Hotspotting was never implemented as an intervention to address upstream social determinants of health, but rather as a cost containment strategy — and it failed at that too.
On January 4, CNBC reported on a Freakonomics radio episode from November 2020 in which Whole Foods CEO John Mackey graced the world with his very astute and novel big idea for twenty-first-century health reform:
I mean, honestly, we talk about health care. The best solution is not to need health care. The best solution is to change the way people eat, the way they live, lifestyles and diet. There’s no reason why people shouldn’t be healthy and have a longer life span.
His statement quickly drew a backlash on Twitter, including from Congresswoman Ilhan Omar, who highlighted the cruelty of Whole Foods slashing healthcare benefits for its part-time employees. This was not the first time Mackey’s messaging fell flat; in 2009, he penned a WSJ piece about the importance of personal responsibility and maintaining healthy lifestyles, attempting to kneecap the proposed ACA. What makes his statements especially repulsive is that people with chronic medical conditions will not be cured of simply buying organic groceries or adopting yoga regimens, nor will eating fewer carbs or cholesterol ward of infection during a deadly global pandemic.
It so turns out that the rhetoric of Mackey’s calls is not too far off from the ethos behind healthcare hotspotting. 10 years ago, general surgeon and author Dr. Atul Gawande detailed the practice in a New Yorker piece (along with its totally cool and not at all offensive thumbnail), in which he demonstrated how a team of community health workers in Camden, NJ brought down ER and hospital visits by using spatial profiling to create maps of select patients whose care comprises a disproportionate amount of healthcare spending. In short, the practice of hotspotting aims to link frequent patients, known as “super-utilizers,” with follow-up outpatient appointments and social services after discharge. The goal is to address patients’ social needs upstream and buffer their otherwise high impact on resource-strapped hospital systems. While out-of-touch multimillionaire CEOs scold about preventative health and pearl-clutch about the deficit, local hotspotters supposedly pick up the slack and meet patients where they are to keep them out of the hospital as much as possible.
Disappointingly for the Camden Coalition, however, the results from a randomized controlled trial by Finkelstein et al. at MIT did not bear out in their favor. The study randomly assigned 800 medically and socially complex patients, half to intervention and half to control, and found that hotspotting yielded no benefit in hospital readmission rates within 180 days of discharge compared with the control group (62.3% vs 61.7%). So much for the hype — perhaps in the future the RCT can be done before the supporting op-ed.
To clarify, the findings from this study are not an argument to stop case management services for high-need patients and should not be used as a pretext for administrators to make such cuts. Community efforts to provide patients with housing vouchers, meals, transportation, mental health services, primary care appointments, etc. should be encouraged because insofar that they are successful, they can contribute to better care and overall more positive experiences with the healthcare system. On-the-ground social support is good on its own accord and need not be tied up with the externalities of meeting readmission caps; routine care should be supplied morally, not budgetarily.
From a health systems standpoint, it is important to examine the original premise of hotspotting — reducing healthcare spending via reducing utilization — and then zoom out. Americans, on average, do not use more healthcare than their counterparts in other countries. They go to the doctor less, spend less time in the hospital, and receive less long-term care than most other industrialized nations, according to a 2019 OECD report.
While hotspotting proponents are correct to note that ~5% of patients drive a majority of healthcare spending, it is remiss to simply home in on utilization as the cause of high healthcare costs. A hyper-focus on heatmaps of high-usage patients elides the fundamental reason that the US spends twice the average OECD nation on healthcare (but still has worse health outcomes). Healthcare costs are astronomically high in the US because the unit prices of healthcare are high, not because of high aggregate utilization. Costs = Prices x Volume. When reading Gawande’s famous “Cost Conundrum” in retrospect, he was wrong not because he could not predict the outcome of the RCT study; he was wrong because he failed to acknowledge that vital fact and the false assumptions he adopted as a result.
On some level, it may seem that hotspotting initiatives channel the vision of Sir Michael Marmot, the professor of public health and epidemiology who showed that to reduce health inequalities on a global scale, we must address the social determinants of health (SDoH), a phrase that he helped popularize. But there is a crucial difference between the comprehensive social investments he prescribes and the practice of using GIS tools to scope out high-risk patient populations. The Camden model was specifically applied to certain patients at select periods, not meant for a broader context. Dr. Marmot’s research group, on the other hand, is explicit about the myriad sociopolitical factors that underlie health disparities across the whole lifespan, emphasizing poor childhood development secondary to child poverty, for example, as a key determinant of poor health later in life.
The question becomes: in which context should care coordination services occur, if not in a bandaid manner under the current fragmented healthcare system? In his 2019 book Health Justice Now, activist Tim Faust argues that a federal single-payer program is a giant instrument for health justice because it bears the costs of both providing care and not providing care. Once everyone is covered under a single risk pool, not only can we most effectively address structural health inequities, but we have an incentive to do so, collectively as a nation, in both the short- and long-term.
As Faust argues, if a family lives in a dilapidated neighborhood or is homeless, then housing becomes healthcare, and the government has an incentive to invest in safe public housing. If a family lives in a food desert and can only access unhealthy fast food, then all of a sudden food becomes healthcare, and the government has an incentive to invest in accessible grocery stores with fresh produce. And so on with air pollution, lack of transportation, etc. Makeshift hotspotters need not apply.
In fact, there is a case to be made that hotspotting only gets in the way. In the 2017 book Subprime Health: Debt and Race in U.S. Medicine, Nadine Ehlers and Shiloh Krupar observe:
In much of the public discourse surrounding medical hot spotting, the imperative of cost containment and minimizing this form of health care expenditure seems to eclipse the need to attend to health inequities. This is evident in the very language used to talk about those who are targeted: they are the “high utilizers” and “the superuser 1 percent”; they are identified as “outliers” and identified as “socially disintegrated.”
As the medical geographer and public health researcher Arrianna Planey notes, the authors continue to describe how hotspotting is often applied across racial lines and reinforces racial capitalism, juxtaposing crime-mapping and broken-windows policing with the present modalities of healthcare utilization analyses. “Superutilizer” has become the new “superpredator.”
The 2010s was a decade of immense solutioneering in healthcare. In addition to several regional hospital systems championing hotspotting in a start-up-style fashion, the ACA rolled out a host of top-down “value-based” programs such as the Hospital Readmission Reduction Program. These schemes, which reward/punish health systems for meeting/failing quality metrics & quotas, not only have little to show but have caused harm. Gawande’s own healthcare venture company Haven, which was bankrolled by JPMorgan Chase and Berkshire Hathaway, recently shuttered “with little to show for” and “whose accomplishments remain unclear.”
It is time to end the crazes around unsustainable stopgap proposals that only tinker around the edges and ignore the big picture. It is time to get serious about healthcare reform and organize for a single-payer Medicare For All system.
Additional References
The Myths and Lies of the Health Care Quality Industrial Complex — Cleveland Review of Books