CODI: Cornucopia of Disability Information

Aging and Disability: A Comparison of Health Related Changes

Aging and Disability:  A Comparison of Health Related Changes

Kenneth Brummel-Smith, M.D.
Director, Geriatric Evaluation and Management Program
Portland Veterans Affairs Medical Center
Associate Professor of Medicine and Family Medicine
Oregon Health Sciences University


My task today is to present some of the initial medical information from our
study, which has been overwhelming me in data. Given the volume of analyses
performed, I'm going limit what I present primarily to significant findings.
For those of you who aren't familiar with scientific presentations, I'll try
to describe the data in a way that is valuable to you. But also you need to
realize the limits of scientific information.  Even though we frequently
observe differences between people and groups, the only ones that we can have
confidence in are those that we consider statistically significant, or
approaching significance, based on a 5% to 10% or lower probability of
observing the same finding by chance alone. I am also going to present a
little information about why we embarked upon this study of aging and
disability, and try to separate out the two processes -- aging and
disability -- if possible.

Demographic Realities

As we all increasingly appreciate, the number of people in the United States
who are aging to advanced years is rising tremendously. Back in 1900, there
were fewer than 10 million people above the age of 55, but in 2040 there are
going to be almost 100 million people in this age group alone. (Remember,
there are a total of only 260 million people in the United States today.)
This estimate of the 55-and-over population uses conservative data from the
U.S. Bureau of the Census.  Recently Schneider at USC re-evaluated these
data, using a different predictive model, and showed that we are probably
underestimating the number of people in this older age category by as much as

Another thing to realize is that the older you get, the longer you are going
to live, not the opposite. Sounds funny, but the reality is that a male born
today has a life expectancy of 72-73 and a female, about 78 years. But if you
reach age 65, your life expectancy goes up an added 18 years for females and
12 years for males. One of the most common mistakes physicians make is to
think that people who are now 70, especially if they are male, don't have
very much more time to live anyway, so why pursue the same treatment that
they would for a younger person? Using age as a criterion for medical
treatment contradicts the reality that most older people, 70 and older, are
likely to be around even longer, so it pays to keep them in the best health

Another interesting aspect of aging is that older people tend to think of
their health in relatively positive terms. When asked in a survey
questionnaire to rate their health, the majority of older people say their
health is good, very good, or excellent. This is true for people with low
incomes as well as those with higher incomes. This information regarding
perceived health status needs to be communicated, both to the population at
large and especially to health-care providers, because, basically, older
people feel they are doing pretty well. There are, of course, problems to be
dealt with, but too often we key in on the negative aspects of aging.

On the other hand, with the increase in life expectancy, the amount of time
the older person is likely to live with a disability is also going up. A big
concept in geriatrics that was pushed hard ten years ago was the theory of
compression of morbidity; that is, with good health care, better general
nutrition, better exercise, and better health promotion activities, people
would continue to age without disability and more people would live to
advanced old age. In other words, we would compress morbidity (or illness)
into the very end of life, with maybe a year or hopefully even less of sick
time before people would die. Obviously, that would be a wonderful goal.
Unfortunately, it doesn't appear to be happening at the present time. Recent
data suggest that beyond a certain age, for many people living longer doesn't
necessarily mean living better or actively. In fact, the longer we live, the
greater the likelihood that we will spend an increasing percentage of our
older years living dependently rather than independently. So one of the
thrusts in both disability and aging research is to focus on what we can do
to push back chronic illness and disability so that people will live the bulk
of their lives independently.

In spite of older people's perceptions that their health is good to
excellent, we know from national surveys that the majority of people 60 years
and over report at least one major physical impairment (visual, hearing, or
orthopedic problems or cataracts) or chronic health problem (heart disease,
hypertension, emphysema, chronic bronchitis, diabetes, or arthritis). In
other words, chronic diseases predominate as we get older, and they
increasingly predominate as the older age group increases. So again the
thrust in our research is to look at the question: Do people who have
disabilities at a younger age have more chronic health problems, maybe as a
result of their disabilities, than their same-age counterparts with no
early-life onset of disability? These are some of the issues that I want to
talk about.

