Full Transcript
HostOkay, so imagine this: You're trying to get people to take life-saving medication, consistently, every single day. The stakes couldn't be higher. And you think, "Right, money. That'll do it. A decent financial reward."
ExpertAnd you'd be right, to a point. Because when researchers offered a financial incentive in a real-world study, it *did* work. It boosted adherence significantly. But then they added something else, something incredibly simple and low-cost.
HostAnd that's where my jaw dropped. Because this tiny, almost insignificant addition, it didn't just *help* the incentive. It more than *doubled* its effectiveness. We're talking about a text message that cost virtually nothing, making the cash bonus profoundly more powerful.
ExpertExactly. It's a fundamental challenge to the idea that just throwing more money at a problem is the solution. This NBER paper is a brilliant demonstration that our brains often need a nudge, not just a bigger bonus, to get things done.
HostThis is "Incentives, Reminders, and Inattention: Evidence from a Field Experiment on HIV Medication Adherence," published in February 2024. And it takes us to Mozambique, a place where medication adherence is quite literally a matter of life and death.
ExpertThat’s right. The context here is HIV patients receiving antiretroviral therapy, or ART. And the critical thing about ART is that it requires near-perfect adherence – typically over 95% of doses – to be truly effective. If you don't take it consistently, you risk viral rebound and drug resistance.
HostAnd in sub-Saharan Africa, where this study took place, adherence can be a huge challenge. We're talking about a complex mix of poverty, logistical hurdles, and just the sheer cognitive burden of remembering to take a pill at the same time, every day, amidst everything else life throws at you.
ExpertSo, the researchers, led by Hang Yu and Dean Yang, set up a randomized controlled trial with 970 adult HIV patients in Beira, Mozambique. They tracked adherence electronically using smart pill bottles that recorded every time the bottle was opened. This wasn't self-reporting; this was hard data.
HostAnd their initial big question was classic economics: can money solve this problem? If you pay people to take their medication, will they do it more often?
ExpertPrecisely. They designed an intervention where patients could earn about $1.50 USD for each week they maintained perfect adherence, plus a monthly bonus of another $1.50 for perfect adherence throughout the month. That adds up to about $7.50 a month, which, in a country where the GDP per capita is around $500, is a significant sum. It was delivered via mobile money, which is a common and effective method there.
HostSo, my first thought, coming from a practical perspective, is "Great! Money on the table, high stakes. People will respond." And the study showed that it did have an impact.
ExpertIt absolutely did. The financial incentive *alone* had a statistically significant and meaningful impact. It increased the share of patient-weeks with perfect adherence by 10.6 percentage points over the control group, which had a baseline adherence of about 40%.
HostSo, a 10.6 percentage point lift is nothing to sneeze at. In a clinical setting, that could be the difference between viral suppression and viral rebound for a meaningful number of patients. Case closed, right? Pay people, get results.
ExpertWell, that's what many might conclude. And yes, a 10.6-point increase is solid. But the academic in me immediately asks: is this the *most* effective way to spend that money? Is the incentive hitting its maximum potential? The authors suspected there was a cognitive bottleneck limiting the power of the cash.
HostAnd this is where the real magic, and the real surprise, happens in the study. Because they didn't just test money versus no money. They tested money versus money *plus* a simple reminder.
ExpertThat's the pivot. They had a second group that received the exact same financial incentive scheme. The only difference was that this group also received a daily SMS reminder.
HostAnd the reminder itself was incredibly basic. It arrived around the same time each day and said, in Portuguese, "This is a reminder from the Atira-te+ project." It didn't mention HIV, didn't mention medication, didn't even mention the financial incentive. It was just a pure, low-friction memory aid.
ExpertRight. It was designed to be as neutral and unobtrusive as possible, simply a cue. And this is where the findings become truly staggering.
HostOkay, so let's set the stage: incentive alone, 10.6 percentage point increase in perfect adherence. Now, what happens when you add that simple, 10-word text message?
ExpertThe combination of the incentive and the reminder led to an increase in perfect adherence of 24.3 percentage points over the control group.
HostWait, 24.3 percent?
