March 28, 2016

Make No Bones About It

For some time, I’ve tried to find an up-to-date list of the medications most commonly prescribed to older people. Sounds like a simple question, but getting an answer has been surprisingly challenging. Most of the available data is ten years old and that’s a long time in an era when medications go off patent, new medications are introduced, and advertising campaigns affect medication use. Much of the information is for the population as a whole—but kids really are very different from octogenarians in their pill-taking. So I was pleased to find an article in JAMA Internal Medicine this week called “Changes in Prescription and Over-the Counter Medication and Dietary Supplement Use Among Older Adults in the United States, 2005 vs 2011.” This nationally representative sample of community dwelling older adults got its information from in-person interviews. It over-sampled certain populations to try to make sure its interviewees were truly representative. And the results are revealing.

The main finding is that fully 87.7 percent of adults over 65 (excluding those in institutions) took at least one prescription drug regularly in 2010-2011, up slightly 2005-2006. Moreover, 35.8 percent of the population take at least five prescription drugs a day (up significantly from 2005, when the rate was 30.6 percent. Lastly, there’s been a 50 percent increase in the number of people taking vitamins or supplements.

The Big Ten medications are pretty much what you would expect, though the actual percentages are a bit surprising. In first place is over-the-counter aspirin (40.2 percent); simvastatin, a cholesterol-lowering medication, is in second place (22.5 percent), and atorvastatin, another statin (formerly sold exclusively as Lipitor, before it lost patent protection) is number ten. The number three, four, six, and seven spots are taken by the anti-hypertensives, lisinopril (19.9 percent), hydrochlorothiazide (19.3 percent), metoprolol (14.9 percent), and amlodipine (13.4 percent) respectively, although it’s worth pointing out that these drugs can be used for other purposes besides lowering blood pressure—hydrochlorothiazide is a diuretic that may be used to treat heart failure, metoprolol is a beta-blocker often used to treat angina, and amlodipine is a calcium-channel blocker that can also be used in coronary artery disease. 

The remaining three drugs on the list are levothyroxine, a thyroid replacement medication, in fifth place, metformin, a drug used to treat diabetes, in eighth place, and omeprazole, a proton-pump inhibitor used for ulcers and acid reflux in the ninth spot. We are left wondering what all this means: are older people getting too many drugs? Not enough drugs? Are they getting the right medications?

Descriptive statistics cannot answer whether some patients are getting medicines they don’t need (though I’m pretty sure that’s the case) and others aren’t getting medicines from which they might benefit (probably also the case).  I think they do tell us something about the effectiveness of the strategies used to promote medications. When medications are categorized by type, statins are actually taken by just over 50 percent of older people (simvastatin and atorvastatin, drugs number 2 and 10 in the list of individual agents are not the only statins available) and anti-hypertensives by just over 65 percent of the elderly. What this tells me is that the combination of direct-to-consumer advertising, drug detailing to physicians, and  professional society guidelines--the methods used to promote statins and anti-hypertensives,at least when new drugs in each of these classes appeared on the scene--really works to change behavior. It doesn’t prove anything, but it’s awfully suggestive.

Also worth exploring is the dramatic increase in the percent of older people who take supplements. The authors of the study assert that this occurred although there is “no evidence of any clinical benefit.” I think this is a distortion. There may be little evidence of clinical benefit for some of the supplements, such as omega 3 fatty acids, but the story for vitamin D and calcium is both messier and more illuminating.

Over the years, Vitamin D has gone from being clearly necessary for strong bones, to very useful in preventing falls, to a dangerous poison, to a useless additive, and back again. Just what do we know as of 2016? We know that vitamin D is essential to human beings and we get it from sun exposure or from diet, although not many foods other than the ones such as milk to which we now add it naturally contain Vitamin D. Actually, that’s not quite accurate either, as what we get from the sun and from food is a pre-cursor of the active form of vitamin D that we need to make bones, and we rely on our kidneys and livers to perform the transformation. We also know from the National Health and Nutrition Examination Study that at least as of 2005-2006, 42 percent of adults had vitamin D levels below 20 ng/ml, which just about all authorities regard as too low. We also know that people who take megadose vitamin D as part of a fad diet, sometimes taking as much as 100 times the recommended daily dose, can get poisoned by such quantities.

The big question remains whether taking supplementary vitamin D—on the order of 800 units a day (not the tens of thousands of units taken by fad dieters)—prevents falls and fractures. Falls and fractures cost over $28 billion in older people, and that’s just the direct costs; it doesn’t include the pain and suffering and the loss of functioning and independence. The data on the efficacy of vitamin D are a mixed bag, with some studies showing strong evidence that it helps and a few failing to show any benefit at all. Putting all the conflicting evidence together, the American Geriatrics Society recommends, based on the preponderance of evidence, that all older adults, whether living in the community or in an institution, take vitamin D supplements of at least 1000 units together with calcium. Judging by the JAMA Internal Medicine article, we have a long way to go to reach this target: while 35 percent of older people do take a multivitamin (which includes 400 units of vitamin D), just under 16 percent take vitamin D alone.

The back story here is that vitamin D is cheap. No drug company is promoting vitamin D. In addition to being cheap, vitamin D has virtually no side effects (unless it is taken at hundreds of times the recommended dose). And it just might work. We should think about the ways that the consumption of cardiac medications have changed—and the ways that these changes have been achieved. We might learn something about how our system operates and how we can change the attitudes and behavior towards therapy that has a good chance of helping without breaking the bank.