Clarifying Concepts in Aging and Disability

A concept that is being pushed right now in gerontology and geriatrics is the
difference between usual aging and successful aging. By usual aging, we mean
that the typical, or average, situation for an older person is to have one or
more of the chronic diseases or impairments that I just mentioned.  The
difficulty occurs when we try to separate out what is statistically common or
typical from what is physiologically normal. The tendency is to misinterpret
what is typical as what is normal, rather than interpreting it as a result of
an underlying disease that either has not expressed itself to the point that
it can be measured, or is just something that is accepted by our society as
being normal, such as walking more slowly. Successful aging is a term that is
used to describe ideal physiologic aging outside the realm of disease, and it
does appear that there are small numbers of people in our society who undergo
this form of aging. Obviously a lot of effort is going into studying what is
happening to those who avoid the most common kinds of changes that the great
majority encounter with age.

Another related concept in gerontology is the so-called "rule of thirds";
that is, what we think of as aging is really a combination of three things --
normal age-related changes plus the effects of disuse and the effects of
age-related diseases.  There is no doubt that normal physiologic changes
occur with age. Obviously someone who is 70 is not the same as someone who is
7. But what we are probably looking at more in terms of the cause of these
functional changes is the addition of disuse and disease.

What do we currently know about aging with a disability? The two major areas
that have received the most attention to date are polio and spinal cord
injury, although now we are starting to look much more intensively at people
who have cerebral palsy and muscular dystrophy. In terms of numbers, we
estimate that there are approximately 1.6 million survivors of polio in the
U.S. today and about 180,000 persons with spinal cord injury.  For cerebral
palsy, the estimate is 720,000.

But, despite the prevalence of people aging with life-long disabilities, we
have tremendous gaps in our knowledge. We don't really know the effect of
aging on people who have disabilities because, by and large, they haven't
been studied systematically, and when they have been studied, most research
has focused on one disability group by itself, not on comparative studies
that include nondisabled people of the same age. Comparative studies provide
the only way to look scientifically at the results of aging and separate
these out, hopefully, from the results of the disability. I do think,
however, that there are lessons that can be learned from applying the
research which has been done in gerontology, and which generally has received
more funding, to research questions associated with aging with a disability.
Conversely, disability studies can also contribute to our understanding of
aging by helping us look at the positive aspects of aging; and we do have
some data health-wise that look at that, too.

Results from the Rancho Study

As part of the health evaluation component of the Rancho Comparative Study of
Aging and Disability, we conducted a very comprehensive physical exam. This
included: a laboratory screen, which covered the primary kinds of tests that
are usually effective in detecting disease; an incentive spirometer, which is
a type of breathing test that measures the force that a person can breathe
out and has been shown to be a valid screening tool for lung function; a bone
density scan for osteoporosis (using dual energy x-ray absorptiometry), which
measures the amount of calcium that is in people's bones; and an
electrocardiogram (EKG) to look for undetected cardiac disease.

Comparison of Polio and Control Groups

Blood Pressure. First, we compared the blood pressure data of polio survivors
and nondisabled controls. From Figure 1,* we can see that most of our
subjects, in both the control and the polio groups, had a blood pressure
reading of less than 140 systolic, which is arbitrarily defined as being
normal. There was no difference between the two groups in the prevalence of
hypertension, which is defined as equal to or greater than 160. So it doesn't
appear that aging with polio negatively affects blood pressure as a health

Cholesterol. Next, we looked at cholesterol, and again found no statistical
difference between the polio and the control groups (see Figure 2). However,
there were a fair number of people, about 40% of each group, with cholesterol
levels above 240, which again is arbitrarily defined as being elevated.  This
raises an important question. How do we know something is normal or abnormal?
How do we make that determination? It's really just a statistical
representation of reality. In the case of cholesterol, we try to associate
high scores with an adverse event like a heart attack or a stroke; but, on
the other hand, for those consumers in the audience, you have to remember
that having a cholesterol level of 240 or higher doesn't necessarily mean you
will have a stroke. All it means is that you are in a risk group that has
been shown to have a higher rate of strokes; it says nothing about the
individual.  But in our sample we didn't see any difference between the two
groups, in the percentage of people with elevated cholesterol levels.