ExpertYes. Incentive only: 10.6 percentage points. Incentive *plus* reminder: 24.3 percentage points.
HostThat's... that's more than double! For the cost of a simple, automated text message, they more than doubled the effectiveness of the cash incentive. Let that sink in for a moment. The text message itself had a greater marginal impact on behavior than the initial $7.50 incentive.
ExpertAbsolutely. The marginal gain from adding the reminder was 13.7 percentage points, which is larger than the 10.6 percentage points from the incentive alone. It completely reframes how we think about the components of an intervention. We spent all this time calibrating the "right" dollar amount, and the real magic, the real catalyst, was in a simple text.
HostIt feels almost… counter-intuitive, doesn't it? That a non-specific text could unlock so much more from a financial reward.
ExpertIt's a beautiful demonstration of an interaction effect. The reminder acts as a catalyst for the incentive. Without the reminder, the incentive's power is blunted. This tells us the primary barrier wasn't a lack of motivation – the money was clearly desirable. The primary barriers, according to the paper, were cognitive.
HostAnd what exactly were those cognitive barriers? The paper breaks it down into two key frictions, right?
ExpertExactly. First, there's **inattention to daily adherence**. This is just simple, everyday forgetfulness. Life is busy. People get distracted. Even with the best intentions and a financial reward on the line, the thought of taking a pill can get crowded out by a million other concerns. The daily SMS reminder directly combats this by bringing the task "top of mind."
HostSo, it's not that they don't *want* the money, or don't *want* to be healthy. It's that they just... forgot. Amidst everything else.
ExpertPrecisely. And the second barrier is **imperfect memory of payment eligibility**. This is more subtle. Patients might remember they can get paid, but forget the *specifics* of the incentive scheme. They might forget that it requires *perfect* daily adherence to get the full weekly or monthly bonus.
HostAh, I see. So, if you miss a dose on Tuesday, you might think, "Oh well, I'll just be extra good for the rest of the week." But you've already lost that week's bonus. The daily reminder, even without mentioning the incentive, reinforces the overall program's salience.
ExpertThat's it. It’s like a gentle tap on the shoulder that keeps the entire incentive structure in the front of their minds, not just the act of pill-taking itself.
HostSo, the researchers didn't just observe this. They actually put a number on this "cognitive tax." They modeled it.
ExpertAnd this is where it gets really interesting for those of us who like to geek out on the mechanics. The authors developed a structural behavioral model. They essentially treat inattention and imperfect memory as "taxes" that reduce the perceived value of the financial incentive.
HostSo, you're saying if you offer someone $7.50, but because they're likely to forget to take their pill, and they might forget the exact rules for getting paid, the *expected value* of that $7.50 in their mind is much lower than $7.50?
ExpertExactly. It's as if a "cognitive tax" is levied on the reward before it even enters their mental calculus. The reminder effectively "removes" that tax, allowing the full potential of the incentive to be realized. And by comparing the effect sizes of the "incentive only" group and the "incentive + reminder" group, the model can estimate the size of this tax.
HostAnd the number they came up with was... what?
ExpertThe model suggests that these two cognitive frictions—forgetting to take the pill and forgetting the payment rules—reduce the incentive's effectiveness by a staggering 56%.
HostFifty-six percent! So, for every dollar you spend on an incentive in this context, 56 cents is essentially being wasted or burned by cognitive friction. The reminder is a tool that lets you reclaim that lost value. That's an incredibly high ROI intervention.
ExpertIt is. What's so elegant here is that they've put a number on it. It’s not just a qualitative story about forgetfulness. They're saying that the core problem is a 'leaky bucket' of attention. Pouring more water—more money—into that leaky bucket won't work beyond a certain point. You have to patch the holes first.
HostSo, a thought experiment: if the researchers had decided to double the incentive to $15 a month instead of adding a reminder, the model predicts this would have been far less effective. You hit a point of diminishing returns where a bigger reward can't overcome the fundamental problem of a person simply not remembering to act at the crucial moment.
ExpertExactly. And this connects to a broader idea in behavioral economics: **rational inattention**. It's a field pioneered by Nobel laureate Christopher Sims, and it treats attention as a scarce economic resource. We can't process all the information available to us, so we make rational, often unconscious, choices about what to pay attention to.
