Aging with a Disability: A Life Course Perspective
Margaret L. Campbell, Ph.D.
Rehabilitation Research and Training Center on Aging
Rancho Los Amigos Medical Center
My role today, in the next 25 minutes, is to set the stage conceptually for
the presentations that are to follow. I'll begin by explaining why we are
interested in studying aging with disability and then go on to evaluate how
we've been approaching this important subject.
Aging with Disability -- A Hot Topic
First, aging with disability has become a hot topic. In fact, you all have
become a hot topic. The study of aging with disability has become a hot topic
in large part for two reasons. The first has to do with the changing
demographics of the life course; in particular, with increased life
expectancy and increased survivorship. Not only are we living longer, but
larger numbers of people, including persons with disabilities, are surviving
longer at all ages.
This affects two groups of people with disabilities -- those who are aging
with life-long or early-onset disabilities, such as polio, cerebral palsy,
and multiple sclerosis, and those who are surviving with mid- to later-life
onset of disability, such as stroke. If you put both groups together, the
estimates range widely from five to ten million people alive today in the
United States who are aging with disability. This is a large group, and it
promises to get even larger as the 60 and older population increases in
absolute size and proportion with the aging of the baby boom generation.
The second factor promoting interest in aging with disability has to do with
the rising costs of health care. To illustrate, in the last three decades the
United States has experienced a sevenfold increase in medical expenditures.
In 1992 alone total health care costs amounted to 14% of our gross domestic
product (or $840 billion), with a disproportionate percentage of expenditures
for Medicare and Medicaid going to the aged and disabled. In the future,
health expenditures could soar even higher under the demographic pressures of
the aging baby boom generation and increased survivorship among persons with
If you put these two trends together -- increased longevity and survivorship
and rising health care costs -- a lot of people, including consumers, policy
makers, and health professionals, are motivated to learn more about aging
with disability. The primary goal of all our studies is to better understand
how to intervene to promote independence and prevent some of the secondary
conditions associated with aging with disability that are so costly in both
human and monetary terms.
That is why we study aging with disability; but what do we actually know
about this important topic? I suggest that we know considerably more about
the late effects of polio than we do about the secondary complications of
stroke. In large part, this is because long-term survivorship from stroke is
an even more recent phenomenon than it is from polio.
But this is only part of the explanation for our lack of knowledge. Another
factor has to do with the models we have used to guide our study of
disability. For the most part, the prevailing models have not incorporated
"aging" or changes over time as a salient feature of the disability
process. In fact, I would assert that the major challenge to studying aging
with disability is a conceptual one. It involves merging the concepts and
theories of aging and adult development from social gerontology with the
definitions and measures of disability from the medical and rehabilitation
fields. Essentially, it involves integrating multiple disciplines together
under a comprehensive conceptual framework that can guide our investigations.
Approaches to Studying Aging with Disability
I will label these two approaches to studying aging with disability the
"aging and adult development" and the "rehabilitation" perspectives. To
contrast these approaches, studying disability from an aging perspective
involves the central concepts of development and change over time. Those who
use this perspective spend a lot of time talking about change, the factors
that bring about change, and the implications of change for psychological
well-being and life satisfaction -- to name just two outcome measures. From
the disability and rehabilitation perspective, the central concept has become
function and how to measure, maintain, or improve it. This second perspective
is also interested in life satisfaction, but the degree of functional
limitations -- both physical and social -- constitutes the primary outcome
variable of interest. My goal is to show how we can integrate these two
perspectives to advance our understanding of aging with disability. I'd like
to begin by describing two of the models that are available in the
rehabilitation field and then talk about the model that we have built for the
Comparative Study of Aging and Disability at Rancho Los Amigos Medical
The Rehabilitation Perspective
Figure 1* contains the first schematic diagram, which Dr. Maynard will
certainly recognize. It represents the Institute of Medicine's Model of
Disability. This model builds on the World Health Organization's
International Classification of Impairment, Disability and Handicap (1980),
depicted in the center of Figure 1. According to this classification scheme,
the disabling process involves four levels: (1) the pathophysiological level,
which is defined as interruption of bodily processes at the tissue level; (2)
the impairment level, which is defined as loss or abnormality of structure or
function at the organ level; (3) functional limitations, which refers to
restriction or lack of ability to perform physical action in the manner
considered normal; and (4) disability, which is defined as inability or
limitation in performing socially defined activities and roles.