EKG Changes. In terms of changes of electrocardiogram results, we did observe
differences between the polio and control samples in the number of people who
had abnormal EKG changes.  Although the magnitude of the difference only
reflects a nonsignificant trend (p = .12), the incidence of abnormal EKG
results was actually higher among controls than it was among polio survivors.
In this context, abnormal EKG changes refer either to something we call
non-specific ST-T wave changes or to a right bundle branch block (RBBB);
these were the two most common changes, but a variety of others were also in
this grab bag. According to Figure 3, there is a trend suggesting that polio
subjects may actually have had fewer abnormal EKG changes compared to their
age-matched controls.

Thyroid Function. Another area where we started to see some potential
differences due to disability was in terms of thyroid function. As Figure 4
indicates, the percentage of individual TSH (thyroid stimulating hormone)
scores greater than or equal to 5.8, indicating hypothyroidism or low thyroid
function, was slightly greater for the polio group than for the controls.
Although this difference, again, only approaches statistical significance (at
a probability level of p = .10), it is an important trend which I will
emphasize later in the presentation.

Differences Within Polio Sample

Blood Pressure.  Turning to subgroup analyses within the polio sample, our
research indicates that most survivors had average to low systolic (ó 159)
and average to low diastolic (ó 80) blood pressure scores. Only a small
percentage (about 20%) had combined high blood pressure scores of 160/90 or
above. (This was not statistically different from our sample of controls, and
reflects a lower percentage of high blood pressure than is reported for the
U.S. population at large.) Interestingly, however, we did observe significant
differences in blood pressure by age. Contrary to conventional wisdom, the
youngest polio survivors in our sample, those 50 to 59, had a significantly
higher incidence of elevated blood pressure compared to their older
counterparts, who are 60 to 69 and 70 and over.  While we are unable to
explain this finding on the basis of one study, it does emphasize the
importance of blood pressure checks among young post-polio survivors.

Cholesterol.  We also observed significant sex differences in total
cholesterol levels within the polio sample: female survivors had more scores
in the 240-and-over range compared to their male counterparts. This is an
important finding because it used to be thought that high cholesterol in
females didn't mean anything. But now we know that, with age, the female rate
for heart disease continues to rise, and that lowering cholesterol probably
has a benefit in this group, especially for those with above 240 levels. This
might be something to consider in caring for women who have had polio.

Why male polio survivors had significantly lower cholesterol levels than
females possibly has to do with the fact that males, in general, tend to get
more medical attention.  Numerous recent studies indicate that less attention
has been paid to women's health issues, in general, especially where
preventive treatment is concerned. Women are also offered less cardiac
surgery and receive fewer admissions to cardiac care units after heart

EKG and TSH. Again, within the polio sample, we also observed significant sex
and age differences in EKG (heart) rhythm disturbances and TSH. More heart
rhythm disturbances occurred among males than among females and, not
surprisingly, there were more abnormalities of heart rate rhythm in the older
group. This last finding makes sense, since as you age, the likelihood that
there will be some cardiac problem also increases.

Females were the only group to have a high TSH score (ò 5.8), indicating
hypothyroidism. This finding emphasizes the importance of thyroid tests for
women who have had polio, because it is possible that the symptoms of
hypothyroidism may often be misinterpreted as due to age. Feeling tired, not
having any energy, poor concentration, and cold intolerance (symptoms) are
often either written off as being due to post-polio syndrome (PPS) or to
normal aging, both of which are typically not treated as medical problems.

Glucose.  Another interesting finding was with regard to blood sugar and
post-polio syndrome (PPS). Most people in this study who reported symptoms of
PPS had low or normal glucose levels, and we didn't have many people who were
diabetics.  Interestingly, the number of people with slightly elevated
glucose levels was higher among non-symptomatic polio survivors. Another way
of looking at this relationship is in terms of severity of initial
impairment. Those with more limbs affected at acute onset also had lower
glucose levels. Together, these results suggest a curious finding: that
either having PPS or having more limbs affected may decrease the chance of

To summarize this first set of analyses, for the most part there are no
significant differences in basic health indicators between polio survivors
and controls. Contrary to expectations, there weren't any differences in the
pattern of age-related changes between the two groups. However, we did
observe two noteworthy trends that approach statistical significance: (1)
post-polio survivors had lower cholesterol levels and fewer EKG changes, but
(2) they did have higher rates of hypothyroidism. So from a purely medical
standpoint the positive effects of polio appear to be lower cholesterol and
fewer EKG changes.

