December 21, 2015

Home Not Alone

Last June, Medicare announced preliminary results from its “Independent at Home” demonstration project. They showed the program provided high quality care and at the same time saved money. Not only that, but the population it served was the sickest of the sick and the frailest of the frail—people with multiple chronic conditions who needed help with personal care. But the reports in the news media didn’t explain what exactly the program did. Now, a study in the latest issue of Health Affairs explains how the program works and analyzes its achievements to date. 

The new study confirms the earlier reports: participants in the study had lower rates of acute hospitalization, which saved Medicare money. They also had a lower risk of nursing home admission, which saved Medicaid money. But the results so far—the analysis is based on data from the first year of a three-year pilot project—are modest. In one group of program participants, the hospitalization rate fell 14%; in a second group (the criteria for enrollment were somewhat different), the hospitalization rate fell only 1%. And understanding the analysis is almost impossible for the general reader. Not only were there two different sets of participants, with slightly different characteristics, but there were four different comparison groups. Then the authors invoked something called “entropy balancing.” The most it seems to me reasonable to conclude at this point in time is that the program isn’t hurting and it might be helping both patients and Medicare’s budget woes. But the benefits, if they exist at all, are likely to be small.

But what really struck me about the program is the description of what it actually involves—and what it doesn’t. Independent at Home does not support ongoing, comprehensive primary care provided in the home. It is nothing like a program reported on in the media at the same time that the Health Affairs article came out, a program on Long Island, New York that in fact moves all of health care into the home of its frail enrollees. That House Calls program sends doctors and nurses into the home and offers simple lab tests, basic x-rays and services such as physical therapy in the home. The doctor or nurse who visits the patient goes back again and again, providing continuity. The Independent at Home initiative, by contrast, provides a single home assessment each year that involves a history, physical exam, and lab tests, including a screen for depression, a review of medications, and a needs assessment. The upshot of this home evaluation is the development of a “plan of care” that is transmitted to the primary care physician for implementation. That plan might include referrals to social work or palliative care, but the individuals providing these services are not members of the primary team. When the patient develops an acute medical problem, nobody comes to the home to assess the patient--the primary care doctor is called. No medical personnel return to the home until it’s time for the next annual home assessment.

How can such a limited program possibly work? And if it does, what does that say about our usual way of providing care? It’s worth noting that the design of Independent at Home took into account the utter failure of an earlier demonstration program that tested the effectiveness of  “care coordination” and “disease management.” In this model, which had been instituted by 34 programs and which the Congressional Budget Office pronounced ineffective, patients were encouraged to participate actively in their own care and a case manager helped them negotiate the system and facilitated communication among all those involved. Perhaps not surprisingly, a system that relies on very frail older people, many of whom suffer from cognitive as well as physical impairments, to self-manage their care is fraught with peril, however effective such an approach might be in other patient populations. So the new model, the basis for the Independent at Home project, has as its centerpiece a plan of care and it’s the primary care physician, not the patient, who is supposed to implement the plan. It draws on studies such as the University of Pennsylvania’s transitional care program that found that a single home visit by a nurse  after a hospitalization could markedly decrease the risk of readmission to the hospital the following month.

Conceptually, the new model is an improvement over its care coordination/disease-management predecessor. It makes sense to determine whether anything short of comprehensive home-based primary care can provide high quality, cost-effective care, since the more all-encompassing home care won’t be cheap. But why place the responsibility for implementing the “care plan” squarely on the shoulders of the primary care physician, when we already know that most primary care physicians aren’t equipped to care for frail elders? Why rely on the primary care doctor to follow the plan when we already know from studies of geriatric consultation done years ago that when doctors receive recommendations for how to take care of patients that are based on unsolicited medical consultations, they ignore them? Why not focus on the caregiver as a crucial and untapped resource? 

The older patients who enroll in Independent at Home need help to get by day to day. Almost by definition, they have caregivers. Only when we involve the caregivers in medical care (and not just in providing personal care or crisis decision-making) will we be able to provide the kind of care that patients want and deserve. Yes, we will need a home visit to develop a plan—a plan which must reflect the values, preferences, and resources of the patient and caregiver. And yes, the physician needs to be involved in implementing the plan. But so, too, does the caregiver.

December 14, 2015

When I was a member of the Massachusetts Public Health Council, a policy-setting division of the Department of Public Health, I had to recuse myself every time we voted on a proposal in which I had a personal financial stake. In fact, I often couldn’t participate in the discussion or make a decision whenever there was an appearance of a conflict of interest, even when there really wasn’t a conflict at all. Massachusetts took this matter very seriously and had a number of lawyers scrutinize every council member and every vote to make sure we were scrupulously following the roles. So it came as quite a shock when I learned that roughly 30% of US senators and 20% of House members hold assets in biomedical and health care companies but are perfectly free to introduce, discuss, and vote on legislation that significantly impacts those companies.

According to a report by STATa new non-profit organization devoted to investigative journalism in the health care industry, our Congressional representatives are substantial investors in Pfizer, Johnson &Johnson, and Merck, along with assorted other companies. Their investments exceeded $68 million in 2014. The people who have the biggest stake often use their influence in Congress to promote legislation that just happens to coincide with their personal interest. Republican representative Chris Collins, for example, has co-sponsored a bill that would repeal the recently instituted tax on medical devices (which is one of the ways the Affordable Care Act pays for health insurance for the previously uninsured) and he just happens to be invested in the medical device industry, which vociferously opposes the tax. He also has supported abandoning FDA surveillance of products after they are on the market and just happens to be the single largest stockholder in a biotech company. This in the face of the recommendations of an independent Institute of Medicine report that the absence of any meaningful post-market surveillance is one of the weakest links in the FDA and has been responsible for delays in discovering drug toxicity that have led to numerous deaths.

