Yes, we should definitely study that!

Two studies published last week emphasize the benefits of randomized trials--not only for medical interventions

Let’s think today about one of evidence-based medicine’s biggest blind spots: the study of policies.

Drugs, devices and treatment strategies often face the challenge of randomized trials. Yet policies mostly escape this challenge. It seems all a policy needs is a strong champion and plausibility.

Yet the failure of the hospital readmission penalty policy and sepsis protocols expose the risks of implementing a policy because it sounds good.

---

Last week, two randomized trials reported results on using cash giveaways to improve health.

The background for cash giveaways stem from the association of low income and worse health outcomes. If low income causes poor health, then treating low income (with cash) should improve health.

The medical journal JAMA published the first trial.

A Harvard group of mostly medical people randomized 2800 low-income people in a Massachusetts town to receive a cash benefit of $400 per month for 9 months or not.

Their primary endpoint was emergency department visits. These were young people (mean age 45) who had a mean family income of $16,000/year. More than 70% were first-language Spanish speaking.

The primary outcome was positive but there were caveats.

ED visits in the cash benefit group were 217 per 1000 person years vs 317 per 1000 person years. Which met statistical significance. But there were no differences in out-patient visits, visits to primary care, or out-patient behavior health. Visits to specialty care were slightly higher in the active arm.

Importantly, however, there were no differences in health outcome measures, such as BP, BMI, HbA1c or cholesterol level.

---

Dr. Sarah Miller, an economist in the University Michigan business school, led the second trial, named the Open Research Unconditional Income Study or ORUS study.

Their team published the paper as a NBER working paper. (The first thing to say is that these working papers must not have a word limit, because, compared to a JAMA or NEJM write-up, the NBER paper is massive.)

The Michigan economists asked the same basic question but used more robust methods. They randomized low-income adults to receive $1000 (tax free) per month or $50 extra per month. The study carried on for 3 years (not 9 months).

Their results were sobering. Relative to the control arm, the cash transfer arm resulted in large but short-lived improvements in stress and food security, greater use of hospital and ED care, and modest increases ($20/month) in medical spending.

But they found no effect on several measures of physical health as measured from surveys and biomarkers.

Unlike many medical researchers, the economists wrote strongly about the null results, as if they were not disappointed.

“We can rule out even very small improvements in physical health and the effect that would be implied by the cross-sectional correlation between income and health lies well outside our confidence intervals.”

We also find precise null effects on self-reported access to health care, physical activity, sleep, and several other measures related to preventive care and health behaviors.”

Comments

I see two big lessons in these trials. First is the idea that we should at least try to empirically study major policy interventions.

We can argue whether these studies adequately answered causal questions, but without any studies, there can be no arguments—only expert opinions. (The first study, with its modest cash benefit over only 9 months, seemed unlikely to make a difference.)

But still, I laud these efforts in the highest terms possible. This, I believe, is the way forward. No major health policy should be foisted on people without attempts at empirical study. If I were head of NIH, this would be one of my major areas of focus.

The second lesson is that the relationship between poverty and health is complicated. And that many are drawn to simple solutions—like cash transfers or expanding care.

In 2019, I wrote about the provocative book Elephant in the Brain--co-authored by George Mason economist Robin Hansen.

Hansen has long argued that there is little evidence supporting the idea that increasing access to healthcare leads to better health. He cites the Rand, Oregon and Indian health insurance studies, which all find little to no effect of giving health insurance to people.

There are plenty of medical studies, too, that find little to no effect of increasing access to medical care.

The MI-Free study found no benefit to giving full coverage of post-MI meds; the ARTEMIS trial found no reduction in 1-year major cardiac events with co-payment vouchers for post-stent platelet inhibitor meds. And the Beth Israel group in Boston could find no benefit in the treatment of cardiac risk factors in low-income adults in states that expanded Medicaid vs those that did not expand Medicaid.

I am not as negative about healthcare as Professor Hansen. And, I am not against improving disparities in outcomes between richer and poorer citizens.

But. Since I am a simple cardiologist not a policy wonk, I cannot offer a solution to the disparity problem in US healthcare.

My point from these studies is that the simple and elegant solution of cash transfers did not work. And. Had it not been studied; we would not have known that.

Knowing what does not work is surely a valuable thing to know.

---

Once again, we thank you for your support of Sensible Medicine. Your support keeps this site free of advertisements. JMM