Comparison of Stroke and Control Groups

First, some general observations on our stroke sample. About one-third of the
sample had a stroke on the left side of their brain (L-CVA), affecting the
right side of their body. The other 68% had strokes on the right side of the
brain (R-CVA).  The predominance of participants with R-CVA's is probably an
artifact of our sample selection criteria, which eliminated individuals with
moderate to severe levels of cognitive impairment. Second, our stroke sample
had a very high rate of hypothyroidism, with 13% having elevated TSH levels
greater than or equal to 5.8. This is well above the national average, which
places the prevalence of hypothyroidism between 0.7% and 1.2% of the U.S.
population. Even among older persons being admitted to the hospital, the rate
of elevated TSH levels is still lower at 7% than was observed in our stroke

Third, about 23% of the stroke sample had evidence of a tuberculosis
infection, which is about the same as in the national population; but a
significant number had not been exposed to mumps; or, more likely, their
immune systems were somewhat depressed. This is an important finding and one
that is consistent with other research in geriatrics; namely, that aging is
associated with a depression of the immune system. In our particular sample,
although most stroke survivors over the age of 50 have had mumps, they failed
to respond positively to the mumps skin test. This indicates that
experiencing a CVA is probably associated with changes in one's immune
system, a finding that is not surprising if we remember that the immune
system is controlled by the brain, not by our lymph glands.

One criticism of the Rancho study is that we had considerably more male
stroke survivors than females (43 vs. 17). This is a statistically
significant difference, and could be explained two different ways. One is
that older male survivors of stroke tend to have female companions or
partners who are willing to bring them in to a research study, whereas older
female stroke survivors are less likely to be living with partners. And this
is true with aging, so it might confound our results that way.

The other possible explanation is that more males have stroke than females,
and this is also true. However, there are important gender differences by age
of onset. If you focus only on those above the age of 70, the occurrence of
stroke is almost equal by gender; but among the younger age groups, those
between 55 and 70, the incidence of stroke is much higher among males.

Blood Pressure and Cholesterol. Turning to comparative data, Figure 5
presents a surprising but encouraging finding regarding blood pressure. In
our sample, the percentage of stroke survivors with diastolic blood pressure
scores within normal limits (ó 80) is significantly larger than that
observed for controls (83% vs. 55%). This same result is also reflected in
tests of mean difference, which indicate that the average diastolic blood
pressure of 76.5 for the stroke sample is significantly lower than that
observed for controls with a mean of 82.2. This finding emphasizes the
importance of blood pressure treatment post-stroke. Interestingly, the
distribution of cholesterol scores (no figure presented) also favors the
stroke sample. Again, it is stroke survivors who are disproportionately
represented among those with low cholesterol scores (37% vs. 17%), while the
incidence of high cholesterol (ò 240) is greater among controls (20% vs.

Taken together, I hope these two findings indicate that after stroke people
do pay more attention both to their diet and to cholesterol treatment to
reduce their further risk.

Peak Flow. Peak flow, which is a measurement of lung function, was
significantly lower among both male and female stroke patients than among
age- and sex-matched controls. This result makes sense, knowing what we know
about the effect of stroke on respiratory muscles. It also emphasizes the
importance of good preventive care and of not smoking. In addition to mean
differences between groups, in both stroke and control samples there were
significant age differences in peak flow, with younger participants having
better lung capacity. However, as Figure 6 illustrates, the negative effects
of age on lung capacity were greater for the stroke sample. In other words,
there is a risk of reduced lung function associated both with having a stroke
and, in general, with getting older. In our study the group of stroke
survivors 70 years and older had the lowest peak flow of all, and so they are
theoretically at the highest risk for lung/pulmonary complications.

