‘ ”theta

nsm usmmes

 

 

 

 

 

 

 

 

 

Illllllllljllilllgllw

This is to certify that the

thesis entitled

Compliance, Memory Loss, and Depression
In the Elderly

presented by ,

Barbara Clark

has been accepted towards fulfillment
of the requirements for

Masters Clinical Psychology

degree in

liars. M

Major professor

 

 

 

 

 

 

Date July 16, 1991

0-7639 MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

 

_LIBRARY w‘
M'cmgan State
Unlverslty

 

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before dmo duo.

DATE DUE DATE DUE

        
  

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Aflirmatlve Action/Equal Opportunity Institution
cW.prn3-p.t

 

COMPLIANCE, MEMORY LOSS, AND DEPRESSION
IN THE ELDERLY

By

Barbara J. Clark

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF ARTS

Department of Psychology

1991

ABSTRACT

COMPLIANCE, MEMORY LOSS, AND DEPRESSION
IN THE ELDERLY

By

Barbara J. Clark

The intent of this study was to identify potential risk
factors for unintentional noncompliance with prescribed
medication regimens in an elderly population. Variables
hypothesized to affect ability to comply included cognitive
impairment, depressed feelings, complexity of dosing regimen,
and age of participant. In addition, two other variables,
reliance on memory aids and intent to comply with medication
instructions, were included in the final analysis.

The 40 elderly participants who completed the study
resided independently in the community and took their
prescribed medications without help. The participants
provided demographic information and a medical history.

Assessment instruments included the Beck Depression
Inventory, the Geriatric Depression Scale, the Mini Mental
State Examination, and the Senile Dementia of the Alzheimer
Type (SDAT) battery. Compliance percentage and complexity
were determined by pill counts.

Results using Chi Square analysis suggest there is a
relationship between memory loss, intent to comply, and
actual compliance. A possible relationship between

depression and compliance is discussed.

ACKNOWLEDGMENTS

I would like to thank Dr. Norman Abeles, my thesis
chairman, for his guidance, support, and encouragement. I
would also like to thank Dr. Robert Caldwell and Dr. Raymond
Frankmann for their guidance and consideration.

I would like the thank my husband, Lee, for his
professional and personal contribution to this study. I

would not have completed this without his help.

ii

TABLE OF CONTENTS

List of Tables. .

Introduction.

Hypotheses. .

Review of the Literature.

Memory Loss in the Elderly.

Depression in the Elderly

O O O O I O

Memory Loss and Depression in the Elderly

Age and Compliance in the Elderly

6

Depression and Compliance in the Elderly.

Complexity of Dosing Schedule and Compliance

in the Elderly.

O O O O O O O O O I O O O

Memory Loss and Compliance in the Elderly

Health Care Providers and Compliance
in the Elderly.

Significance of the Study
Method. .

Participants.

0 O O O O 0

Procedure .

Instrumentation

Statistical Analysis.

Results .

O O O O O O O O O O O O O 0

Descriptive Statistics.

iii

10
10
12
14
16

17

19

20

22
24
26
26
26
29
33
35

35

Pearson Product Moment Analysis

Chi Square Analysis .
T-test Analysis . . .
Regression Analysis .
Discussion. . . . . . .
Summary of Findings

List of References. . .

iv

Page
37
37
4O
43
45
55

58

LIST OF TABLES

Page
Pearson Product Moment Correlations for
Compliance and Independent Variables . . . . . . . 37
T-test on Use of Memory Aids and Independent
variables. I O O O O O O O I 0 O O I O O O O I O O 41
T—test on Intent to Comply and Independent
variables. 0 O O Q 0 O O O O I O O O O O O O O O O 42

INTRODUCTION

Elderly persons represent 12% of the population in the
United States, yet they consume over 30% of all drugs
prescribed (Landry, Smyer, Tubman, Lago, Roberts, & Simonson,
1989). By the year 2000, they will represent 17% of the
population and will account for over 50% of prescription drug
use. Elderly Americans living independently in the community
receive at least 20 prescriptions each year (Beller & Evans,
1987). It is not surprising that the elderly’s expenditure
on drugs is disproportionately greater than that of the
general population (Swift, 1988). It is also more likely for
an elderly individual to have a chronic disease requiring
treatment with drugs (German, Klein, McPhee, & Smith, 1982).
The simultaneous treatment of several diseases increases the
risk of adverse drug reactions, and drugs have been estimated
to be directly related to 20-33% of all hospital admissions
of those over 85. It is clear that drug toxicity is more
severe and frequent in the elderly and that they are more
sensitive to adverse side effects and drug interactions than
are younger adults who consume the same drugs.

Both normal aging and the disease processes of chronic
illness make older adults more susceptible to adverse

reactions to prescription medicines. As an individual ages,

1

their body composition changes, resulting in alterations of
the distribution, metabolism, and renal excretion of drugs
(Kazis & Friedman, 1988). The elderly are more vulnerable
physiologically, psychologically, and anatomically to the
adverse effects of drugs, and the potential for serious harm
is significantly greater.

One goal of a responsive health care system should be to
assist the elderly in maintaining functional status. The
majority of elderly patients live at home, so it appears that
attempts to preserve their independence can be a considerable
benefit, both in terms of quality of life to the individual
and cost to the community (Currie, Moore, Friedman, &
Warshaw, 1981). Understanding the factors which permit or
threaten the maintenance of the elderly in their homes is
helpful in keeping them there. One of these factors is the
ability to take medication as prescribed. Meichenbaum and
Turk (1987) View compliance as "the extent to which patients
are obedient and follow instructions, proscriptions, and
prescriptions of health care providers" (p. 20).
Noncompliance with prescribed medication can take two forms--
intentional and unintentional. Intentional noncompliers
choose not to follow instructions on how to take medication,
while unintentional noncompliers try to follow instructions
but are unable to for some reason.

According to some studies, up to 45% of the elderly do
not take their medications as prescribed (Folkman, Bernstein,

& Lazarus, 1987; Gabriel, Gagnon, & Bryan, 1977; Martin &

Mead, 1986; Morgan, Nowson, Murphy, & Snowden, 1986; Morrow,
Von Leirer, & Sheikh, 1988). Furthermore, rates of
noncompliance increase with the number of medications taken
and the number of times a single medication is taken (Morrow
et al., 1988). Statistics such as these underscore the need
to study factors affecting noncompliance in the elderly since
surveys show that 85% of patients over the age of 75 are
responsible for taking their own medication (Wanderless &
Davie, 1977).

Thus, it appears that the successful application of
ambulatory programs for the control of disease demands that
the health care provider be able to detect those elderly
individuals who will be noncompliant. Furthermore, it has
been documented that physicians overestimate the degree of
compliance among their patients and that their ability to
predict whether or not a patient will comply with the
medication regimen is no better than chance (Beck, 1985). It
is well documented that patients generally overstate the
extent of their compliance (Weibert & Dee, 1980). Over 25
years ago Schwartz, Wang, Zietz, and Goss (1962) recognized
the need to identify reliable risk factors for noncompliance
so that the health care provider might be able to detect the
potentially unreliable patient before treatment was begun.
More recently, Pace (1989) has reported that the
incorporation of functional assessment into the routine
ambulatory care of the elderly is supported in the

literature. Assessment of the impact of aging on an elderly

individual’s ability to function independently in the
community is critical in providing appropriate health care.
Since it appears that the majority of ambulatory elderly
patients with chronic illness do not follow their prescribed
treatment and that the physician is often the last to know
that the patient is not complying, it is imperative to
identify factors that influence noncompliant behavior.

