What If a Physician Prescribed a Soccer League Instead of Therapy?

“When someone is anxious, lonely, or depressed, the standard advice is to see a therapist. But therapists are scarce, waitlists are long, and many people—especially teenagers— never make it through the door . So what if your doctor could skip the waitlist and prescribe gardening, a walking group, a choir, cooking classes, or a soccer league instead? It’s called social prescribing, and it’s coming to the U.S. What is Social Prescribing? The idea is a primary care doctor refers a patient to a modestly paid community connector who helps the patient find and stick with activities outside the home. In the UK, over a million people a year now receive a social prescription, and the National Health Service has made extending the model to teenagers a national priority . The approach has deep roots in something therapists already know works. In a recent piece for Education Next , we argued that behavioral activation is the phonics of therapy: It works, but therapists don’t like it. The basic idea is simple: When people feel bad, they withdraw, and withdrawing makes them feel worse. Behavioral activation reverses that cycle by getting people back into meaningful activities. Social prescribing partly solves the adoption problem by bypassing reluctant therapists altogether, handing patients off instead to “ link workers,” who are perhaps akin to paraprofessionals in the K–12 world. The link worker doesn’t ask why you feel bad. The link worker says, “Let’s find ways to get you out your front door to do something that matters to you.” As social prescribing has scaled, though, it has generated controversy that is reminiscent of debates over tutoring in K–12: promising results in select contexts, lots of overstated claims, and a push to grow faster than quality control systems can handle. We think the uneven evidence has to do less with whether social prescribing “works” and more with how much of it people actually get and how well it’s run. We think this because we watched it happen with tutoring. A Trip Down Tutoring Lane: Basic, High Dosage, and High Quality High Dosage Mike remembers Wednesdays in Fall 2001, when the volunteer tutors would show up at our little Boston charter school, called Match. A dozen, maybe two dozen. Many grad students, a few retirees. They’d come once a week. Sometimes you’d see a kid have the absolute best 90 minutes of her week—sitting side-by-side with her tutor, actually learning, defenses temporarily down, maybe even a smile or a hug. It was magic. The logistics were bad, though. We had a volunteer coordinator whose 50 hours a week of herding volunteer cats would sometimes generate just 40 hours a week of actual volunteer hours. A typical student might get just 20 hours of tutoring a year. Tutor supply was high for September and October and February and March but always waned as college exams approached. We could count on just four months out of 10. Then we radically overhauled our school. We coined a phrase—“High Dosage Tutoring”—and massively upped the supply. Your friendly reminder that the catalyst of the high-dosage tutoring movement — MATCH Corps / SAGA Education — was a model where tutoring was a regular class every day of the week ~165 total hours. Most tutoring programs we are evaluating today provide 15-30 total hours AT BEST. — Matthew A. Kraft (@MatthewAKraft) March 27, 2026 To our dear friend Professor Matthew Kraft (who back then was our wunderkind doctoral student advisor), it was actually 165 hours per year in both math and English! Roughly 330 hours per year total, per kid, no exceptions. That story was told in 2015 in Education Next . Dosage mattered, but so did a million details: quality leaders like Kamala, Lisa, and Alia; measurement; memorable Thanksgiving dinners. The movement to “scale up” tutoring over the past six years has been choppy. Mike Duffy, in his Education Next review of Liz Cohen’s book The Future Of Tutoring for Education Next , writes : One of the more refreshing aspects of Cohen’s analysis is her willingness to surface uncomfortable truths about where tutoring failed. Tutoring programs that began as emergency relief often faltered when federal pandemic funds tapered off. Some districts underestimated the logistics of matching students, coordinating schedules, or retaining consistent tutors. Others failed to integrate tutoring deeply enough into the instructional core, treating it as “remediation” for struggling students rather than opportunity to personalize learning for all students. By being candid about these failures, Cohen gives her leaders permission to anticipate challenges and build in adaptations from the start. If social prescribing is to have a better fate than the complicated legacy of scaled-up high-dosage tutoring, it will need to be candid about its own struggles to date. The Evidence on Social Prescribing The most comprehensive evidence to date comes from a March 2026 nationwide analysis published in Nature Health . Analyzing records from nearly 20,000 patients across more than 300 social prescribing sites in the UK, University College London researchers found consistent and sizeable improvements across five well-being measures in the one-to-six months following referral. That’s a compelling headline. But the study has no control group, so regression to the mean cannot be ruled out. More importantly for our purposes, it has almost no information about what patients actually did after referral, how many sessions they had with their link worker, or how many hours they spent engaged in community activities. The signal is there. The dosage is invisible. This is exactly the problem a systematic review in BMJ Open surfaced when it examined eight controlled studies of social prescribing programs for adults and found little impact on mental health or quality of life. The research design here is stronger. Rob Poole and Peter Huxley went further in a 2024 piece: They made the case that that the gains from social prescribing tend to disappear once staff involvement ends, and the people who benefit most are those with the mildest needs. In other words, the critics found roughly what you’d find if you evaluated volunteer tutoring without distinguishing it from high-dosage tutoring: a mush of weak implementation producing weak results. Social prescribing in these studies was often a quick referral, a few loosely defined sessions, and very little follow-up. In one study, link workers met with their clients