TSH. We also observed significant between-sample differences in thyroid
functioning. As Figure 7 shows, mean levels of TSH vary more by gender than
by sample designation. For males, TSH scores were almost identical for both
the stroke sample and controls, whereas female stroke survivors had
significantly higher TSH scores compared to their non-disabled counterparts
(mean = 3.8 vs. 2.4). Again, this emphasizes that women who have had strokes
should have their thyroid checked, because the symptoms of hypothyroidism --
tiredness, weakness, lack of energy, poor concentration -- are often
misinterpreted. These symptoms may be totally unrelated to aging or to
stroke; and hypothyroidism is a very easily treatable problem.

Finally, we considered smoking. We are pleased to report that most of our
subjects, almost 90%, either never smoked or stopped smoking, which is
fabulous. Only 10% still continue smoking.

For the future, we need longitudinal studies. The problem with our study is
that it was a one-place-in-time study. It concerned only five years in the
lives of our subjects. What we need in disability research is the same type
of financial incentive the government has given, for instance, to the
Baltimore Longitudinal Study of Normal Aging, to conduct a 45-year
investigation following people with disabilities throughout their whole
lives. From a medical research perspective, what we need is to identify
people with early- to mid-life onset of disability as soon as possible,
follow them for the rest of their lives, and compare them across a broad
range of health indicators with a group who don't have life-long

We also need to add the next level of research, which is stressing people.
The tests I reported on were all done with subjects sitting down, which gives
only basic information.  What really determines how people function
physically is to study them under physiological stress. So exercising them,
doing tests of breathing over time, those kinds of tests, which are much more
expensive, would give us different information. We also need to look more at
diet. Our study didn't include diet, which I think is an important aspect to
study, especially regarding cholesterol.

Finally, a lot of you have had difficulty working with physicians. Part of
the problem is that they don't have this information on aging with disability
because it is brand new; and part of the problem is they aren't oriented
toward rehabilitation or helping people who have disabilities. I like Ambrose
Bierce's quote: "A physician is one upon whom we set our hopes when ill and
our dogs when well." Most physicians have no access to the kinds of data I
presented to you today.  Your advocacy and assertiveness in getting
physicians to learn more about disability, both in aging and otherwise, is a
very important part of successful health care reform. Thank you.


Questions and Answers

Questions for Dr. Margaret Campbell (Plenary #1) and Dr.
Brummel-Smith were submitted in writing from the audience.

Question:  Does glucose intolerance usually go up with age?  

Brummel-Smith:  Glucose intolerance refers to the gradual rise
in blood sugar with age. Glucose rises gradually with age but
still stays within what is called normal limits. 

      For instance, the vast majority of people under 55 don't
have diabetes, and obviously have blood glucose below 110. But
with age, especially above age 75, we usually have blood
glucose up to 140. Anybody with a reading above 140 is still
considered to be diabetic, even at age 74. The reason for the
rise in glucose levels with age is that there appears to be a
change in the way insulin functions in the body with age and
that people don't respond as well to the same circulating
level of insulin as they did when they were younger.  

Question:  Did you find a correlation between acceptance of
disability and depression in the polio group? If so, what was
the correlation?  

Campbell:  Yes, we did. They're inversely related, as you
might expect. In general, the higher the mean score on
depression, the lower the acceptance of disability. However,
we find significant differences in the strength of this
(negative) relationship within the polio group by sex, by age
of onset, and by historical period of onset. There were also
significant differences between the polio and the non-disabled
control sample in scores on a standardized measure of

Two related questions:  How can you tell if you are
experiencing normal aging or post-polio syndrome? How does one
identify pain as a result of post-polio syndrome, arthritis,
or the aging process?  

Brummel-Smith:  I want to shift some of this to Dr.
(Frederick) Maynard because he is going to speak more
specifically about post-polio syndrome, and I was looking more
at general health effects. But one of the things that you
should be aware of is that there is no pain normally with age.