March 14, 2016

Pay to Plan

When Medicare began allowing payment to physicians for advance care planning on January 1, bloggers and editorialists and columnists all commented on the new rule. Many said what a great new advance this is. Dr. Diane Meier, founder of the Center for the Advancement of Palliative Care, said it was “substantive and symbolic.” Others were more guarded. Dr. Robert Wachter of the University of California San Francisco said he expected a “modest uptick” in the number of advance care planning conversations, but did not anticipate the rule would be “transformative.” Now, an essay in Health Affairs pulls together the various comments and critiques and concludes that unless we overcome the prevailing “training deficit,” the pervasive inability of physicians to carry out such conversations, and unless we develop a health care system that allows for the implementation of whatever choices patients make, the reform will be meaningless.

The reason I didn’t blog on this topic—apart from the fact that so many other voices were chiming in—is that it seemed obvious to me that the change was exclusively symbolic. It was obvious because the truth is that doctors have been able to bill for advance care planning visits for years. Instead of using the elaborate “coding system” that most doctors use for billing purposes, in which you have to assess the level of complexity of the history, physical exam, and something nebulous called “medical decision making,” it is perfectly legitimate to bill based on time. All you have to do is state how long you spent with the patient (and family) and write that “over 50% of the visit was devoted to counseling.” If you do that, then there are no specific rules about “documenting” the physical examination and the history--your note can focus primarily on the substance of the visit, advance care planning. And you get paid more for a 40 minute office visit using the old system of time and counseling--$145.82 in 2016—than you do for a 30 minute advance care planning visit using the new code--$86.66 in 2016 (though you can also tack on an extra 30 minutes for filling out forms with the new system and bill another $75.11). 

To be sure, to use the old system, the patient has to be physically present. You have to write something in your note about the medical history, but “77 year old retired lawyer with stage 4 non-small cell lung cancer, unresponsive to chemotherapy” is good enough; and you have to write something in your note about the physical examination, but  “patient is pale and cachectic; he is hoarse and dyspneic" should suffice. The remainder of the note can address goals of care, choices about limitations of treatment, designation of a health care proxy, and so forth.

Now I don’t want to underestimate the power of symbols—especially since the effort to include any kind of mention of advance care planning in the Affordable Care Act was dead on arrival after Sarah Palin made her notorious “death panel” comments. But sometimes adopting health policies that are doomed to be ineffective leads us astray because we think we have “solved” whatever problem led to the introduction of that policy in the first place, and result in our failing to solve the problem over the coming years. Perhaps the Patient Self-Determination Act of 1990 was a legislative example of the same phenomenon—as a result of this law, every state and the District of Columbia passed some sort of advance directive legislation over the next decade, but we now recognize that this kind of “legal transactional” approach to advance care planning, rather than a more communications-based approach, doesn’t work: either people don’t use it, doctors don’t follow the directive, or the directive doesn’t apply in precisely the clinical situations that real people find themselves in. 

So yes, Medicare’s decision to reimburse doctors for time explicitly spent on advance care planning is symbolically important. But I worry that it will result in unwarranted complacence, in our checking off advance care planning reform as “accomplished” on our national to-do list. Now that would be a serious mistake.

March 07, 2016

What We Have in Common(wealth)

A year ago, I reported on an interesting comparative study of older adults: the Commonwealth Fund surveyed the health care experience of adults 55 and older in eleven developed countries and found some striking differences. Now Commonwealth has drilled deeper into its data and analyzed differences among “high need” patients in the US and eight other countries (Australia, Canada, France, Germany, the Netherlands, Norway, Sweden, and Switzerland). As usual when we compare ourselves to other countries in the health arena, we don't do so well. And as usual, the differences are enlightening.

This new study looks at patients who are “high need.” I like that term: instead of talking about “high risk” patients--patients who are really at high risk of hospitalization, institutionalization, or death because they already have a lot of needs, not because we can magically determine that they might develop needs in the future--we focus on people who have problems now. The study defines people as high need if they have either three or more chronic conditions or need help in one of their basic daily activities. I might have preferred a composite measure of chronic diseases and functional difficulties, but it turns out to be useful to separate the two for purposes of international comparisons. 

Which brings me to the first interesting observation: the US has more people with at least 3 chronic diseases than anyone else, by a lot. In the US, 42 percent of people over 65 have at least 3 chronic diseases. No other country even comes close. Switzerland is the best, at 19 percent. Everyone else is in the 20-29 range. Does this reflect actual disease rates? Or is it just that we are more thorough in diagnosis—some might say by over-diagnosing disease? The flip side of this finding is that the US performed best in ADLs—only 14 percent of Americans reported they needed a moderate amount or a lot of help, compared to 50 percent of the French. Surely this is cultural—I can imagine that individualistic Americans like to be self-reliant and don’t want to accept help; perhaps the residents of other countries are far more likely to feel that as they get older, they deserve help.

If the populations are as different as the disease and ADL prevalence variability suggests, then the differences that were found in access, costs, and coordination may be meaningless. But for what it’s worth, here's what the study found: the US has a high rate of preventable emergency room visits (19 percent, compared to a low of 4 percent in Germany); and a high rate of cost-related access problems (22 percent vs a low of 5 percent in Switzerland).

Coordination of care was poor across the board—except in France. It sounds as though French people like to have things done for them, both in terms of assistance in basic activities and having someone arrange their health care for them. They report that they actually get help with coordination; I wonder if they feel they get the help they need in other domains as well.

The US came out on top in a single area: the proportion of older people who report they have a “plan of care.” Since having a plan doesn’t amount to much if you can’t access the services you need in order to implement the plan and you don’t have anyone to help you make sure you get what the plan says you need, this accomplishment isn’t terribly impressive. But I think it does tell us something—just as in the earlier Commonwealth study, which found that American patients were more likely than their European counterparts to have designated a health care proxy, what we see here is that America does well on form. We don’t do as well on substance. That’s the disconnect we need to remedy.