March 29, 2006

Adding Life to Years

The latest statistics on life expectancy were released by the National Center for Health Statistics in January, 2006 and they show continued improvement in the prospects for old age (DL Hoyert, M. Heron, SL Murphy, and HJ Kung, “Deaths: Final Data for 2003,” National Vital Statistics Reports. National Center for Health Statistics, Hyattsville, MD, released January 19, 2006). Life-expectancy at birth is 77.5 years, as of 2003, the year on which the results are based, surpassing the previous record, from 2002, by 0.2 years. The age-adjusted death rate (that is, the death rate taking into consideration the distribution of ages in the population) is also at a record low, down 1.5% from 2002.

If we examine the statistics more closely, we discover several interesting observations. First, let’s look at life expectancy at age 65. Life-expectancy at birth reflects the risk of dying of congenital abnormalities or of the dangers that affect young adults, such as AIDS and homicide. For older people, what is of concern is how much time they are likely to have left if they have made it through middle age. The answer is that life-expectancy at age 65 is an impressive 18.4 years. It is even higher for white women (19.8 years) and a little lower for white men (16.9 years). What is disturbing, however, is how much lower life-expectancy at age 65 is for black men (14.9 years). Black women, by contrast, are just about exactly at the national average (18.5 years).

Since age 65 really is not considered “old” any more, what is of even greater interest is the data on life-expectancy at age 75. The 2003 data indicate that 75-year olds can expect, on average, to live for another 11.8 years. Again, the situation is a little better for white women (12.6 years) and a little worse for white men (10.5 years). And again, the life-expectancy for black men is unfortunately only 9.7 years, although black women are almost as well off as white women, in terms of years of life left (12.4 years compared to 12.6 years).

Another important result reported by the government is that while the age-adjusted death rate is down for several of the leading causes of death, including heart disease, cancer, and stroke, the rate of Alzheimer’s disease continues to rise. Regrettably, this means that as we make strides in the treatment of many of the diseases afflicting the elderly, the net result is often that people live long enough to develop and ultimately die of dementia.

What is not discussed in the government statistics is the rate of chronic disability in old age. While we all aspire to live longer, for many of us it is critical that those additional years be good ones. We want to live longer but we want to be as independent as possible. We don’t want to require the help of others for our most basic activities—dressing, bathing, eating, walking. It turns out that the studies that have been done analyzing disability rates show that from 1982 to 1999 (no more recent figures are available for the U.S.) disability declined: one report calculates that the decline from 1982 to 1989 was 0.26% per year, the decline from 1989 to 1994 was 0.38% per year, and the decline from 1994 to 1999 was 0.56% per year (Kenneth Manton and XiLiang Gu, “Changes in the Prevalence of Chronic Disability in the United States Black and Nonblack Population Above Age 65 from 1982 to 1999,” Proceedings of the National Academy of Science (2001); 98: 6354-6359). A second report, summarizing the results of various other studies, suggests that the rate of decline in disability among the elderly has been 1% per year for several decades (David Cutler, “Declining Disability Among the Elderly,” Health Affairs (2001); 20: 11-27).

But if we examine the data on chronic disability carefully, what we find is that the improvements are most pronounced for domains such as the ability to cook or shop or clean house. The fraction of the elderly population with severe disability remains virtually unchanged. Moreover, historical research indicates that it is hazardous to predict future rates of decline based on recent trends: in the past, sharp declines in disability rates have often been followed by stability or increases in disability. (See: Donald Redfoot and Sheel Pandya, “Before the Boom: Trends in Long-Term Supportive Services for Older Americans with Disabilities,” AARP Policy Institute, October, 2002.) Recent work on the potential long term outcome of the obesity epidemic, for example, predicts that even the gains in life expectancy may be reversed (see S. Jay Olshansky, Douglas Passaro, Ronald Hershow et al, “A Potential Decline in Life Expectancy in the United states in the 21st Century,” New England Journal of Medicine; 2005; 352: 1188-95).

What should we conclude from all of this? We can conclude that most of us who reach retirement age will live past 80. During much of our old age, we will be quite vigorous. But we are still at considerable risk for developing frailty (severe physical disability) and dementia (cognitive disability). Preventing these conditions, treating these conditions, and developing supportive care that will enable us to make the most of our old age despite these conditions, are the challenges that lie ahead.

March 14, 2006

Break a Leg?

Sometimes I wonder what disease will get me in the end: statistically speaking, it’s a toss up between cancer and heart disease. But I worry even more about breaking a hip. For many older people, a hip fracture is the beginning of the end. After sustaining a hip fracture, only 50-60% of people walk as well as they did previously. About 20% never walk again and 40% end up in a nursing home. Within a year, 37% of elderly Americans who break a hip are dead.