      In general, you should always start with the presumption
that the symptoms are not due to post-polio syndrome. There
are a number of other conditions with which it may be
confused. A comprehensive evaluation is needed by someone who
knows about polio, preferably not just by a physician, but
also by a physical therapist or some other allied health
therapist. Something that is often missed in the definition of
post-polio syndrome is the criterion of "no other explanation"
for these symptoms. So the most important thing is to look for
other possible explanations. In general, I would say that
post-polio syndrome is similar to the rule of thirds, that it
has to do with aging, with specific changes associated with
having had the disease, and with some of the aspects of disuse
or misuse. 

Campbell:  Lots of good questions. I'm going to try to put
some of these together. A couple of them are on the general
outline of the Comparative Study of Aging and Disability
itself. Again, I will refer you to your syllabus and the
tables that are there. In a nutshell, we had 120 polio
survivors as participants in our study and 60, or half that
number, of stroke survivors. We recruited our subjects using
two strategies. The first was general solicitation, relying
primarily on newspaper advertising, and the second was
recruiting from stroke or polio support groups.  

      Next are questions on how we measured acceptance of
disability and how that differs from adjustment. Adjustment is
a much broader concept and could be measured with a variety of
indicators, including marital 

satisfaction, marital happiness, and life satisfaction; you
could also use measures of coping. The scale I was referring
to contains 50 items and purports to measure acceptance of
disability based on how people evaluate the impact of
disability in their lives. 

Brummel-Smith:  On the last question about post-polio
syndrome, I neglected to answer one other question:  "Is it
possible for someone to have polio but not develop post-polio
syndrome?" Absolutely.  But some people are non-symptomatic
for PPS; why, we don't know, and that is a very interesting
subject for further research. 

      There are a bunch of questions about TSH, which is
thyroid stimulating hormone. When your thyroid starts to go
out, the pituitary gland in the brain produces a hormone
called TSH, trying to get the thyroid going again. But if the
problem is that the thyroid is out, it won't go, and so the
TSH just keeps rising. So nowadays we use the blood test TSH
to screen for low thyroid.  

      Another question:  "I have had one doctor say thyroid
replacement medication was OK and another say it was not OK;
but in each case the doctors used different blood tests. Which
is the better blood test, and should retesting be done?" In
general, TSH testing is the best blood test for thyroid
function. However, there is another test called the sensitive
TSH, which is the very best. It should be retested every six
months if it is slightly elevated because there may be a
progression into worsening of the thyroid function. That's why
I promote people taking thyroid medication when needed. But
they should take it under a doctor's supervision and take the
lowest dose that 
returns their TSH to a normal level. 

      "What does our study mean with reference to TSH levels
in women over 45?" If you have had polio, and especially if
you have had stroke, having your TSH measured as part of 


an annual evaluation would be a good idea. You have to
remember that in studies on health promotion, people with
disability are excluded because the purpose is to look at
"normal aging." These studies have clearly shown that doing
TSH testing is not necessary, and I don't think you should ask
for that as a routine screen. But if you've had polio or a
stroke, you should.  

Campbell:  I'm going to try to pull four questions together
quickly. First, we need more people to be involved in our
study. If we get refunded -- our renewal application is pending
right now -- we will continue doing this research on both stroke
and polio survivors, so please stay tuned. Second, "For
Hillary Rodham Clinton's task force on health insurance, what
do you suggest I give them on aging with a disability?" To
copy Ken, tell them about the Rancho study. Moreover, I would
give them some of the data that Ken has presented. 

      Third, "When does aging with a disability begin?" Well,
it's a life course process. There is evidence that in terms of
osteoporosis, which is bone thinning, for those aging with
spinal cord injury it starts within a year to two. All you
polio survivors in the audience know about the 30-year
problem. We have used 30 years as the guideline for monitoring
when post-polio syndrome or later-life effects begin. For
stroke, it's a shorter interval, and we really don't know when
aging with stroke begins, because increased survivorship from
stroke is so recent. We have the fewest studies on aging with
stroke. But in terms of what to tell Hillary, we obviously
need to look at function, not only physical function but
social function. We also need to look at changes in function
over time and how people with disabilities integrate into the
rest of society; here we need to identify the barriers to
increased independence and community integration.