In research on unintentional noncompliance, Wong and
Norman (1987) state that poor compliance may be attributed to
cognitive impairments associated with aging. Between 5-20%
of persons over 65 suffer from dementia, a deterioration of
intellectual capacity that includes serious declines in areas
of cognition, memory, language, visual-spatial skills, and
personality (Benson, 1982; Larson, Reifler, Featherstone, &
English, 1984). Meyer and Schuna (1989) say that demented
patients may have decreased ability to understand, remember,
and follow through with the directions that would enable them
to take their medications as prescribed. Since a large
proportion of patients discharged from a general hospital are
elderly and a substantial number are likely to exhibit
impairment, this population would appear to be at high risk
for cognitive failure to comply (Johnston, Clarke, Mundy,
Cromarty, & Ridout, 1986). Thus, it appears that dementia as
well as age-associated cognitive decline would affect the
ability to comprehend and remember text, a vital component in
the task of following dosage instructions. Elderly

individuals are most susceptible to making mistakes because

they have significantly more complicated medications
schedules, and the more drugs they take, the greater the
likelihood of adverse drug reactions which may further impair
cognitive abilities. Morrell, Park, and Poon (1989) reported
that elderly persons are unable to process information as
well as younger adults whenever the amount of information
presented is increased. In addition, gradual reductions in
cognitive abilities may be adversely affected by diseases
that elderly persons seek health care for (Morrow et al.,
1989).

Disease is not the only factor that effects cognitive
abilities; mood has been implicated in many studies. It is
apparent that depression in particular is common among the
elderly, with prevalence rates ranging from 10-20% (Haggerty,
Golden, Evans, & Janowsky, 1988). As the intensity of the
depression increases, cognitive processes are disrupted, and
less can be learned and remembered (Weingartner, Cohen,
Murphy, Martello, & Gerdt, 1981). Furthermore, both dementia
and depression can coexist in the elderly, and there is a
trend for those with moderate to severe depression to
manifest more cognitive dysfunction (Cavanaugh & Wettstein,
1983). A depressed emotional state could be a cause of
noncompliance in that the drug reminds the elderly individual
they are sick, so they consequently "forget" to take the
medication (Weibert & Dee, 1980). Therefore, although there
is a paucity of research on the subject of unintentional

noncompliance, it seems possible that the existence of

depression in the elderly could adversely effect their
ability to comply with complicated dosing schedules.

No conclusive data have emerged that identify the
factors that affect unintentional noncompliance in the
elderly, but it is important to do so since the effects of
inappropriate drug use would be more severe among the elderly
as compared to the general population. Few researchers
address cognitive function in the elderly and relate it
specifically to drug compliance; even fewer studies relate
depression specifically to drug compliance. Concern about
the misuse of drugs among the elderly centers on the
deleterious effects on day-to~day functioning of mistakes in
dosing. The effects of hyper and/or hypo compliance can be
seen physiologically by identifying somatic health symptoms
and dysfunctions in the central nervous system, as well as
psychologically in the way the elderly cope with everyday
stresses (Folkman et al., 1987). Spagnoli et al. (1989)
state that factors which influence compliance must be
researched since errors in drug taking have potentially more
serious effects in the elderly. The two factors mentioned
here, the effects of dementia and/or depression on
unintentional noncompliance in combination with increasing
age, should be significant and most severe when the elderly
try to understand and remember complex medication schedules.

While it is important to identify risk factors that
would identify elderly who are unintentional noncompliers,

the elderly who are intentional noncompliers should be

mentioned. Elderly patients choose to discontinue drug
therapy for many reasons, some of them being the failure to
recognize the seriousness of their condition, the belief that
once their symptoms disappear they are cured, the erroneous
conclusion that the medication is not working as it should,
the cost of the medication, and the deleterious side effects
of the medication (Falvo, 1985; Weibert & Dee, 1980). It is
not uncommon for elderly individuals to comply with one
prescribed medication but to choose not to take their other
prescribed drugs (Richardson, 1986). Richardson (1986) cites
one study indicating that 73% of noncompliance is intentional
with the most common reason for not complying being the
belief that the drug was not needed in the dosage prescribed.
Evidence seems to suggest that intentional noncompliance is
prevalent among the elderly.

Finally, while there are many articles dealing with the
problem of noncompliance, health care providers have
neglected to integrate the management of the problem into
their clinical practices. Improved function is an
appropriate therapeutic goal for the elderly, especially if
they are chronically ill. For busy, office-based health care
professionals, "that goal is more easily realized if some
means are available to regularly assess the functional
abilities of elderly patients at risk” (Pace, 1989, p. 29).
Distinguishing the causes of noncompliance, both
unintentional and intentional, should be an important avenue

for research. This study focuses specifically on identifying

possible causes of unintentional noncompliance.

The ability to screen for loss of functions on a routine
basis represents a potential first step in identifying
overlooked problems that demand attention, such as dementia
and/or depression (Pace, 1989). If it is found that dementia
and depression are factors that effect medication-taking
behavior in the elderly, it will be useful for health care
providers to have access to standardized, structured
assessment instruments and questionnaires that will enable
them to identify patients at risk for unintentional
noncompliance so they can take preventive measures.

Therefore, it is the intent of this study to identify
factors that affect unintentional noncompliance in the
elderly and in turn to provide health care professionals with
standardized assessment instruments and questionnaires which
will reliably identify those elderly at risk.

Accordingly, the present study was set up to test six
specific hypotheses:

1. There will be a statistically significant correlation
between memory loss and compliance. Specifically, the
community-dwelling elderly with more memory problems will be
significantly less compliant than those with less memory
problems.

2. There will be a statistically significant correlation
between depression and compliance. Specifically, the
community-dwelling elderly who are more depressed will be

significantly less compliant than those who are not

depressed.

3. There will be a statistically significant correlation
between the presence of dementia as well as depression and
the ability to comply. Specifically, the community-dwelling
elderly who are both demented and depressed will be
significantly less compliant than those who are only
depressed, only demented, or not depressed or demented.

4. There will be a statistically significant correlation
between the complexity of the dosing regimen and compliance.
Specifically, the community-dwelling elderly who have more
complex regimens will be significantly less compliant than
those who have less complex regimens.

5. There will be a statistically significant correlation
between the age of participant and compliance. Specifically,
as age of participant increases, ability to comply with
dosing regimens will decrease.

6. These four variables-~memory loss, depression, complexity
of dosing schedule, and age--will fit a multiple regression
model of compliance for the community-dwelling elderly

population.