It’s not just Republicans who influence health care policy despite their clear and unequivocal personal financial stake. Democrat Scott Peters has spearheaded an effort to essentially allow drug companies to prolong their already lengthy patent protection of drugs, the principal bulwark against price competition in the pharmaceutical industry. His wife is a major investor in the industry, buying over half a million dollars worth of stock in 2015 alone.

Even when they have what appears to be a conflict of interest, our congressional representatives sometimes vote against their own interests. STAT cites the case of Senator David Ritter who introduced a bill that would allow US consumers to buy medications from Canada, something that the pharmaceutical industry opposes. He has a modest personal investment (in the neighborhood of $100,000) in health care stocks. But by and large, those with the greatest influence in Congress also have the largest stakes. Collins, for instance, is a member of the House Energy and Commerce Committee—which is charged with overseeing the FDA.

Why is this a geriatric issue? Older  people are disproportionately large users of medications and devices. By way of example, more than 60% of all total knee replacements are inserted in people over age 65. The mean for pacemaker insertion is 75.5 and the mean age for ICD (implantable cardiac defibrillator) is 66.2. In the drug arena, data from 2010 showed that seniors account for 13% of the population but consumer 34% of all prescription drugs. So while conflict of interest in Washington affects us all, it affects older people most dramatically. And older people vote—they can vote out those who support their personal interests and not those of their constituents.


We've known about the perils of conflict of interest among legislators for some time, just as we've known about potential conflict of interest among physicians who prescribe drugs and devices for their patients. The widely accepted solution to the problem is disclosure. And in fact it is the federal disclosure requirement that generated the thousands of pages of congressional disclosure forms which STAT used to generate its report. Maybe this is just the first step: maybe disclosure, when analyzed the way STAT has done, will lead to shaming and behavioral change. More likely, it won’t lead to any meaningful change. Disclosure isn’t enough. Transparency isn’t enough. We need to change the rules of the game.                                                                                              

December 07, 2015

The Stealth Solution

After 22 posts about dementia in which I repeatedly inveigh against the all-too prevalent hype about how the cure for Alzheimer’s disease is just around the corner, I finally have some encouraging news. A piece in the “Medical News and Perspectives” section of the Journal of the American Medical Association highlights a radically new approach to preventing dementia.

At the heart of the approach is the recognition, as the JAMA article leads off by saying, that “pathology is not destiny.” We know that this is true in large part because of the nun study, an ingenious epidemiological study that’s been ongoing since 1986. In this study, 678 of the School Sisters of Notre Dame agreed to annual assessments of their cognitive and physical function as well as to blood tests for genetic studies. They also agreed to donate their brains to the researchers upon their deaths. Taken together, these tests allow researchers to uncover the relationship between cognitive function during life (the presence, absence or degree of dementia), biochemical markers in the blood, and pathologic changes in the brain.

The study population is particularly useful because the nuns tend to stay within their order—they don’t move around the way many Americans do—simplifying longitudinal follow-up. Moreover, because they all live in the same order, where they have a similar diet and participate in many of the same activities, they exhibit a degree of environmental homogeneity seldom found in other groups. One of the many things we have learned from the nun study is that there are people with evidence of advanced Alzheimer’s disease at autopsy who clinically had very mild disease and conversely, there are people whose brain pathology is consistent with early Alzheimer’s disease who were clinically quite severely affected. This observation strongly suggests that factors other than the brain plaques and tangles that are the hallmarks of Alzheimer’s can mediate the severity of a person’s symptoms.

One of these mediators is other brain pathology, such as vascular disease, which causes strokes. It turns out that vascular changes plus plaques and tangles are far more devastating than would be expected by simply summing the expected effects of each alone. The implication is that preventing atherosclerosis (what once upon a time was known as hardening of the arteries), which involves smoking cessation and treatment or prevention of high blood pressure, diabetes, and elevated cholesterol, is an effective means of attenuating dementia, even dementia due to Alzheimer’s disease. But another mediator is entirely new and different: recent results from the longitudinal Framingham Heart study point to special proteins that the brain manufactures to protect itself. Remarkably, a stimulus for the body to make more of these magic molecules is social engagement.

One such brain protein has the awkward name of brain-derived neurotrophic factor, abbreviated BDNF. This substance appears to reduce the risk of both dementia and stroke, probably by promoting the growth of new neurons and of synapses, the connections between neurons, as well as by protecting neurons from dying. Another protein, with the equally euphonious name of vascular endothelial growth factor, abbreviated VEGF, may also play a protective role. VEGF promotes the growth of new blood vessels, which may counteract the damage to the brain resulting from amyloid, a principal factor in the development of Alzheimer’s dementia. Just as social networks seem to boost BDNF, exercise seems to increase the levels of VEGF.


So far, no one has found a drug that either mimics the effects of BDNF or VEGF or stimulates the brain to make more of these substances. Or at least, the one such drug that has been studied, davunetide, failed to show any benefit in a clinical trial. It’s conceivable that a combination of pharmacology and life style interventions will be needed to make a clinical difference. But the idea of harnessing the body’s own repair mechanisms is promising; it seems much better than trying to remove amyloid deposits after they’ve formed (by which time the damage has already been done) or to boost levels of the neurotransmitters needed for neuron to neuron communication (which doesn’t work if the neurons that are supposed to communicate are already dead). 

If one way to coax the brain into producing more protective protein is by exercising and remaining socially engaged, both of which have other benefits and almost no side effects, that would be welcome news indeed.

December 02, 2015

A Biochemical Helmet?

At last, an approach to preventing dementia that just may work. See blog post next weekend.