This week in health & care (2.24.17)
Clinical: Vitamin D cut the risk of respiratory infections so much that some are advocating for more food products to be fortified with it. The studies on vitamin D and (you name it disease) are vast and conflicting, but especially on infection prevention, so the Brits did a meta-analysis on 25 high quality studies (randomized controlled trials). They found a significant decrease in the rate of acute respiratory infections (colds, flu, pneumonia) overall with a number needed to treat (NNT) of 33, and an even greater benefit among those who have profound vitamin D deficiency (NNT = 4) and those receiving regular vitamin D supplementation as opposed to single large dose (NNT = 20). It was also found to be safe without an increase in the feared side effects of high calcium levels and kidney stones. I’m usually the first one to say that taking vitamin supplements without a known deficiency will do nothing for you but give you very expensive urine. After learning of my own vitamin D deficiency and seeing the results of this study, I think the conventional wisdom that “the sun is enough” will evolve. Sure, the studies are inconsistently done and show conflicting results, but this appears to be a well-done meta-analysis with a strong conclusion and meaningful outcome. All of your questions and concerns about confounding and other limitations are likely addressed in the actual paper (http://www.bmj.com/content/356/bmj.i6583). Here’s the easy-to-read article with experts quoted http://bit.ly/2ksDKBS. My simplest, and perhaps oversimplified, explanation as to why we get so many colds in the winter? Less sunlight = less vitamin D = relatively weaker immune system. Merely spending more time inside is a very weak hypothesis alone but may contribute. I’m very open to being contradicted, but right now I’m a believer in D (and admittedly biased).
Mobile: The World’s first *approved* app for contraception — input daily temperature and an algorithm tells a woman if she is fertile that day. This specific method was as effective as the pill for birth control in one study (I haven’t read the study). The app was developed by a Swedish nuclear physicist and is now approved in the UK (while awaiting FDA review). The concept isn’t new, but it looks like her algorithm is superior. https://getpocket.com/a/read/1603534916
AI: MD Anderson pauses its collaboration with IBM’s Watson product. If someone asks you to name the top cancer hospitals in the world, you better include Memorial Sloan Kettering in New York, and MD Anderson in Houston, both of which partnered with IBM several years ago to incorporate the AI of Watson into their oncology treatment decisions. It’s unclear if the decision from MD Anderson is more related to odd violations of internal protocols, not achieving expected results from the IBM product, both, or something else entirely. I’m not surprised, however, if Watson isn’t the lone decision support genius of the machine world at this point in time, 6 years after it won Jeopardy. It may be a setback for AI in Medicine on the surface, but IBM has recently announced many new partnerships. My bet is that this will end up being one small bump in the road but a friendly reminder to demand evidence before spending millions on a new product in healthcare. http://bit.ly/2mjkl3L
Obituary: Kenneth Arrow passed away at 95 — the legendary economist made numerous contributions to the field and was a giant, to say the least. For healthcare, his famous 1963 paper on “Uncertainty and the Welfare Economics of Medical Care" laid the foundation for how we think about markets with information asymmetry in general, and how we think about “moral hazard” in relation to health insurance, more specifically. It’s imperative to understand his work and its derivatives in the history and future of health policy.
Barry Breaux is a Board-certified Internal Medicine physician practicing and teaching at Stanford. His healthcare posts consist of brief reactions to select pieces of health and medical news to help bridge the profession-public divide. Thoughts and opinions are his own and not meant to serve as the advice of your doctor. Interesting articles? tweet @barry_breaux