Not only are your odds of returning to your baseline level of functioning poor, but the chance of breaking a hip is substantial: 350,000 Americans break a hip each year. Older people are at risk of fractures because they fall and because they have thin (osteoporotic) bones. So articles in two major medical journals in the last month that call into question the efficacy of two leading preventive strategies—calcium and vitamin D on the one hand and hip protector pads on the other—are worrisome. Should all older people give up their calcium supplements and forget about wearing hip pads?

The study of calcium and vitamin D appeared in the New England Journal of Medicine (R. Jackson, A. Lacroix, M. Gass et al, “Calcium Plus Vitamin D Supplements and the Risk of Fractures,” NEJM 2006; 354:669-83). The authors had the opportunity to randomize over 36,000 women who had previously enrolled in a Women’s Health Initiative trial to receive either 1000 mg of calcium carbonate and 400 units of vitamin D or placebo. They were monitored for fractures for the next 7 years and their bone density was measured. The results? There was a small but not statistically significant decrease in the risk of hip fracture, a small improvement in bone density, and a small increased risk of kidney stones.

Before throwing away those calcium and vitamin D pills, it’s important to realize that only 59% of the women were actually taking the recommended dose at the end of the study. When the authors evaluated the outcomes in women who really took the medicine, there was a 29% statistically significant decrease in the likelihood of hip fracture. For scientific reasons, the analysis should be done the way the authors did it—using an “intent to treat” approach. This allows them to conclude that from a public health or policy perspective, they cannot recommend the use of calcium and vitamin D because prescribing those supplements is not likely to produce the desired outcome. But that’s different from concluding that taking the medication will not result in lowering your chance of hip fracture. It’s also important to realize that the 36,000 women in the study were healthy, community-dwelling post-menopausal women with an average age of 62. Inferring from this study that calcium and vitamin D do not prevent hip fractures in older women with other medical problems is not warranted. And in fact there was some suggestion that women over 60 did in fact benefit.

The second study on a related subject was a meta-analysis or pooled analysis of previous studies that was published in the British Medical Journal. (M. Parker, W. Gillespie and R. Gillespie, “Effectiveness of Hip Protection for Preventing Hip Fractures in Elderly People: Systematic Review,” BMJ 2006; 332: 571-4). It asked whether hip pads achieved their initial promise: earlier studies of nursing home patients had found that hip protectors could halve the rate of hip fractures. After combining the information from 14 studies, 11 of which involved nursing home patients and 3 of which were conducted in community-dwellers, the authors were unimpressed. They concluded that the pads were ineffective in the home setting and conferred only a small (but statistically significant) benefit in institutionalized individuals. But very much like the work on calcium, it turns out that many of the people who were supposed to wear the hip pads didn’t do so for much of the time. They found them inconvenient or awkward or uncomfortable, despite efforts to streamline the pads. Once again, it’s reasonable to conclude that hip pads should not at this point be the “standard of care.” We shouldn’t routinely expect insurance companies to pay for them or nursing homes to provide them. But there may well be a benefit for the person who is at high risk of falls and fractures and who is motivated to use them.

The data aren’t all in yet—we may find that calcium and vitamin D are effective, but higher doses are required, or it may turn out they are principally useful in an as yet undefined subgroup of the population. We may discover that certain types of hip protectors work and others, made of different materials, do not. In the meantime, these two interventions may make a lot of sense for people who want to do everything they can to avoid a hip fracture. And that’s not such a bad idea.

March 09, 2006

How Much will Medical Care Cost After Retirement?

Fidelity Investments just released a report estimating what the average 65-year old couple can expect to pay in medical expenses during their retirement (Reuters: “US Retirees’ Health Care to Cost $200,000-Fidelity”) Fidelity’s prediction is that a couple can anticipate spending $200,000 over the next 20 years. This is up from $190,000 in 2005 and $175,000 in 2004.

This figure—and clearly it’s just an estimate—is based on the expected cost of premiums for Medicare Part B and Medicare Part D, plus the cost of co-payments, deductibles, and out-of-pocket prescription drug costs. It does not include dental care, over-the-counter prescriptions, or long term care (ie nursing home costs).

It’s important to realize that this figure is based on the fact that the average life expectancy for a man at age 65 is 15 years and for a woman is 20 years. Actually, a 65-year old white man will, on average, live another 16.6 years; a 65 year old white woman will live 19.5 years; a 65 year old black man can expect to live another 14.6 years; and at age 65, a black woman will, on average, live for 18.3 years (see Health United States, 2005). Since these are averages, some people will live less long, and others a good deal longer.

Fidelity’s estimate assumes the couple will enroll in Medicare but have no supplementary employee health insurance coverage. Increasingly, corporations are discontinuing their medical care coverage for retirees in order to save money, so couples should not count on having such benefits. Out-of-pocket medical costs are the single largest expense for most retired couples, so start saving now!