REVIEW OF THE LITERATURE

Mgmgry Loss in the Elderly

Dementia is one of the major health problems in the
elderly. Ciocon and Potter (1988) report that 50% of
individuals over 60 complain of memory problems, with
dementia the most common cause of memory impairment in the
elderly. Stern and Bernick (1987) state that "dementia is a
syndrome of diffuse disturbance of cognitive functions in an
alert patient of sufficient severity to interfere with social
or occupational performance; a defect in memory is always
present" (p. 44). Although it is accepted that as
individuals age there are declines in intellectual processing
and memory, it is hard to distinguish age-associated declines
from a dementing process (Morrow et al., 1988; Stern &
Bernick, 1987). Early dementia is a disorder that primarily
affects the elderly and is characterized by a wide range of
symptoms such as failing attention and memory, errors of
judgment, irritability, personality changes, and poor
orientation (Haggerty et al., 1988; Meyer & Schuna, 1989;
National Institute on Aging Task Force, 1980). The
prevalence rate of dementia has been estimated to be between
5—20% for individuals over 65 years of age (Larson et al.,
1989; Lazarus, Newton, Cohler, Lesser, & Schweon, 1987;
McCartney & Palmateer, 1985; McIntyre & Frank, 1987; Small &
Jarvik, 1982). Furthermore, dementia of the Alzheimer Type

is the most common cause of dementia in the elderly

10

11

population (Larson et al., 1986; Bayer, Pathy, & Twining,
1987). The elderly individual with memory problems forgets
names, overlooks appointments, fails to recall the day’s
events, leaves tasks unfinished, and becomes ignorant of the
date, time of day, or season of the year (Roth, 1976).

Memory problems may be concealed by elderly persons who have
high initial intelligence, but as these problems magnify, the
individual is unable to recall past events as well.

Mild degrees of intellectual impairment are relatively
common in the older adult population, and these individuals
may be at risk for further dementing processes, or these may
be age-associated (late-life forgetting) memory impairments.
The earliest change is usually identified as a selective
impairment of memory for recent events. Roth (1976) states
that an elderly individual with dementia is characterized as
having progressive impairment of memory and orientation for
time and place, a decline in intellectual grasp and
comprehension, deterioration of the personality, and mood
lability.

It is important to assess for dementia in the elderly
population even though symptoms of forgetfulness and
difficulties in understanding and coping with life stresses
are often attributed to old age (Bayer et al., 1987). In
fact, many physicians overlook such symptoms and fail to

recognize a dementing process (Bayer et al., 1987; Pace,

1989).

12

Depression in the Elderly

Another major problem in the elderly population is the
presence of depression. The incidence of depression
increases as age increases (NIA Task Force, 1980). The
prevalence of depression in the elderly population has been
estimated to be between 10-15%, and the prevalence of
depressive symptoms in disproportionately greater in the
elderly (Blazer, Hughers, & George, 1987; Good,
Vlachonikolis, Griffiths, & Griffiths, 1987; Griffiths, et
al., 1987; Reifler, Larson, & Hanley, 1982). Reifler,
Larson, Teri, and Poulsen (1986) reported that 24% of the
elderly have significant depressive symptoms. Freedman,
Bucci, and Elkowitz (1982) report that the highest rate of
depressive symptoms is found in those over 65 who live in the
community. Indeed, depression is the most common
psychological complaint among the aged. Blazer et al.
(1987) cite studies which found the prevalence rate of
depression to be 21% in one cohort of older adults and
pervasive depression to be 13% for persons over the age of
65. Henderson and Kay (1984) cite a study that found the
prevalence rate of depression in individuals over the age of
65 to be 8.1%. This sample was only made of community~
dwelling elderly. They also reviewed another study that
randomly selected persons over the age of 65 living in the
community and found prevalence rates of depression to be 13%.

Because of the increased likelihood of multiple

psychological, social, physical, and cognitive losses, the

13

elderly are at a greater risk for developing depression than
any other age group. Salzman and Shader (1979) point out
that in the context of declining function, limited coping
strategies, loss of interpersonal support, and stress such as
disease, it is not surprising that late-life depression
evolves. They go on to say that the loss of health,
cognitive abilities, coping strategies, employment, and
independence, work to lower an older adult’s self-esteem and
bring on a depressive episode.

Dementia and depression are related, but the overlap
between the two represents a diagnostic challenge (Spar,
1982). Unfortunately, it is difficult in some instances to
determine whether an elderly individual is clearly
experiencing memory problems as a result of intellectual
impairment and decline in cognitive functioning. At times,
it may appear that an older adult suffers from dementia, when
in fact a depressive state appears as dementia (Thompson,
1986). This condition is referred to as pseudodementia, and
it is different from true dementia in that the underlying
depression of pseudodementia is treatable. Roth (1976)
states that 15% of elderly persons who are depressed exhibit
cognitive impairment. Clinical research has shown that the
memory problems of older adults with pseudodementia improve
once the depression is effectively treated (Salzman & Shader,
1979). In a case of true dementia this would not happen. It
should be noted that some authorities question the entire

concept of pseudodementia (Salzman & Gutfreund, 1986).

14

Pseudodementia is described by Wells (1970) as the
syndrome in which dementia is mimicked by depression.
Although their performance on mental status examinations may
point to organic brain disease, this is usually not the case;
their affective states usually inhibit their performance
making it appear that they are demented. Many researchers
have identified the presentation of an elderly person with
pseudodementia as follows: a) complaints and distress over
declining cognitive abilities, b) history of recent, abrupt
onset of symptoms that fluctuate over time, and c) appearance
of depression at examination (Benson, 1982; Black & Hughes,
1987; Caine, 1981; Feinberg & Goodman, 1984; Fopma-Loy, 1986;
Haggerty et al., 1988; Jorm, 1986; Reding, Haycox, & Blass,
1985; Small & Jarvik, 1982; Spar, 1982).

MQEQEY Loss and Depression in the Elderly

Although it appears true that depression can be masked
by findings suggestive of dementia, depression and dementia
can also coexist in an elderly individual (Devanand & Nelson,
1985; Reding et al., 1985; Stein & Bernick, 1987). Reifler
et al. (1982) state that the concept of pseudodementia did
not seem helpful in dealing with the mixed presentation of
coexistent dementing and depressive disorders. Griffiths et
a1. (1987) found that increasing dementia is associated with
increasing depression and that depression and dementia were
mutually exclusive diagnoses. Reding et a1. (1985) suggested
that a relatively severe impairment of intellectual

functioning might suggest the presence of a coexistent

15

dementing disorder along with depression. Davis and Robins
(1989) found the presence of concurrent psychiatric disorders
in demented elderly subjects, one of which was depression.
Lazarus et al. (1987) found that elderly patients diagnosed
with dementia had a significantly higher frequency of
depressive symptoms and suggested that there might be a high
frequency of concomitant depressive symptoms in elderly
patients diagnosed with dementia. Reifler et al. (1982)
found that depressive and dementia were distinct disorders in
the elderly in that depression was not a universal feature of
early dementia; two-thirds of the mildly depressed elderly
patients did not meet the criteria for depression, and one-
third had no recognizable symptoms of depression. Finally,
Reifler et al. (1986) found that 25% of demented elderly
patients also suffered from a depressive disorder.

In a study which used the Mini Mental State and the Beck
Depression Inventory, researchers sought to relate the
severity of depression and cognitive dysfunction when they
coexist in an inpatient pOpulation (Cavanaugh & Wettstein,
1983). They found no relationship between the severity of
depression and cognitive dysfunction among subjects younger
than 65, and in those older than 65, only an insignificant
trend was reported. Although the trend indicated that
elderly patients with moderate to severe depression tend to
have more cognitive dysfunction, the authors state that both
conditions should be evaluated independently, as it appears

the severity of depression is not directly related to

16

alterations in cognitive processes (Cavanaugh & Wettstein,
1983).
Age_and Compliance in the Elderly

One of the factors that has been studied in association
with compliance is the effect of increasing age. The
evidence regarding the influence of age is confusing, and it
is still not clear whether or not compliance decreases as age
increases. German et al. (1982) found that patients over 65
were significantly less knowledgeable about their diseases
than were younger patients and had less knowledge of the
drugs prescribed, although knowledge was not related to
compliance. They did not look at the specific relationship
between age and compliance. Davis and Eichhorn (1960) found
that the age of patients affected the direction and degree of
change in compliance, with the younger age group continuing
the highest rate of compliance. Some years later, Stewart
and Cliff (1972) found that the tendency to be noncompliant
seemed to be correlated with increasing age, although this
was not consistent. In a review of the literature, Marston
(1970) states that only two studies found younger patients
are more likely to compliant than older. Lastly, Schwartz et
al. (1962) found that when they separated their sample into
two groups based on age (60-74; 75 and over), age was
directly correlated to error-making, with the patients 75 and
over making significantly more errors than patients under 75
years of age.