The model in Figure 1 represents an advance over the traditional medical
model of disability for a couple of reasons. First, it looks at disability as
a process that focuses on function rather than just impairment and places
functional abilities and limitations within a larger context of both risk
factors and psycho-social outcomes. Second, it is an interactional model, as
indicated by the bi-directional arrows linking together risk factors and the
disabling process. Viewed this way, impairment and functional limitations are
seen as both the result of risk factors and as contributors to the increased
risk of new health problems. Despite these advantages, the Institute of
Medicine's model still contains problems. Chief among them is that aging, or
change over time, is not included as part of the disability process. The net
result is a static model.
The second, and more recent, model is contained in Figure 2. This model has
been proposed by the newly established National Center for Medical
Rehabilitation Research (NCMRR), which is under the umbrella of the National
Institutes of Health. The architects of the NCMRR model readily acknowledge
that this may not be the Center's final model. As we learn more about the
long-term effects of chronic disability, we will, hopefully, change our
models to reflect these new discoveries. For obvious reasons, we refer to
this second model as the egg.
The NCMRR model represents an advance over the previous model I presented
because it is more dynamic, which is what the egg shape is attempting to
communicate. In this model, disability is viewed as a nonlinear process
depicted by the pentagon in the center of the large oval. It also includes
risk factors, or influencing factors, and quality of life outcomes.
Another innovative feature of Figure 2 is that the scope of what is
considered an outcome of the disability and rehabilitation process has been
expanded to incorporate more quality of life measures from the Independent
Living Movement. In addition to functional status, this includes other life
domains such as self-determination, productivity, and consumer assertiveness
or advocacy. At the bottom of the NCMRR model is a line, labeled Time, which
represents the passage of time. This constitutes a third advantage of the
NCMRR model over the previous model of disability.
Despite these advances, my criticism of the NCMRR model is that time,
although included, is treated as a single or one-dimensional phenomenon. This
approach violates what we know about the aging process from a social
gerontology perspective. Viewed from a life course perspective, aging is
described as the result of changes associated with the multiple meaning of
time -- namely, chronological time, social time, and historical time. The
pace and process of aging is seen as a complex interaction of these three
dimensions of time.
To remedy the omissions of previous models, the Comparative Study has
developed a new framework which integrates concepts from both the adult
development and the rehabilitation perspectives. We refer to the Rancho
Model, depicted in Figure 3, as "Aging with a Disability -- A Life Course
Perspective." The first thing you will notice about our model is that it is
more complicated than the previous models of disability. I would suggest that
this is because aging with disability is a more complicated process than
either of its component processes taken alone.
The Life Course Perspective
What are the benefits of looking at disability from a life course
perspective? The most important benefit is that our understanding of the
long-term effects of disability is enhanced through incorporating the
multiple dimensions of time or aging. Let me trace out some of the features
of our Rancho model.
First, for persons with a disability, individual aging, as indicated along
the bottom axis of Figure 3, is made up of three different components. For
all of us it includes chronological age, which is represented by the diagonal
arrow, showing aging from birth to death. The second component consists of
social aging. This includes all those life course transitions we make from
childhood to adolescence to adulthood to midlife, to retirement and old age.
Each of these social transitions is further associated with a set of
age-graded developmental tasks. Examples include separation and
individuation, identity formation, attachment, maturity, retirement, and
For persons with disability, there is a third component of individual aging
which affects the timing and structure of their life course. Jessica Scheer
of the National Rehabilitation Hospital in Washington D.C. refers to this
component of individual aging as the "polio trajectory" (1991). For our
purpose, we have renamed it the "disability timeline" to apply to both polio
and stroke populations. This timeline refers to the disability-related events
of acute onset of disability, initial rehabilitation and recovery, a period
of stability, and perhaps a period of decline followed by the potential onset
of secondary complications of disability. The disability timeline is
superimposed on the other two components of individual time to define the
individual aging process for persons with disability.
