by 1.45 times compared to the control group (no message sent).
- Implementation intention to increase influenza vaccination. A team of researchers let by Wharton’s Katherine Milkman evaluated a field experiment designed to measure the effect of prompts on changing health behaviors of employees in a large firm. In this case, the behavior of interest was whether employees received the free flu vaccination provided by the firm. Employees were randomly assigned to categories, and those in the treatment groups were mailed a reminder with (i) the date the employee planned to be vaccinated or (ii) the date and time the employee planned to be vaccinated. In both cases, the vaccination rates increased relative to the control group — and those employees who received the more specific prompt to write down both date and time saw the biggest increase in vaccination rates (4.2 percentage points higher).
The AI Governance Challenge
We all make complex decisions everyday, many of which impact our lives or those of our peers. While there are always ethical considerations when attempting to influence people’s behavior, in the domain of health care, these interventions can quite literally be the difference between life and death. In situations where both individual and societal welfare can be improved through simple, low-cost nudges, the question becomes: should we let patients and GPs suffer from their own inherent biases, or intervene and improve their health outcomes?
Recognizing this need is critical to improving health system. By taking advantage of these insights into how people behave, and translating it into policy interventions, we can take another step toward improving health and health care.
Behavioural Insights Team, 2016. Update Report 2015-2016, available at:
Hallsworth et al, 2016. Applying Behavioral Insights: Simple Ways to Improve Health Outcomes. Doha, Qatar: World Innovation Summit for Health.
Hallsworth et al, 2016. Provision of social norm feedback to high prescribers of antibiotics in general practice: a pragmatic national randomised controlled trial. The Lancet.
Haynes et al, 2009. A surgical safety checklist to reduce morbidity and mortality in a global population. New England Journal of Medicine.
Milkman et al, 2011. Using implementation intentions prompts to enhance influenza vaccination rates, Proceedings of the National Academy of Sciences (PNAS).
OECD, 2017. Behavioral Insights and Public Policy: Lessons from Around the World. Available at: https://dx.doi.org/10.1787/9789264270480-en
Raifman et al, 2014. The impact of text message reminders on adherence to antimalarial treatment in northern Ghana: a randomized trial. PloS one.
Thaler, R. H., & Sunstein, C. R., 2009. Nudge: Improving decisions about health, wealth, and happiness.
About the Author
Gabriella is a research assistant at Innovations for Poverty Action and analyst at PsychoLAWgy, conducting research regarding behavioral interventions on social problems in Latin America. She obtained a B.A in Economics in 2016 from Universidad del Pacifico, Peru. She is interested in applying behavioural economics to public policy problems related to health, education, and finance.