On the other hand, several studies determined that age

17

was not correlated with compliance. Malah (1966) and Neely
and Patrick (1968) found that in an elderly population age
made no difference in the identification of noncompliers.
More recently, Spagnoli et al. (1989) found that the effect
of age on compliance was minimal when comparing the age

groups of "young-old" versus "very-old" (over 75). Wong and

Norman (1987) also found that in an elderly subject
population age was not a factor affecting compliance.
Darnell, Murray, Martz, and Weinberger (1986) found the
effect of age on compliance in an elderly population was not
significant. Rost and Rober (1987) found that age was not
significantly related to recall of medication instructions in
an elderly population. In a review of the literature,
Marston (1970) concluded that age is probably not
significantly related to compliance. Finally, Morrell et al.
(1989) reported that age does not seem to be a contributing
factor to noncompliance but that there are underlying factors
in each age group that effect compliance behavior. In the
present study, factors thought to effect compliance in the
elderly include the presence of dementia and/or depression,
and/or the complexity of the medication schedule, although
the effect of age will also be studied.
Depression and Compliance in the Elderly

Reifler et al. (1986) state that 24% of elderly medical
outpatients have significant depressive symptoms and that 11%
suffer from a major affective disorder. Kinzie, Lewinsohn,

Maricle, and Teri (1986) state that medical illness and

18

medication use are common in the elderly population and that
they can contribute to depression. They found that 52% of
middle-age and elderly subjects had depressions definitely or
probably associated with a medical illness or medication.

It is disturbing that in many cases, health care
providers rarely recognize the presence of depression
(Fitton, Morley, Gross, Petry, & Cole, 1989; Kinzie et al.,
1986). Rodin and Voshart (1986) report that since depression
is common in the medically ill, it is important for the
clinician to detect depressive symptoms in order to provide
better overall health care. Folkman et al. (1987) recognize
that health care professionals should be aware of symptoms of
depression in an elderly patient and that these symptoms may
be the result of drug misuse. Other authors have reported
that physical illness and medication usage may precipitate or
contribute to the development of depressive symptoms
(Haggerty et al., 1988).

Considering this evidence, it is surprising that
research has not been conducted to determine the specific
relationship between depression and the ability to comply
with medication instructions. In a review of the literature,
Marston (1970) could find no study that assessed the presence
or absence of depression and measured actual compliant
behavior. Fitten et al. (1989) state that many health care
providers may fail to identify symptoms of depression and
monitor compliance in the elderly. No studies could be found

that explored the specific relationship between depression

19

and compliance, but findings seem to support the hypothesis
that the presence of depression could adversely affect the
ability to remember to take medication. Weingartner et al.
(1981) found that cognitive processes such as the ability to
learn, remember, and encode information were disrupted in
depressed patients. Morrell et al. (1989) suggest that
anxiety associated with the presence of a medical illness in
elderly individuals could adversely affect rates of
compliance but did not test this hypothesis.
Complexity of DosingASchedules and Compliance in the Elderly
There has been a great deal of research conducted on how
the complexity of the dosing regimen affects compliance among
the elderly. Complexity is an important factor because of
the increased likelihood of polypharmacy and multiple illness
in the elderly population (Morrell et al., 1989). Morrell et
al. found that as the amount of drug information presented to
old and young subjects increased, recall was significantly
affected. The more complex the drug information, the less
likely both groups were to remember it. Many researchers
have found that as the number of prescribed medications
increases, the more likely the elderly patient is to make
errors in medication taking (Darnell et al., 1986; Malahy,
1966; Morgan et al., 1986; Neely & Patrick, 1968; Spagnoli et
al., 1989; Wandless & Davie, 1977). In the majority of
research found on this relationship, the participants who
were studied were not cognitively impaired and were oriented.

In contrast to these findings, Schwartz et al. (1962) found

20

that although patients who took 2-3 prescribed medications
were more likely to make errors than were those taking only
one, larger numbers of prescribed medication and increasing
complexity did not produce a larger proportion of errors.
Finally, Wong and Norman (1987) found that the total number
of prescribed doses did not correlate with compliant
behavior.
Memory Loss and Compliance in the Elderly

The relationship between declining cognitive function
and compliance has received attention. Dementia causes loss
of memory and impairments in cognitive function that could
impair a patient’s ability to do everyday tasks such as
remembering dosing schedules (Mulley, 1986). To be able to
comply with dosing schedules, an elderly patient must be able
to remember instructions and retrieve them at the proper time
and then must be able to remember to take the medication at
the proper time (Morrell et al., 1989). Cavanaugh, Grady,
and Perlmutter (1983) found that the elderly encounter more
memory failures in their everyday memory experiences and that
forgetting routines was a significant problem. Morrell et
al. (1989) point out that loss of memory could affect
compliance and tested the ability of elderly patients to
recall medication instructions. These researchers did find
that the elderly consistently manifested poorer recall of
medication instructions. However, they did not assess memory
functions using scorable measures which could be correlated

with compliant behavior. Furthermore, they did not test

21

actual compliance since they simply asked participants fill
out a medication instruction recall sheet. Meyer and Schuna
(1989) did use a cognitive screening instrument and found
that elderly patients scoring in the impaired range were less
likely to correctly read and interpret prescription labels.
However, they did not assess the actual medication-taking
behavior of the participants because the purpose of the study
was to only assess functional ability to take medication. It
appears that many studies which have assessed functional
competence in relation to compliant behavior have used
participants who are not experiencing significant memory
problems (Darnell et al., 1986). Spagnoli et al. (1989)
found that poor understanding of the drug regimen was the
most frequent reason for noncompliance, but they did not have
an index of cognitive functioning. Wong and Norman (1987)
gave a mental status exam to their elderly participants and
found that scores on the exam did not correlate with
compliant behavior. However, they only included individuals
in their sample who did not fall into the demented range.
Therefore, although many researchers have stated that
cognition appears to be a crucial factor in assessing elderly
person’s ability to comply with dosing schedules, very few
have actually addressed the problem specifically. Only one
study could be found that assessed memory using instruments
that would supply an index of cognitive functioning that
could be directly correlated with the ability of elderly

individual to recall medication instructions they would use

22

in their everyday environment.