A second major dimension of time that shapes the life course is "historical
period." Historical period refers to the social and cultural context in which
a person is born, acquires a disability, and ages. This dimension of time is
indicated on the vertical axis of Figure 3. To illustrate the significance of
historical context, Figure 3 identifies some of the key decades associated
with polio and stroke development. For example, 1940 marked a watershed
period for polio survivors. Prior to this time few, if any, people survived
with severe respiratory complications. Before 1940, we did not have the
technology, the ventilators, and the respiratory aids to adequately help
those individuals. Also 1948 to 1955, the peak years of the epidemic,
represents a significant historical period for polio survivors which, as I
will argue later, is positively correlated with the severity of polio
For stroke survivors, the advent of diagnostic- related groups (DRGs) in the
1980s marks a major watershed period. People who had their strokes prior to
the introduction of DRGs received, on average, more rehabilitation over a
longer period of time compared to people who had strokes after the advent of
this cost-control practice. In sum, historical periods are important because
they shape our philosophies about rehabilitation, affect the technology and
medical treatments that are available, influence social policies, and
essentially shape the social and cultural opportunities available to all of
us to maximize our potential.
Figure 4 illustrates the theoretical implications of disability when it
occurs at different stages of the life course. The three stages associated
with early-life onset of disability include infancy and early childhood, late
childhood, and adolescence. The differential impact of the timing of
disability is examined for three different areas of human development: family
relations, education and employment opportunities, and personal development.
Differences in the age of onset have tremendous implications for the meaning
and consequences of disability. For example, when the onset of polio occurred
in late childhood or adolescence, people experienced interruptions in their
education that affected their preparation for careers and their achievement
of life goals.
Differences in the timing of onset of later-life disability can also result
in significant differences in life consequences and adjustment. For example,
what are the implications of having a stroke when you are still in midlife,
age 40-60, versus having a stroke when you are in the post-retirement years,
after age 65? Again, Figure 4 traces out some of the theoretical implications
of differences in the timing of onset of stroke for three different areas or
Findings from the Comparative Study
Now, I'd like to shift our attention from theoretical models to data to show
you a sample of some of the results of our Comparative Study.* The question
we are asking here is: What do we gain empirically by adopting a life course
approach to aging with disability? First, I want to apologize. I will be
presenting more data on the polio sample than on the stroke sample. This
discrepancy stems from the fact that we started the polio sample earlier,
finished data collection sooner, and therefore had more time to analyze the
data. In the four months remaining in the study, we'll be busy completing the
Polio Sample: Effects of Timing and Severity of Impairment
Figure 5a examines the interrelationship between age of acute onset of polio
and severity of initial impairment. In this analysis we followed Dr. Lauro
Halstead's lead and conceptualized severity of initial impairment in terms of
the number of limbs affected at acute onset (Halstead, 1985). The categories
are: 1 limb, 2 limbs, and 3 & 4 limbs; "affected" refers to muscle weakness,
partial paralysis, or complete paralysis. The 1 limb category also includes
five people for whom only their trunk muscles were affected. "Age of onset"
is measured by a four-level variable: 0 to 2 years, 3 to 9, 10 to 19, and 20
to 39 years.
The pattern in Figure 5a is very clear. The percentage of people with only
one limb affected goes down dramatically as age of onset goes up, while the
percentage of people with three and four limbs affected is positively related
to an increase in age of onset. This positive relationship between age of
onset and severity of initial impairment is similar to the pattern Halstead
detected in his data. This pattern was also replicated in analyses we
performed with data from Richard Daggett's 1986 Family Circle magazine survey
of polio survivors. Although the underlying explanation for this relationship
is complicated, the interpretation is clear: the age at which you had polio
makes a difference in how severely impaired you were.
*See Appendix 2 for a summary of the socio-demographic characteristics of the
polio, stroke, and non-disabled samples of the Comparative Study.
Looking at the timing of polio another way, in terms of the historical period
in which it occurred, we see a similar pattern. In Figure 5b we have grouped
together people who had polio early in the century, up to 1929, versus those
who had polio in the 1930s, the 1940s, and the 1950s. Hopefully, you can
locate yourself in one of these historical periods. Again, the percentage of
people with only one limb affected at onset goes down dramatically across the
century, while the percentage of people with three or four limbs affected
goes up significantly. These data are summarized at the top of the graph.
Prior to 1940, only 27% of our sample had three or four limbs affected,
compared to 45% after 1940. Again, we have strong evidence that the timing of
polio affects the severity of initial impairment.
Unfortunately, we were not able to find a significant direct relationship
between the timing of acute onset of polio and the presence of post-polio
syndrome some 37 to 80 years later. We did, however, detect a very strong
relationship between severity of initial impairment and post-polio syndrome
(see Figure 6). Again we followed Dr. Halstead's lead and defined post-polio
syndrome (PPS) in terms of the dual criteria of: (1) the onset of new
symptoms of muscle weakness, fatigue, and pain, and (2) new functional losses
in activities of daily living (e.g., walking, climbing stairs, transferring,
etc). In addition to persons meeting these dual criteria, our approach
identified polio survivors who met neither criterion and those who reported
new symptoms only but no new functional losses (the "maybe" category).
