When getting healthier means your family gets healthier, too

There’s a concept in health economics that can be simplified to ​“you are who you hang out with.” The more technical term in the literature is ​“peer effects” or ​“network effects,” but it all boils down to the same thing. We’re influenced by the health behaviors practiced by our friends, our families, even our colleagues. These effects have been established across different age groups, different lifestyle choices – both ​“good” and ​“bad” ones – and different geographic regions. It’s a well-accepted concept. 

With network effect in mind, think about how much harder it can be to change your behavior. You’re not only trying to get out of a personal rut or pattern, but you’ve got the inertia of those around you, too, whether you’re trying to eat healthier, drink less, quit smoking, whatever. Of course, some medical interventions can help change behavior and sometimes can have interesting impacts on our networks. 

And so we were happy to see a recent New York Times article about how GLP‑1 use by one member of a family can lead to a ​“surprising side effect” for other family members: a better relationship with food, increased physical activity, and sometimes even weight loss. 

This is certainly newsworthy. But I personally wasn’t particularly surprised when a colleague shared the NYT piece – as an economist I am familiar with the peer-effect literature. I did see it as an excuse to reflect on the nuances of what could also be seen as family or community spillover of the benefits of certain medicines (or ​“externalities,” if we want to get technical about it), and as an industry, our inability or reluctance to try to measure them. 

If you’re a regular Rapport reader you know that we’ve been huge proponents of generalized cost effectiveness analysis (GCEA), the practice of adding petals to the so-called ​“Value Flower” in a kind of reverse ​“he loves me, he loves me not” exercise that attempts to show a fuller picture of the societal value of a medicine. (Try our GCEA Calculator!)

GCEA encourages us to consider family and caregiver spillover. But we mostly think about spillovers as alleviating a burden – freeing up a caregiver’s time and effort so they can resume their life, boosting productivity, spending less money caring for a sick relative, lowering rates of depression. We need to also consider that some treatments – like GLP-1s – may have other positive spillovers for network health that are measurable, too.

The positive spillover from GLP‑1 treatments also made us wonder about other disease areas and treatments where we might observe similar network effects. We can surmise that this is more likely to occur in cases where we observe more ​“health-seeking” behaviors. For example, treatments that are likely to help individuals overcome nicotine dependence, substance abuse, or engage in early and routine cancer screening in a family with a history of cancer diagnosis.

Understanding the relevance and scale of these positive impacts in different disease areas will help us design more effective interventions and improve health and wellbeing across social circles. However, in our current approaches to understanding the clinical and economic value of medicines, family spillover is inconsistently measured and rarely incorporated into economic evaluations. In fact, it is largely missing from formalized data collection efforts such as clinical trials or real-world data generation efforts (e.g., patient registries). For example, from a quick web-based search, we were not able to find any evidence that innovators like Novo Nordisk or Eli Lilly measured this kind of positive spillover in their trials or value demonstration efforts for GLP-1s. 

This gap calls for a more consistent approach to measuring family spillover and incorporating it into decision-making processes over the product lifecycle. A recent paper by Leech and colleagues summarized the latest methodological developments and made some concrete suggestions on how we could do better: incorporating these efforts into existing data collection efforts and processes, eliciting expert opinions, and further developing and testing novel methods and algorithms. The SHEERS Task Force also developed 11 consensus recommendations and outlined 12 avenues to advance for future research. 

As GCEA typically focuses on the patient population a treatment is intended for, we might also need to consider broader population-level/economy-wide models that allow us to more comprehensively capture the value of treatments to different stakeholders and in different sectors. Computable general equilibrium models in the macroeconomic literature offer an alternative. 

So where does this leave us? 

If medicines (or other types of healthcare interventions) can improve not only the lives of the people who take them but also the health trajectories of everyone gathered around their dinner tables, then ignoring those ripple effects means we are systematically undervaluing innovation. That should matter to public and private payors deliberating on coverage and reimbursement decisions, regulators who calibrate evidence requirements, HTA bodies that seek to approximate societal value of innovative treatments, and to investors/​innovators who decide which pipeline products to advance.

The fix is not conceptually hard, though it will take intent and coordination. 

• Trials and real-world evidence programs can start by fielding a handful of validated spillover questions to first-degree relatives or household members. These questions about health-related activities such as diet and exercise, caregiving hours, and even psychological well-being can be quite informative. 
• Payers and HTA agencies can publish explicit guidance that family/out-of-network benefits will be considered if and when they are credibly measured, lowering the perceived risk of inclusion for sponsors. 
• Researchers can pressure-test new general-equilibrium or agent-based models that translate those survey data into population-level health and productivity gains. 
• And journals can insist that, at minimum, authors disclose whether positive spillovers were looked for and how.

Peer effects remind us that health, like disease, is contagious. Health is passed across lunch tables, living rooms and conference rooms, and even group texts. Capturing and attempting to measure those hidden transmissions won’t just make our models prettier; it will help steer capital and clinical energy toward interventions capable of lifting whole families at once. 

The next time a GLP‑1 script quiets a household’s nightly snack run or an addiction therapy breaks a multigenerational cycle, we should be ready. That means collecting better data, building better methods, and embracing a better, more inclusive value framework – one that counts not just those who received the injection or swallowed the pill, but the full spectrum of beneficiaries who enjoy quantifiable health benefits.