That study used short, standardized depression and
dementia instruments, the Zung Depression Scale and the Mini
Mental State, and attempted to assess and correlate cognitive
and affective functioning with the ability to recall
medication regimens (Rost & Roter, 1987). Their sample
included elderly outpatients, 84% of whom were oriented to
the date and 66% of whom could remember at least 2 of 3 words
on a memory test. Thus, even before they were given the
mental status exam it appears that the majority of the
participants were not experiencing major memory problems and
might not have been representative of the general ambulatory
elderly population. Recall was assessed by asking
participants to list all medications recommended by their
physicians. Thus, compliance was assessed indirectly. It
does not seem surprising, therefore, that in this study
memory loss and/or depression were not correlated with
compliance. There is a need for a study designed to
assess more accurately the relationship between memory loss
and/or depression and other factors which might correlate
with compliance in the elderly.

Health Care Providers and Compliance in the Elderly

Since there is a large body of research substantiating
the importance of the relationship between the health care
provider and the elderly patient, it seems practical to find
new ways in which clinicians can better meet the needs of

their elderly patients. Bergmann, Foster, Justice, and

23

Matthews (1978) sought to determine how demented patients who
are able to live independently in the community can be
identified from those who will not be able to function
independently and whether other factors such as depression
that threaten the independence of the community-dwelling
elderly with dementia can be identified. The presence of
dementia and/or depression in the elderly population is
frequently undetected even though suitable instruments and
questionnaires are available for assessing this p0pulation’s
cognitive and emotional functioning (Currie et al., 1981;
Freedman et al., 1982; Knights & Folstein, 1977; Williamson,
et al., 1964). It seems necessary to develop a routine
method to test the cognitive abilities and emotional status
of elderly individuals who could benefit from
multidimensional assessment in that clinicians could tailor
an appropriate health care plan to the level of functioning
(American Geriatric Society Public Policy Committee, 1989).
Since the present study is interested in developing a
routine assessment procedure for the office-based
professional, it must be taken into account that instruments
requiring a long administration time will meet with
resistance and that targeted shorter instruments would have
the greatest utility and would better meet the needs of the
patient and clinician (Applegate, 1987). Instruments
appr0priate for the health care provider must also meet the
criteria set forth by Pace (1989) who states that what

clinicians desire most is a way to assess clinically

24

significant change in their patients. Finally, instruments
and questionnaires selected for use must be reliable and
valid for use in the elderly population, and they should
provide information that will aid the clinician in helping
the elderly to maintain functional status is the community.

The most obvious health care provider to make use of a
routine assessment method is the office-based physician.
Most of the general health care of the elderly is done by
family practitioners, and the office setting provides a
suitable place in terms of the practicality of doing
assessments (Freedman et al., 1982). In addition, practicing
physicians should be encouraged to utilize the expertise of
other disciplines when developing an appropriate health care
plan (AGS Public Policy Committee, 1989). There are several
studies which espouse the use of the pharmacist as an
important health care professional who can influence patient
compliance (Gabriel et al., 1977; Kazis & Friedman, 1988).
Johnston et al. (1986) also found that a pharmacist was an
appropriate and cost-effective professional whose expertise
could be tapped when dealing with compliant behavior.
Therefore, it appears that both physicians and pharmacists
could be trained to use assessment tools which would identify
those elderly patients at risk for noncompliant behavior.
Significance of the Study

Many researchers have addressed the problem of
compliance in the elderly. However, few have studied

specific factors which directly affect how compliant an

25

elderly individual will be. Not only will this study try to
identify specific risk factors for unintentional noncompliant
behavior, but it will also provide a way for the trained
health care professional to identify those community-dwelling
elderly at risk. This study may help in facilitating better
health care for the elderly and enabling them to more

successfully function.

METHOD

Participants

The participants for the present study were over 60
years of age, ambulatory, residing independently, and
caring for their own needs in the community. Each
participant was taking at least one medication on a scheduled
basis with no help from care givers. They volunteered to
participate with the expectation that they would receive
feedback about cognitive functioning and information about
prescribed medications.
Procedure

Participants, after learning about the study through
advertisements and health care providers, contacted the
Psychological Clinic at Michigan State University to schedule
appointments with investigators. They were encouraged to
participate after being told a little about the study, namely
that many people worry about memory problems as they get
older and that this study would provide them with useful
information about their memory. They were also told that the
study included asking them questions about the way they were
feeling and doing a survey of the medication they were
taking, since in some cases medication can affect memory and
mood. Therefore, demand characteristics were minimized in
that compliance was never directly addressed.

Participants were asked to bring all of their scheduled

medications (those not taken on a prn basis) to the initial

26

27

testing session. They were also informed that they would be
expected to return in 3—4 weeks in order to complete the
study and receive feedback and information about their
medications. At this time any further questions they had
about the study were answered.

When participants first arrived, they were greeted by
the principle investigator, a graduate student in clinical
psychology, and asked to enter a room in which memory and
mood were assessed. Participants completed a subject
information sheet which obtained their name, address, age,
phone number, and primary care physician. Participants were
assured of confidentiality at this time and were told their
subject number would be used to identify them at a later
date. The principle investigator had access to participants’
names in order to match interview one and interview two data;
others have access to subject numbers only.

Participants were then instructed to read and sign two
research consent forms, one of which they kept. The consent
form contained information about the study and testing
procedure.

The actual assessment procedure lasted approximately
one hour. The first test administered was the Mini Mental
State Examination (MMSE) and was followed by a battery to
assess Senile Dementia of the Alzheimer Type (SDAT). Both of
these instruments were used to determine whether or not
participants suffered from dementia. Next they completed a

self~report questionnaire, the Beck Depression Inventory

28

(BDI). If participants had vision problems and/or were not
able to write, the questionnaire was read by the examiner,
and participants answers were recorded. These instruments
were used to determine whether or not participants were
mildly, moderately, or severely depressed.

The first investigator then led participants to an area
where the second examiner, a pharmacist, was waiting. The
pharmacist recorded the drug regimen from the label of each
prescribed medication and did a pill count for each drug. At
this time he explained that participants would need to return
in 3-4 weeks with their medication in order to receive
feedback about their memory and information about their
medication. A follow-up appointment was made, and
participants were free to leave.

At the follow—up session, participants were greeted by
the principle investigator and received feedback about their
cognitive function. They were then taken to the second
investigator where their pills were recounted. Compliance
scores were calculated from the following formula:

a) ideal count - actual count x 100% = compliance score
ideal count

Complexity was considered to equal the exact number of
pills participants had to remember to take in one day and was
calculated from the dosage directions recorded from each
scheduled medication,

To determine the proportion of participants who were
intentional noncompliers, the pharmacist asked, "Do you

always try to take your pills in the right dose at the right

29

time, or do you sometimes decide not to take your medication
for some reason?" The pharmacist recorded answers and
queried participants if needed about reasons for intentional
noncompliance. The pharmacist also asked participants the
following: (a) "If your physician or pharmacist asked you
whether you try to take your medication as prescribed, what
would you tell them?" (b) "In the past, have you stopped
taking your pills before the prescription ran out, and if so
why?" and (c) "In the past, have you ever taken different
dosages of medication than what was prescribed, and if so
why?" In asking these questions, the examiners were able to
record the number of intentional noncompliers, the reasons
they chose not to comply, and whether a self-report
assessment for intentional noncompliers would serve to
identify them.

Finally, the pharmacist gave participants information
about their medications and answered any questions they had
about the drugs they were taking. Participant were then free
to leave.

Instrumentation

The assessment tools used in the present study were the
following: (a) the Senile Dementia of the Alzheimer Type
(SDAT) test battery, (b) the Mini Mental State Examination
(MMSE), and (c) the Beck Depression Inventory (BDI).