As Figure 6 demonstrates, the percentage of people who met the criteria of
PPS in 1990 is strongly related to the number of limbs affected at acute
onset. The greater the severity of initial impairment, the greater the
likelihood of experiencing PPS an average of 45 to 50 years post onset.
The next figure rules out the simple explanation of "premature aging" as a
causal factor in PPS. As Figure 7 indicates, the percentage of people who
meet the criteria for PPS is not related to chronological age. There is no
significant difference in average age between symptomatic and nonsymptomatic
polio survivors. In fact, there is a slight trend for those without PPS to be
slightly older than those who report new symptoms and new functional
limitations. Similarly, the duration of disability does not differ between
the two groups. On the basis of our data, therefore, we conclude that the
simple passage of time is not an adequate explanation to account for the
development of PPS.
Polio Sample: Consequences for Psychological Well-being
Continuing this line of analysis, we now ask the important "So what?"
question. Specifically, we are interested in what implications timing of
acute onset and severity of initial impairment have for the psychological
well-being of polio survivors four to eight decades later. To address this
issue, we examined mean scores on a standardized measure of depression by
both age of onset and the historical period in which polio occurred to ask:
Does timing of acute onset of affect current psychological well-being? The
answer is yes.
As Figure 8 indicates, the mean score on depression varies by both the
historical period in which polio occurred and the age at onset. However, it
is important to point out that we are not dealing with a group of clinically
depressed people. In fact, the mean scores for all six groups of polio
survivors fall significantly below 15, which is the cut-off for clinical
depression. There is, however, one group whose mean score does cross the
threshold for clinically significant depressive symptoms (i.e., mean ò 8).
This group includes people who had polio after 1940 and at an early age in
the life cycle.
Who are these individuals and why do they have significantly higher
depression scores? I suggest that the answer to this question lies in
applying our life course perspective. Those individuals who had polio early
in life and later in the epidemic were, on average, as we saw in Figures 5a
and 5b, more severely impaired than their counterparts who had polio earlier
in the century and at an older age.
Moreover, because of the previously detected relationship between initial
severity and later-life effects, we can infer that persons in this group are
now experiencing the debilitating new symptoms and functional losses of PPS
at an earlier stage in the life cycle, during their 50s, when the competing
demands of work and family are the greatest. We refer to this group as the
"sandwiched generation," because they are squeezed between caring for aging
parents and/or maintaining careers and launching the next generation of
children. Viewed from this life course perspective, it is not surprising to
discover that this group of survivors is most vulnerable to the negative
psychological consequences of aging with polio.
Stroke Sample: Effects of Timing on Psychological Well-being
Moving on, we examined two sets of results from our stroke sample. Here we
asked a similar question: Does the timing of acute onset of stroke affect the
current level of psychological well-being? Again, we used the same
standardized measure of depression. Age of onset for the stroke sample is
defined in terms of three levels: (1) people 47 to 60, whom we refer to as
the early-onset or pre-retirement group; (2) people 61 to 69, the retirement
group; and (3) people 70 and older, the late-onset or post-retirement group.
We also distinguish between people who had left CVAs and right CVAs affecting
different sides of their bodies.
Using the life course approach also yields a couple of noteworthy findings
for the stroke sample. First, Figure 9 reveals that the mean scores on
depression are considerably higher for our stroke sub-samples than we
observed for polio survivors (overall mean score of 9.7 vs. 5.5). But again,
we are not talking about a group of clinically depressed individuals.
However, we do observe three groups of stroke survivors, compared to only one
group of polio survivors, whose mean levels of depression reach or exceed the
cut-off for clinically significant symptoms. Those with the highest
depression scores were people whose first stroke occurred early in their 50s
or in their 60s. Again, I suggest that the key to understanding this pattern
lies in examining the differential impact of stroke for individuals and
families when it occurs at different stages of the life course.
Finally, in Figure 10 we examine the effect of the timing of stroke on
another standardized measure of psychological well-being -- namely,
acceptance of disability. In our sample, with an overall mean score of 192,
acceptance of disability for stroke survivors is considerably lower than the
population mean of 264, based on a large sample of persons with disabilities.
