#costsofcare

If you've been paying attention to the literature over the last 6 months, you might feel completely befuddled by all of the conflicting information on osetltamivir (Tamiflu). I know I am!

In the spring of 2014, the Cochrane group published an updated review of data on oseltamivir's effectiveness. This review incorporated previously unpublished and withheld data sets. In short, the Cochrane group found that oseltamivir had no effect on hospitalization rates for influenza. The Cochrane authors argued that the decision to stockpile neuraminidase inhibitors like oseltamivir might be misguided. (For those interested in the full story of the withheld data, I would direct you to this masterful piece.) Additionally, a review in Annals of Internal Medicine found that industry-supported reviews of the drug had more favorable findings and conclusions than studies completed by researchers without a financial conflict of interest.

On the other hand, at least one major review in 2014 found that hospitalized patients treated with oseltamivir derive real benefits including reduced mortality. The CDC's guidelines for this flu season recommend prompt therapy with a neuraminidase inhibitors for high-risk outpatients and for all hospitalized patients with suspected or confirmed influenza. An editorial in this week's NEJM also argues for prompt treatment of pregnant patients.

So what's the deal with oseltamivir anyways? Is every prescription for it just helping to fill Roche's corporate coffers? Or is oseltamivir's routine use justified?

Here's my take. In otherwise healthy adults, there is probably not a role of treating influenza. The relative shortening of a flu-like illness by about one half-day may be of some value to patients, but there is real risk of adverse effects and resistance. I myself probably wouldn't take an oseltamivir script this year -- even though I admit to having placed myself on two separate courses of it over the last several years. 

This week, I did something that would once have been considered unimaginable: I started a patient on a statin without checking a lipid panel!

As recently as one year ago, such practice would probably be considered frankly negligent. Since starting med school in 2006, I had been taught to treat to specific target levels of cholesterol. 

However, this week, I took a trip into uncharted territory in my own practice -- but did something that is clearly evidence-based in my mind!

I just saw a sprite lady of advancing age in clinic who frankly had no major CV risk factors aside from her chronological age and borderline hypertension/prehypertension. She came to establish with a new primary doctor and hadn't seen a physician regularly in years. 

Using the 2013 pooled cohort calculator, I started plugging in her values. "Wait though... There's no lipid panel!" No worries, I put in some nearly irrationally idealized lipids: total cholesterol of 130 and HDL-C of 70. According to the calculator, she has more than a 25% risk of a major atherosclerosis-related cardiovascular event (also called ASCVD) in the next 10 years.

Do I really believe she is at such a high risk? Yes, probably! In my mind, there is little doubt that a low-dose statin will alter her future risk of ASCVD.

Starting a statin without a lipid panel for the first time was a baby step. But why should I not feel so emboldened to start high-dose statins for my patients with diabetes irrespective of their LDL-C? Or for all of my middle-aged and elderly patients with hypertension?

I have often questioned the role of checking serum folate levels in patients with anemia, particularly those that lack macrocytosis. Folic acid is widely supplemented in the United States. While some drugs are known to deplete folate, most good docs know to put patients on 1 mg a day if using these drugs. Even people with alcoholism likely are getting some folate in their diet given the extent to which our food is supplemented with this critical vitamin.

I can recall having heard at least two practicing benign hematologists say that they have never seen a case of folate deficiency causing anemia. Of course, if anyone is going to see a case, it ought to be the specialists who already experience a referral bias. So if the prevalence of folate deficiency is vanishingly low even in specialty clinics, it probably doesn't exist.

Given these considerations, I have vowed to never check a serum folate level myself. I'm sure at some point a subspecialty consultant over-ruled my restraint, but I know I've also stuck to my guns at times... particularly with neurologists who are disturbingly enamored with this test.

I was relieved to see that my suspicions on the rarity of folate deficiency appear correct. And we indeed are greatly overusing and abusing serum tests of folate. A great JAMA Internal Medicine study and accompanying editorial were recently published online showing that folate deficiency is about as rare as can be. 

Will these findings change practice? Unfortunately, I doubt it. Until we strike folate deficiency questions from the USMLE board examinations and eliminate it from available lab tests in our hospitals, we will perpetuate the myth that folate deficiency still exists.

Jesse Theisen-Toupal, MD; Gary Horowitz, MD; Anthony Breu, MD. Low Yield of Outpatient Serum Folate Testing. JAMA Intern Med. Published online August 11, 2014.