The SDAT battery consists of tests of logical memory,
trailmaking, word fluency, and mental control (Storandt,

Botwinick, Danziger, Gerg, & Hughes, 1984). Storandt et al.

30

(1984) defined this battery of tests in order to
differentiate older individuals with mild SDAT from normal
older adults. Logical memory is assessed by asking the
participant to recall a prose passage after it is read to
them. Trailmaking A is a neuropsychological test involving
tracing lines from one circled number to another in
consecutive order. Word fluency is assessed by counting how
many words a participant could think of that began with the
letter S and P in a 60 second time limit. Mental control is
assessed by determining how quickly and accurately the
participants can recite the alphabet, count backwards from 20
to 1, and count by threes (1, 4, 7, etc). Following the
equation given by Storandt et al. (1984), a participant’s raw
score on the test is multiplied by the corresponding
unstandardized coefficient and then summed with the constant
to produce the canonical score. Those with canonical scores
equal or greater than 0 are classified as having dementia,
and those scoring less than 0 were classified as normal. The
norms on each test are as follows: (a) Mental Control
demented = 4.50; normal = 7.21, (b) Logical Memory demented =
2.02; normal = 8.65, (c) Trailmaking A demented = 12.40;
normal = 25.48, and (d) Word Fluency demented = 13.29; normal
= 27.07. Major analysis involved a stepwise discriminant
function analysis designed to determine which measure and
their weights would discriminate maximally between subjects
with mild SDAT and normal controls. The first discriminate

analysis conducted with 42 subjects correctly classified all

31

subjects on the basis of a discriminate function using the
four measures mentioned above. Applying weights obtained
from this analysis to the cross-validation sample (n = 42),
all but 2 persons were classified correctly (98%). The two
samples were then combined to obtain more stable
coefficients.

The Mini Mental State (MMSE) is a mental status
examination developed to provide a practical method for
grading the cognitive state of patients using a brief test
designed to assist clinicians in recognizing cognitive
impairment (Folstein & McHugh, 1975). The MMSE is divided
into two sections; the first covers orientation, memory, and
attention, and the second covers the ability to name, follow
written and verbal commands, write a sentence spontaneously,
and copy a simple design. Folstein et al. (1975) used 206
patients with dementia syndromes, affective disorder with or
without cognitive impairment, mania, schizophrenia, and
personality disorder, as well as 63 normal subjects to test
the reliability and validity of the MMSE. In validating the
MMSE, it was determined that the MMSE scores agreed with the
clinical opinion of the presence of cognitive difficulty and
as the cognitive difficulty is usually less is depression
than in dementia the scores dispersed in a fashion agreeing
the severity of the illness. Scores were not due to age
effects and unrelated clinical conditions (M = 24.5 for
patients with depression under 60 years; M = 25.7 for

patients with depression over 60 years). Graphs show that

32

the change over time in the MMSE score matches the actual
improvement or decline in cognitive state. Folstein et al.
determined that the MMSE is reliable on a 24 hour or 28 day
retest by single or multiple examiners (Pearson r = .887 when
given 24 hours apart; r = .98 when given 28 days apart).

The maximum score on the 5-10 minute test is 30.
Bleecker, Bolla-Wilson, Kawas, and Agnew (1988) established
age-specific norms for the MMSE, and the lowest quartile
cutoff scores by decade are as follows: 603 = 28; 70s = 28;
and 803 = 26. Using these age—specific norms provides
greater sensitivity than the recommended cutoff score of less
than 24.

The final instrument to used in this study assessed
depression using a self—report format. The Beck Depression
Inventory (BDI) consists of 21 self~report symptoms and
attitudes which are rated from 0 to 3 in terms of intensity
by the participant (Beck, Steer, & Garbin, 1988). The cut-
off scores established to distinguish those who have
affective disorder from those who do not are as follows: (a)
less than 10 is considered minimal depression, (b) 10-18 is
considered mild to moderate depression, and (c) 30-63 is

considered to be severe depression (Beck et al., 1988). The

mean BDI scores are as follows: (a) minimal 10.9, (b) mild

= 18.7, (c) moderate = 18.7, and (d) severe 30.0 (Beck et
al., 1988). Beck et al. (1988) report that for psychiatric
populations, the mean coefficient alpha from 25 previous

studies for internal consistency was .86. For 15

33

nonpsychiatric studies, the mean alpha for internal
consistency was .81. Split-half reliability has been shown
to be .93, and test-retest reliability is .74 for 3 months.
The BDI correlates well with other symptom inventories and
can discriminate between diagnostic groups. For a detailed
discussion on the excellent content, concurrent,
discriminant, construct, and factorial validity of the BDI,
Beck et al. (1988) have reviewed the psychometric properties
of the inventory.

Statistical Analysis

Pearson Product Moment Correlational analysis and Chi
Square analysis were used to determine whether there were
statistically significant differences between proportions of
the following variables and compliance--memory loss,
depression, complexity of regimen, and age. These analyses
examined whether there was a signficant relationship between
any of the preceeding variables and the ability to comply
with medication instructions.

Multiple regression was used in order to study the
ability to comply as a function of these variables of
interest: (a) age, (b) complexity of dosing regimen, (c)
presence of dementia and/or depression. Employing such a
strategy yielded results indicating whether these independent
variables were correlated with each other as well as with the
dependent variable--the ability to comply.

This method of statistical analysis determined the

following: (+a) the tendency of the variables to covary, (b)

34

whether compliance is associated with each of the independent
variables, (c) whether an independent variable is a surrogate
for another independent variable in the relationship with
compliance, and (d) whether an independent variable in
enhancing the effect of another independent variable on the
ability to comply.

Therefore, multiple regression was used to determine the
strength and nature of the relationships among the
variables-~age, complexity, depression, dementia, and

compliance.

RESULTS

Data for the present study was collected during the
summer and fall of 1990. The 40 participants in the study
ranged in age from 60 to 92 years and included 37 females and
3 males. The average age of the participants was 75.4 years.
Descriptive Statistics

On the Beck Depression Inventory (BDI), 50% of
participants were classified as depressed, with 16
participants scoring in the mildly depressed range, 3 in the
moderately depressed range, and 1 in the severely depressed
range. The mean BDI score of the 40 participants was 10,
which is in the mildly depressed range.

On the Mini Mental State Examination, the majority of
the participants scored in the normal range of cognitive
functioning; 9 scored in the demented range. The mean score
on the Mini Mental State was 26.1, which is indicative of
normal cognitive functioning. On the Senile Dementia of the
Alzheimer Type (SDAT) battery, scores ranged from -3.160 to
+2.103, and the mean score was -0.226. Scores less than 0
are considered normal, and scores greater than 0 are
indicative of memory impairment. Seventeen participants
(42.5%) scored in the demented range.

Total number of chronic diagnoses per participant ranged
from 1 to 9, and the mean number of diagnoses was 4.7.

Total number of scheduled prescribed medications per
participant ranged from 1 to 11, and the mean number of

35

36

scheduled drugs was 4.3. Fifty-five percent of the
participants received 4 or less prescriptions.

Nine of 19 participants questioned used some type of
memory aid to help them remember to take their medications as
instructed. Twenty-one of 27 participants questioned
indicated that they tried to comply with medication
instructions from physicians. Only 6 out of 27 admitted
intentional noncompliance.

Complexity, based on the total number of pills a
participant would have to take each day, ranged from 1 pill
to 18 pills. The mean number of pills participants took
every day was 6.4 pills. Of the 40 participants, 72% took 8
or less pills each day.

