Less is More

I had wanted to write on the utility and interpretation of serial CBCs. I do it all the time and I’m sure my practice is not well-informed. However, it’s doing. It’s active. And we love doing.

(Check out Drs Sinert and Zehtabchi’s paper₁₎

Speaking about doing, yesterday I was listening to the freakonomics podcast entitled, “How Many Doctors Does it Take to Start a Healthcare Revolution₂” and my mind blew. The two-part series is broadly about our imploding system, its dysfunction, perverse incentives and consequences. A Macarthur-award winning family doc from Camden, NJ talks about shutting down his practice while two mega-hospital systems blossomed within 10 miles. About how reimbursement structure incentivizes procedures, testing and hospitalization. And outcomes? Who cares? Patient-doctor communication, patient-education and information, trust, relationships: not billable and money-losing. Our trajectory is towards more doing (less talking) and is consuming our national GDP like a malignancy. And yet our outcomes are mediocre, at best. Value is absent.

Speaking on value, on doing and on outcomes, the meatiest part of the podcast is when Dr Jena, an MGH internist, gets interviewed. In February, 2015, Jena published an article entitled “Mortality and Treatment Patterns among Patients Hospitalized with Acute Cardiovascular Conditions during Dates of National Cardiology Meetings₃” in JAMA. Or, what happens to patients with acute cardiologic badness when cardiologists are gone? The author’s diplomatic hypothesis is that patient’s outcomes during these international conferences would be worse. What they found blows the mind.

A retrospective review of all Medicare beneficiaries from 2002-2011: they looked at all patients admitted with cardiac arrest, acute myocardial infarction and congestive heart failure during the dates of international cardiology conferences and compared them with those admitted on the 3 weeks before and after the conferences. They divided patients into high- and low-risk categories and analyzed teaching vs non-teaching hospitals. Primary outcome: 30 day mortality. Secondary outcomes: Length of stay, procedure utilization, charges.

What they found: Those admitted with cardiac arrest to a teaching hospital during the dates of an international cardiology conference were ~10% less likely to die within 30 days compared with those admitted during non-conference dates. Those admitted with high-risk heart failure to a teaching hospital during a conference were of 7.5% less likely to be dead in 30 days compared with those admitted during non-conference dates.

Notably, high-risk AMI patients did not share the mortality benefit. However, there was a statistically significant difference in rates of PCI. 20.8% of high-risk AMI patients underwent PCI during conferences dates vs 28.2 during non-conference dates. Again, with no difference in mortality.

What does it mean? It’s a retrospective, observational study. The national cardiology associations are quick to point out the lack of causality. The president of the American College of Cardiology said he was “reassured by the finding that patient mortality didn’t increase when those doctors were away₅.” Talk about an understatement.

What we do know is that patients suffering cardiac arrest and high-risk CHF patients do better if they present to a teaching hospital during a conference. But we don’t know why. We know that high-risk AMI patients admitted during conferences undergo PCI less frequently but have unchanged mortality. But we don’t know why.

The reason I love this study is that it spotlights how little we know about what we’re doing and how confidently we do it. If this study is “true,” the number needed to “treat” to prevent one death is 10 and 12.4, respectively for cardiac arrest and heart failure. Keeping in mind that “treatment” in this case is having been admitted during a cardiology conference. And if real, this “treatment” is one of the most effective interventions in recent medical history, far more powerful than aspirin or thrombolytics in STEMI_4.

At least if we’re screwing up, let’s have the security and dignity to name it and address it. And refocus on why we do. Patients and outcomes.

If we learn one thing from this paper, we should have more cardiology conferences.

Zehtabchi, Shahriar, et al. “Diagnostic performance of serial haematocrit measurements in identifying major injury in adult trauma patients.”Injury 1 (2006): 46-52.
http://freakonomics.com/2015/04/09/how-many-doctors-does-it-take-to-start-a-healthcare-revolution-a-new-freakonomics-radio-podcast/
Jena, Anupam B., et al. “Mortality and Treatment Patterns Among Patients Hospitalized With Acute Cardiovascular Conditions During Dates of National Cardiology Meetings.”JAMA internal medicine (2014).
http://www.thennt.com/nnt/thrombolytics-for-major-heart-attack/
http://www.npr.org/blogs/health/2014/12/22/372508398/is-your-heart-doctor-in-if-not-you-might-not-be-any-worse-off