an average of 1.7 times. These were mild interventions that quickly sputtered out. This version of social prescribing we’d liken to “typical tutoring” in the early years of Match, when there was neither meaningful dosage nor quality control nor deep relationship. The strongest positive evidence came from within that same BMJ Open review. Two of the eight studies stood out. Link workers met with patients weekly for six months, helping them set individualized goals and work through whatever social frictions kept getting in the way. Here, 91 percent of patients stayed engaged for the full intervention. The review drew the lesson plainly: Intensity mattered, and current plans to scale social prescribing with higher caseloads and shorter interventions risked undermining the exact ingredient that made it work. Teenagers and the Dosage Problem So, does the dosage lesson hold for teenagers who get social prescribing? INSPYRE , a program making one of the first rigorous attempts to test social prescribing with young people (instead of with adults), paired link workers with teenagers on waiting lists for therapists. In practice, most young people received about six sessions with their link workers, though 18 percent did not engage at all. The study showed null results on anxiety and depression. Behavioral activation works in randomized trials. Social prescribing for teenagers didn’t work in randomized trials. Why not? Dosage and quality. We think there are actually two dosage problems, and the social prescribing field is only paying attention to the first one. The first variable is time with the link worker. Six social prescribing sessions is thin compared to what worked for adults. The programs in the BMJ Open review that moved outcomes involved weekly contact for six months. But the second dosage problem is bigger, and almost nobody is measuring it: How many hours per year the teenager spends doing the prescribed activity. In INSPYRE, many link worker sessions were consumed figuring out what activities existed in the community. Some kids settled on one-off experiences. Some couldn’t get a ride or didn’t have the money. (Participants were granted about $50 each.) Going to a museum once for an hour is lovely. But forming a habit of going to the gym three times a week, or rehearsing and performing with a little band—those can be commitments of 100, 200, 300 hours per year. For kids who want a job to get them off the couch, a grocery store gig can be transformative . It means perhaps 1,000 hours per year away from your phone, away from your anxious thoughts, spending time in a group instead of alone. The social prescribing field hasn’t yet figured out that it needs to count total activity hours to distinguish a one-off experience from a life change. And it may be that social prescribing needs to be paired with a hefty funding source, like Flourishing Scholarship Accounts . Professional Quality Matters—but Qualifications Don’t Predict Quality Before concluding that social prescribing just needs more navigation sessions with link workers, who then become conditioned to find habit-forming activities for the long haul (i.e., solving the dosage problem), it’s worth addressing an obvious objection. Maybe social prescribing underperforms its cousin behavioral activation because link workers aren’t trained therapists. Hire better-credentialed people, run more training programs, and maybe the results will follow. Right? The evidence doesn’t support this. Nobody has done more to demolish the notion than Harvard’s Vikram Patel . He has spent decades demonstrating that effective psychological treatment does not require a credentialed clinician to deliver it. His Healthy Activity Program (HAP) , tested in a randomized trial published in The Lancet , is behavioral activation delivered not by therapists but by lay counselors—people drawn from local communities with no prior mental health training, akin to link workers. Against moderately severe to severe depression in primary care in India, it worked. Effects were sustained at 12 months, and the cost per recovery was a fraction of specialist-delivered care. EdNext in your inbox Sign up for the EdNext Weekly newsletter, and stay up to date with the Daily Digest, delivered straight to your inbox. Email Name Opt in to another list EdNext Daily Digest Subscribe The same lesson has been learned repeatedly in education. Writing in these pages, Dan Goldhaber concluded that the proxies most school systems use to screen and pay teachers—certification, years of experience, education level—are “well researched, but there is little definitive empirical evidence that these characteristics, defined in general terms, are associated with higher student achievement.” Eric Hanushek went further in a subsequent Education Next analysis : Master’s degrees are not only uncorrelated with value-added but may be weakly negatively associated with it. Thus the lesson in teaching, in tutoring, and in therapy: Quality is real and consequential, but credential proxies don’t capture it. You cannot know who has it until you watch them do the work. The potentially good news is that although behavioral activation hasn’t taken off clinically, social prescribing has. The Lancet recently mapped 23 US programs already up and running. SocialRx, an Atlanta-based operator working in nine states with partners that include Stanford and Mass General Brigham , describes its model as grounded in behavioral activation and long-term habit formation. The question is whether all that momentum produces more six-session programs that fizzle out or something that works. To us, high-dosage, high-quality social prescribing means connecting young people to activities that generate hundreds of hours of engagement per year—the kind of sustained involvement that actually changes how someone feels. Let’s try to make it good social prescribing for teens. We need High-Dosage High-Quality Social Prescriptions here in the U.S. We need to roll up our sleeves and find the link workers who are wired just right to really help teens. And we need to valorize them, compensate them, retain them, and study them. Sean Geraghty and Mike Goldstein are the co-founders of the Center For Teen Flourishing. The post What If a Physician Prescribed a Soccer League Instead of Therapy? appeared first on Education Next .

Summary generated from the RSS feed of Ed Next. All article rights belong to the original publisher. Click through to read the full piece on www.educationnext.org.