Compliance percentages ranged from -25% to +100%, with
perfect compliance equalling 0, hypercompliance being —1 to
~100 and hypocompliance being +1 to +100. Compliance
percentages were calculated using a formula previously
discussed. Readers should note that there is a fundamental
lack of any agreed definition of noncompliance, and various
studies have described the noncompliant patient as one who
takes less than 3%, 50%, or 90% of prescribed drugs.
Patients in this study were considered to be compliant if
they took at least 90% of their prescribed medications
correctly. Fourteen participants met this criterion for
acceptable compliance. The mean compliance percentage was
+14.484, which is in the range of noncompliance with drug

regimens due to missed doses.

37

Correlational Analysis With Pearson Product Moment and Chi
Square
Pearson Product Moment and Chi Square analyses were
conducted to determine whether there were statistically
significant differences between proportions. Table 1
presents the findings of Pearson Product Moment analysis.
Table 1

Pearson Product Moment Correlations for Compliance
With Independent Variables

 

MMSE Score SDAT Score BDI Score Complexity Age
Comply -.03 .09 .18 -.06 -.05

N 40

 

Hypothesis 1 stated that there would be a statistically
significant correlation between memory loss and compliance.
Pearson Product Moment correlation analysis did not approach
statistical significance (3 = .09, r = -.03).

Additional analysis using Chi Square was not
statistically significant though the results were in the
expected direction (Chi Square(1) = 1.78, p = .18). In
analyzing overall dementia score (MMSE and SDAT battery) and
the ability to comply, the hypothesized relationship between
memory loss and compliance was not supported. Participants
falling within the demented range split almost equally
between either taking at least 90% of their medication

correctly or not complying with dosing regimens.

38

Furthermore, participants with memory loss who did comply
significantly outnumbered those participants without memory
loss who were also compliant. Participants without memory
problems were more likely to be noncompliant than compliant.
When Chi Square analysis was conducted using only the
score from the Mini Mental State Examination, a possible
relationship between memory loss and the ability to comply

was discovered (Chi SquareLl) = 3.17, p = .07). Twenty of

 

the 28 participants who scored in the normal range on the
Mini Mental State were not compliant with prescribed
medications. Scores indicate that those without memory
problems as measured by the MMSE were significantly less
compliant than their demented counterparts.

Statistical significance was achieved for an additional
analysis conducted using Chi Square which examined the
relationship between score on the MMSE and the ability to
comply with dosing regimens 90% of the time while controlling
for intent to comply. For those who intended to comply, Chi
Square(1) = 6.11, p = .01. Clearly, those who intended not
to comply were intentional noncompliers who did not take
their medications as instructed. Significant findings
occurred when analyzing participants who indicated that they
intended to comply with dosing schedules. Participants
without memory problems who intended to comply were
significantly less compliant, while participants with memory
problems who intended to comply were significantly more

compliant.

39

Hypothesis 2 stated that there would be a statistically
significant relationship between compliance and depression.
Pearson Product Moment correlation analysis did not approach
statistical significance (r = .18). However, additional
analysis using Chi Square yielded a result that does approach

significance (Chi Square(1) = 2.67, p = .10). Participants

 

who were not depressed were just as likely to be compliant as
noncompliant. However, those participants who were depressed
were more likely to be noncompliant than compliant.

Hypothesis 3 stated that there would be a statistically
significant correlation between the presence of both dementia
and depression and the ability to comply with medication
regimens. Chi Square analysis was conducted to determine the
relationship between memory loss and depression while
controlling for compliance. For those were compliant, Chi
Square(1) = 1.25, p = .26. As in all other analyses,
compliance was defined as taking medication correctly 90% of
the time. Findings, although not statistically significant,
suggest that those who are not demented or depressed account
for the largest number of noncompliers. Interestingly, those
who are both depressed and demented did not appear to be at a
greater risk for noncompliance.

Hypothesis 4 stated that there would be a statistically
significant correlation between complexity of dosing regimen
and compliance. This hypothesis was not supported by Pearson
Product Moment analysis (; = -.06), although the findings

were in the expected direction.

40

Hypothesis 5 stated that there would be a statistically
significant relationship between age of participant and
compliance. Analysis determined that there was no
statistically significant correlation between age of
participant and the ability to comply with dosing schedules
(g = -.O5).

There was a significant relationship between age of
participant and score on the Mini Mental State Examination
(3 = —.34, p = .05). This finding can be described as an
inverse relationship between the variables, with score on the
MMSE decreasing significantly as age of participant
increases.

Analysis With T-Tests

In order to determine whether there were statistically
significant differences between group means, t~tests were
conducted on memory strategies and intent and the following
variables—~depression, memory loss, complexity, and
compliance. This was a post-hoc analysis without specific

hypotheses, and results are shown in Tables 2 and 3.

41

Table 2

Comparison of participants using memory strategies with those
not using memory strategies on each of the following
variables: depression, memory loss, complexity of dosing
regimen, and ability to comply.

 

fl Mean §D L Q:
BDI Score
Uses memory aids 9 8.89 4.94 -.66 17
No memory aids 10 11.10 8.86
SDAT Score
Uses memory aids 9 —.47 1.75 —.84 17
No memory aids 10 .08 .99
Compliance
Uses memory aids 9 8.62 16.00 -1.02 17
No memory aids 10 17.75 21.96
Complexity
Uses memory aids 9 7.00 3.90 1.07 17

No memory aids 10 5.10 3.81

 

42

Table 3

Comparison of participants intending to comply with
instructions with those intending not to comply with
instructions on each of the following variables: depression,
memory loss, complexity of dosing regimen, and ability to

comply.

 

.1! _M e._a_._n_ 5.2 :9 if
BDI Score
Intend to comply 21 8.90 5.32 -1.90 25
Intend not to comply 6 14.50 9.40
SDAT Score
Intend to comply 21 -.27 1.45 —.10 25
Intend not to comply 6 -.21 .83

Compliance
Intend to comply 21 7.05 16.11 -4.66* 25

Intend not to comply 6 50.13 30.98

Complexity
Intend to comply 21 6.17 4.30 .00 25
Intend not to comply 6 6.17 3.31

*p = .001

 

There was one statistically significant difference
between the independent variable, intent to comply with
medication instructions, and actual compliance, the dependent
variable (3(251 = —4.66, p = .001). Intent to comply was
determined by asking participants whether they attempted to
comply with prescribed medications and identifying specific
reasons for intentional noncompliance. Those participants
who stated their intent to comply were significantly more
compliant than those who were intentionally noncompliant.

There was not a statistically significant difference

between scores on the EDI and using memory strategies.

43

Memory strategies were defined as any aid participants
admitted to that made it easier to remember to take
medication in the right dose at the right time. Although the
difference was not significant, participants who were not
depressed used more memory strategies than those who were
depressed.

There was not a statistically significant difference
between noncompliance score and using memory strategies. The
difference observed did suggest that those who used memory
strategies were more compliant (+8.623) than those who did
not use memory strategies (+17.526).

There was not a statistically significant difference
between complexity of dosing regimen and using memory
strategies. Although the difference was not significant,
those with more complicated dosing regimens used more memory
strategies than those with less complicated regimens. There

was not a significant difference between intent to
comply with medication instructions and scores on the BDI.
Even so, those who intended to comply had a mean depression
score in the normal range, while those who were intentional
noncompliers had a mean depression score in the mildly
depressed range.