The average level of acceptance of disability is also significantly lower for
the stroke sample than we observed for the polio sample with a mean of 246.
However, within this pattern, acceptance of disability varies significantly
by age of onset of stroke and by sex.
Females, on average, report higher levels of acceptance of disability than do
males (mean scores = 206 vs. 175), but among female stroke survivors the
level of acceptance varies significantly by age of onset. As Figure 10
indicates, women in our sample who had their first disabling stroke in their
50s report a mean acceptance score of only 153 compared to a mean score of
222 for those whose age of onset was between 61 and 69. Interestingly, we
observe no such interaction between age of onset and acceptance of disability
for male stroke survivors in our sample. The question emerges, Why do females
who experienced an early onset of stroke report the lowest levels of
acceptance of disability in the total sample? Again, I suggest that the
explanation has to do with the role of middle-aged women in our society and
the greater competing demands they experience during this stage of life
compared to their male counterparts.
To conclude, we return to Figure 3 to view our schematic diagram of "Aging
with a Disability -- A Life Course Perspective." We are convinced that the
life course perspective provides an important framework for moving beyond the
traditional medical model of impairment and the rehabilitation model of
function to gain a better understanding of how the meaning and consequences
of disability are affected by the timing of onset.
Viewed from a life course perspective, the onset of both acute and secondary
disability represents more than a medical problem or a further restriction of
functional activity. Rather, both represent unscheduled life transitions or
disruptions which threaten to undermine an individual's sense of continuity.
Therefore, persons who expeience these unscheduled life transitions at
different chronological ages face the challenge of integrating potentially
profound disruptions into their continuous sense of self at different stages
of the life course.
I want to take a moment to thank all of those who participated in our study
who are here today.
Aging with a Disability: A Life Course Perspective
Baum HM, Manton KG (1987). National trends in stroke-related
mortality: A comparison of multiple cause mortality data with
survey and other health data. The Gerontologist, 27:293-300.
Dauer C (1948). Trends in age distribution of poliomyelitis in the
United States. American Journal of Hygiene, 48:133-146.
Davis F (1963). Passage through crisis: Polio victims and
families. New York:Bobbs-Merrill.
Eisenberg MG, Sutkin LC, Jansen MA (1984). Chronic illness and
disability through the life span: Effects on self and family. New
York: Springer Publishing.
Erikson E, Erikson J, Kivnik H (1986). Vital involvement in old
age. New York: W.W. Norton.
Fisks M, Chiriboga D (1990). Change and continuity in adult life.
Halstead LS, Rossi CD (1985). New problems in old polio patients:
Results of a survey of 539 polio survivors. Orthopedics, 8:845-850.
Kotre L (1984). Outliving the self: Generativity and the
interpretation of lives. Baltimore, MD: Johns Hopkins University
Maurer JR, Strasberg PD (1989). Building a new dream: A family
guide to coping with chronic illness and disability. Reading, MA:
Addison Wesley, Inc.
Mulder DW (1992). Post-polio syndrome -- past, present, and future. In
TL Munsat (Ed.) Post-polio syndrome (pp. 1-8). Boston:
Norris VK, Stephens-Parris MA, Kinny JM (1990). The impact of
family interactions on recovery from stroke: Help or hindrance?
The Gerontologist, 30:535-542.
Scheer J (1984). They act like it was contagious. In SC Hey, G
Kiger, J Seidel (Eds.) Social aspects of chronic illness,
impairment, and disability (pp. 62-69). Salem, OR: Willamette
Scheer J, Groce N (1988). Impairment as a human constant:
Cross-cultural and historical perspectives on variation. Journal of
Social Issues, 44:23-37.
Scheer J, Luborsky ML (1991). The cultural context of polio
biographies. Orthopedics, 14:1173-81.
Task Force on Stroke Impairment, Disability and Handicap (1990).
Symposium Recommendations for Methodology in Stroke Outcome
Research, Stroke, 21(suppl.II):68-73.
Treischmann RB (1987). Aging with a disability. New York: Demos
When stroke strikes -- Information for the family: The road back
resocialization program. Stroke Association of Southern California, 2001 S.
Barrington Ave., Suite 308, Los Angeles, CA 90025.
Zola IK (1989). Aging and disability: Toward a unified agenda.
Journal of Rehabilitation, 28:6-11.