|WHAT WE’RE READING
Stuff we read this week that made us think.
Health insurers “make it a bigger bowl”
ProPublica and NPR just launched a new joint reporting series on the economics of the health insurance business, “The Health Insurance Hustle”. It promises to be a really interesting project, judging from today’s first installment. The piece tells the story of a New York man whose hip-replacement surgery resulted in an unexpectedly huge bill—a large chunk of which he had to pay out-of-pocket. Unfortunately for all involved, the man is an insurance actuary himself, and he was surprised to find that his insurer, Aetna, had agreed to a “negotiated rate” of more than $70,000 for the surgery (not including the surgeon’s professional fee), and left the man with a 10% coinsurance payment of more than $7,000. Which led him to ask—why didn’t Aetna get a better deal?
The answer, of course, is that the insurer is simply taking a percentage of the fee as profit, and the larger the fee, the larger the profit. Another term for this is “medical loss ratio”, which the ACA enshrined in law as at least 85% for an insurer. The obvious incentive, which hasn’t gotten nearly enough attention (but which this new reporting series begins to unpack), is for the insurer to agree to a larger rate from the provider, in order to generate a higher dollar amount for the 15% they’re allowed to take off the top. In the words of the article, “It’s like if a mom told her son he could have 3 percent of a bowl of ice cream. A clever child would say, ‘Make it a bigger bowl’.” As much as we assign blame to providers for our high healthcare spending, it’s past time to recognize that the insurance companies—along with many other middlemen (benefit managers, purchasing organizations, distributors)—have been equally complicit in enabling runaway costs. As always, it’s the consumer who gets left holding the bag.
Shave and a haircut, and an ACE inhibitor
The always-excellent Indiana University pediatrician and New York Times contributor Aaron Carroll was out with a piece this week highlighting recent research on community-based management of hypertension. Carroll describes the results of a study conducted by doctors in Los Angeles, which looked at the impact of a barbershop-based intervention program targeted at lower-income, African-American men suffering from high blood pressure. The study built on earlier findings that targeting patients in the barbershop established a level of trust and communication that enhanced compliance with treatment—based on the relationships between patrons and their barbers. The new study went further, finding that coupling the barber-driven program with clinical pharmacist intervention in the barbershop itself, rather than in a clinic setting, yielded even better results in terms of reduced blood pressure. Carroll makes the important point that setting of care matters more than we give it credit for—and that moving care out of traditional settings into community-based venues may be an important approach, especially for populations that either distrust or can’t access clinical care. We know that community-based healthcare workers, and even non-clinical caregivers, can be an important part of improving access and patient engagement—the new study emphasizes the need to widen the definition of “care setting” as well.
Making AI part of the care team
A well-known study (to us care management junkies, anyway) examined the use of a “virtual nurse interface” created to allow patients unlimited, two-way interaction to answer questions prior to discharge. Researchers at Boston Medical Center were surprised to find that 74 percent of patients preferred interacting with virtual nurses “Louise” and “Elizabeth” over their human counterparts; preference for a virtual provider was even higher for patients with lower incomes or poor literacy.
A recent New Yorker article uses this research as a jumping-off point to examine the larger question of artificial intelligence (AI) and its potential to replace—and in some cases, perform better than—“real” providers. When interacting with Louise and Elizabeth, patients found them…more patient. Virtual providers will spend all the time you need and aren’t bothered if they have to repeat themselves over and over. Other studies have shown that patients are more honest with a virtual provider about sexual history, mental illness, substance abuse and other issues they may be reticent to discuss with a doctor. Patients view nurses and physicians as very busy, and frankly, high-status, people. They don’t want to waste their time or embarrass themselves.
The article (correctly) stresses that AI doesn’t replace the human relationship. In its current state, AI is repetitive in providing ongoing support for chronic issues and is a far cry from actual human touch or empathy. But the piece stops short of recognizing that the real question is not whether AI will replace clinicians, but how it can supplement the care team, reduce labor costs, gather detailed information, and create an experience that’s better than human interaction alone.