HostSo, it's not about people being lazy or irrational. It's about them being rationally selective with their limited cognitive bandwidth.
ExpertPrecisely. Acquiring and processing information has a cost. A person with HIV has to manage their health, work, family, and a thousand other daily inputs. The cognitive effort required to remember a pill at the exact same time every single day is a real cost. The SMS reminder dramatically lowers this 'information processing cost,' making it easier to allocate attention to the desired behavior. The study shows that instead of trying to make the *reward* for attention bigger, it's often more effective to make the *cost* of attention smaller.
HostAnd this insight, this idea that a nudge can multiply the power of an incentive, feels incredibly relevant to so many other domains. Like, for instance, the $8 billion dollar industry of corporate wellness programs.
ExpertOh, absolutely. This is a perfect real-world application. Most workplace wellness programs are built almost entirely on the logic of financial incentives. You offer a premium reduction, a bonus, or gift cards for completing health risk assessments, biometric screenings, or participating in fitness challenges.
HostAnd the data shows that incentives *do* boost participation, at least in the short term. Like, a 2019 KFF survey found that at large firms with an incentive for completing an HRA, 50% of workers participate, compared to just 31% at firms with no incentive. So, people show up for the money.
ExpertThey do. And the RAND Corporation found that incentives are effective for HRA completion, especially above a $50 threshold. But here's the kicker: the evidence that these programs lead to *sustained* health improvements or cost savings is highly debated, and often, quite weak.
HostI've seen that. A large-scale randomized controlled trial published in *JAMA* found that while wellness programs increased some healthy behaviors, they had no significant effect on clinical health measures like BMI or blood pressure, healthcare spending, or absenteeism after 18 months.
ExpertAnd another large RCT at the University of Illinois found that while incentives boosted screening participation, the program had no causal effects on medical spending, productivity, or self-reported health after two years. Even employers seem skeptical, with many finding financial incentives only "somewhat effective" or unsure of their effectiveness at all.
HostThis feels like a classic case of what we could call "nudge-washing." A company wants to be seen as caring about employee well-being. The easiest, most visible, and most quantifiable thing to do is throw money at the problem. You offer a $100 premium reduction for filling out a survey. It looks good on a report, HR can show a 40% participation rate, and everyone pats themselves on the back. But did it actually make anyone healthier? The evidence says... probably not.
ExpertIt's the "leaky bucket" problem again. The Mozambique study suggests the real issue might be inattention and habit formation. Does a one-time $100 reward help an employee remember to go to the gym on a rainy Tuesday in February? Does it make it easier to choose a salad when their colleagues are all getting pizza? Probably not.
HostSo, what if companies reallocated their incentive budgets based on this insight? Instead of a $100 bonus, what if they offered free apps that provide reminders for workouts or medication? What if they focused on reducing friction, like offering on-site fitness classes, pre-scheduled slots for health screenings, or making healthy food options the default in the cafeteria?
ExpertExactly. What if they focused on social support systems, which research shows can be a powerful motivator? The core question is a direct parallel to the NBER study: Is the barrier a lack of *motivation* (solved by money) or a lack of *attention* and an excess of *friction* (solved by reminders, defaults, and simplification)? If it's the latter, the money is often just going to waste.
HostAnd this tension between reminders and rewards plays out everywhere, particularly in the apps on our phones. It's a perfect illustration of different behavioral strategies in action. Take medication adherence apps, for example.
ExpertA great example. You have the **reminder-first approach**, like Medisafe or MyTherapy. Their core functionality is sophisticated reminder systems: customizable alerts, caregiver notifications, scheduling tools. Medisafe even lets you tailor it with fun voices like Darth Vader to deliver the reminder.
HostThe primary goal there is clearly to combat the "inattention tax." Make it as easy as possible to remember. And studies have shown these apps can significantly improve adherence; some research suggests Medisafe users have a 30% better adherence rate.
ExpertThen you have the **rewards-based approach**, like Mango Health. They combine reminders with a gamified rewards system. Users earn points for taking their medication on time, which can be redeemed for gift cards or donated to charity.
HostSo, Mango Health is assuming that the barrier is, at least in part, motivational, and tries to make adherence feel rewarding rather than a chore. A clinical trial found a 45% drop in missed doses using that app. So, both seem to work, but the Mozambique study really makes you question which component is doing the heavy lifting.
ExpertIt does. The study implies the reminder is the most critical component. The rewards may provide an extra layer of motivation, but without a reliable system to overcome forgetfulness, the rewards are likely less effective. The reminder is the foundation.
HostAnd we see this same strategic split in personal finance apps. How do you get people to save money?
ExpertRight. You have apps like Acorns or Digit, which use **automation to bypass inattention**. They're designed to make saving happen *without* requiring attention. Acorns automatically 'rounds up' purchases and invests the spare change. Digit uses an algorithm to analyze your spending and income, then automatically transfers small, supposedly unnoticeable amounts to savings.
HostThat's the ultimate solution to inattention—it removes the human from the loop entirely. It acknowledges that people are bad at remembering to save, so it automates the process.
ExpertThen you have apps like YNAB (You Need A Budget) or Mint, which are designed to make you *more* attentive. They force you to categorize every expense and confront your spending habits. They use alerts and visualizations to keep your financial goals top-of-mind. This strategy tries to train the user's "attention muscle."
HostBut that's a higher-friction approach, isn't it? It requires active engagement, which can lead to users abandoning them if they're not fully committed.
ExpertExactly. And finally, you have apps like Qapital, which try to **boost motivation** through goal-setting and visualization. They allow users to set specific, visual goals, like a picture of a vacation destination, and create rules to save towards them. This leans on research showing that visualizing a goal makes a person more likely to achieve it. It tries to increase the *motivation* to save by making the future reward more salient and emotionally resonant.
HostIt's fascinating how these apps embody different behavioral science principles. But we have to be careful, don't we? While these tools can be powerful, they can also create their own problems.
ExpertAbsolutely. The constant notifications from a budgeting app can create anxiety and a feeling of being policed. The gamification in a rewards-based app can lead to over-checking and a focus on points rather than the underlying health behavior. And the automation of an app like Acorns can be so seamless that users forget it's happening, leading to potential overdrafts if they aren't careful. It’s a double-edged sword.
HostSo, as we wrap up, what are the big takeaways listeners should carry with them from this discussion?
ExpertFirst, financial rewards are not a magic bullet. Their effectiveness is fundamentally limited by cognitive frictions like forgetfulness and inattention. The Mozambique study powerfully demonstrates that these frictions can cut an incentive's impact by more than half.
HostAnd second, simple, low-cost nudges that address these cognitive frictions can be incredibly powerful. They can act as multipliers, dramatically increasing the ROI of an incentive program. In the NBER study, a simple SMS reminder was more effective, in terms of marginal gain, than the initial cash incentive itself.
ExpertThird, and perhaps most critically, always diagnose the barrier before prescribing the solution. Is it a lack of motivation? Is it a problem of attention or memory? Or is it simply too much friction or hassle? Your intervention needs to match the underlying problem.
HostAnd finally, the future is integrated. The most powerful interventions will likely combine these approaches: a reminder to cut through the noise, an incentive to reinforce the action, and a simple design to reduce friction.
ExpertAbsolutely. It's about designing for how human brains actually work, not how we wish they would.
HostSo, thinking about all this, as we move towards more personalized digital health, could we design "smart" interventions that adapt to an individual's specific barriers? For example, an app that detects a user is forgetting doses and automatically switches from a rewards-based interface to a more prominent reminder-based one?
ExpertThat’s a fascinating thought. And it raises ethical questions too, doesn't it? When does a helpful reminder become an intrusive nag? When does a financial incentive become coercive, particularly in low-income populations where that $7.50 might be a substantial percentage of their monthly income?
HostAnd zooming out even further, could this 'inattention tax' framework be applied to other major societal challenges, like climate change – remembering to recycle or reduce energy consumption – or education, like completing complex financial aid forms? What other areas are we pouring money into, only to have it drained by the leaky bucket of human attention?