Regression Analysis

An analysis of variance was conducted using stepwise
multiple regression. Hypothesis 6 stated that the variables
(memory impairment, depression, complexity of dosing regimen,

age) would account for a significant proportion of the

44

variance. This hypothesis was not supported. None of the
variables fit into a step-wise regression model of

compliance.

DISCUSSION

Compliance with prescribed medication therapy is
essential for successful medical care. Studies have
estimated that a large percentage of the elderly population
does not follow prescribed treatment, and physicians are
often the last to know when patients are noncompliant.
Misuse of prescribed drugs-~taking too much or too little, at
the wrong time or in the wrong way—~can cause treatment
failures and pose serious threats to health.

Variables hypothesized to affect unintentional
noncompliance include poor cognitive functioning, depressed
mood, complex dosing regimens, and a host of others. While
some studies have attempted to identify correlates of poor
compliance, results have been inconclusive. Most studies
have relied on patient self~report in order to measure rates
of compliance. Yet it is well documented that patients
generally overstate the extent of their compliance.

Several studies attempted to examine compliance rates
using pill counts but did not isolate specific variables that
may have been related to noncompliance. Other studies have
used more sophisticated methods for determining compliance,
such as measuring blood levels of medication, but most
control for specific disease populations and do not attempt
to correlate variable to rates of compliance. Ideally,
researchers should specify for each disease and treatment the

degree of departure from medication instructions known to be

45

46

associated with a clinically significant deterioration in
condition in order to define what must be considered
noncompliance with prescribed treatment regimens.

The intent of this study was to isolate factors that
would identify those elderly patients who are at risk for
unintentional noncompliance with prescribed medication
regimens. The assessment procedure made use of pill counts
and standardized assessment instruments to collect data on
specific variables hypothesized to contribute to
unintentional noncompliance in the community—dwelling elderly
population. Variables hypothesized to
be related to noncompliance included cognitive function,
emotional status, age of participant, and complexity of
dosing regimen. Furthermore, intent to comply and use of
memory aids were included in the final analysis.

Results indicated that slightly over half of the
participants were not experiencing significant memory
problems. Among those with memory problems, most declines
appear to be attributable to either late—life forgetting or
to the early stages of a dementing condition. It is unclear
whether there was a selection bias in that participants who
volunteered may have had fewer memory problems than other
community-dwelling elderly. The proportion of participants
experiencing memory impairment in this sample is in the range
quoted by other studies.

Results also indicate that the elderly living

independently in the community often experience a mildly

47

depressed mood. The prevalence rate of depression in this
study was higher than that found in other studies, but 16 of
the 20 who were depressed scored in the mildly depressed
range.

Almost all participants questioned stated that they felt
it was important to comply with instructions given by
physicians and that they tried to take their medication as
prescribed. Several participants did admit to intentional
noncompliance, but this proportion was lower than expected.
It is possible that the participants wanted to look good for
the investigators and did not accurately report their actual
pill-taking behavior. Those who did intentionally noncomply
cited factors that are substantiated in other research.
Reasons given included cost of medication, busy schedule, and
adverse side effects of medication.

Results indicated that the elderly who reside
independently are able to care for themselves even if they
suffer from mild memory loss and/or depression. Most
participants were fairly well informed about their diagnoses
and prescriptions, and it may be possible that physicians and
other health care providers do not give these patients enough
credit for their ability to cope successfully. Some
participants documented their condition, allergies, and
treatment protocol with elaborate notes. Almost half used
one or more memory aids to remind them to take pills in the
right doses at the right times. Most kept fewer than 10

over-the-counter medications in the house, and few kept

48

expired medication. Those with chronic medical conditions
who were compliant with 90% of their prescribed pills had
developed a routine to cope effectively with any number of
prescribed medications. Nevertheless, the sample was not
compliant (65%); missed doses accounted for the majority of
noncompliers. This percentage is supported by findings of
other studies.

This study tested the hypothesis that age and complexity
of dosing regimen are correlated with the ability to comply
with prescribed medication instructions. Results indicate
that the age of participants and the complexity of dosing
regimen do not significantly affect the ability of elderly
participants to comply with dosing schedules. Although it is
not surprising that age was not related to compliance,
it is surprising that complexity of dosing regimen was not
correlated with ability to comply.

Most studies have found that as complexity of regimen
increases, compliance rates decrease. It is possible that
this sample had less complex schedules because they were
prescribed less medication than samples from other studies.
Since over half of the 40 patients who participated in this
study took 4 or less prescribed drugs, it may be possible
that a more medicated sample would have yielded statistically
significant results. However, since this sample was two
thirds noncompliant, it appears that variables other than
complexity were important risk factors.

This study hypothesized that the elderly who are

49

depressed will be significantly less compliant with
prescribed medication regimens. Results indicate that
elderly patients who are depressed are more likely to be
noncompliant. The findings, while not statistically
significant, appear to be clinically meaningful, since 15 of
the 20 depressed participants were not taking their
medications as prescribed. Furthermore, since the majority
of these patients scored in the mildly depressed range, it is
likely that the number of noncompliers would escalate as
intensity of depression increases.

This study hypothesized that there would be a
statistically significant relationship between memory
impairment in the elderly and ability to remember to take
prescribed medication. When analyzing the effect of memory
loss on the ability to comply, results were not clear cut.
When an overall memory score using both assessment tests of
cognitive functioning was correlated with compliance
percentage, there appeared to be no statistically significant
relationship between memory impairment and the ability to
comply.

However, when scores on the Mini Mental State alone were
correlated with compliance, a possible relationship was
discovered. It appears that for this sample of relatively
able elderly, those without memory problems are less
compliant than those with memory problems. It is possible
that those elderly patients who recognize that they are

experiencing memory loss try to compensate for their

50

declining cognitive functions. Perhaps they perseverate in
that they persistently remind themselves to comply, resulting
in the counterintuitive result of increased compliance.
Paradoxically, those individuals who are not experiencing
significant memory impairment appear to be at greater risk
for noncompliance.

There is further evidence for this speculation when
intent to comply is controlled in the analysis of scores on
the Mini Mental State and compliance percentage. Elderly
patients who suffer from memory impairment and intend to
comply do so at a statistically significant higher rate than
their nondemented counterparts. Even though just as many
nondemented participants said they meant to take medication
as prescribed, they were significantly less successful than
those with memory impairment who also indicated they tried to
comply. All but one of the admitted intentional noncompliers
were individuals who were not cognitively impaired. It is
possible that those with memory impairment rely on memory
aids much more often and with more success than more
cognitively healthy elderly patients. This could not be
accurately assessed from the data gathered in this study.

It was also hypothesized that the presence of both
dementia and depression would adversely affect the ability to
take medication as prescribed. Findings lean in the opposite
direction. While not statistically significant, it appears
that those who are not depressed or demented are actually

worse at complying, while those individuals who are both

 

51

depressed and demented do not appear to be at greater risk
for noncompliance. Although the sample was not severely
depressed or demented, this population may be representative
of a more competently functioning elderly population than has
been tested previously. It should be noted that this result
was obtained when only the Mini Mental State examination, a
measure of global cognitive functioning, was used as the
measure of impairment. Analysis of the Senile Dementia of

the Alzheimer Type (SDAT) battery alone as well as the SDAT

 

battery and the MMSE did not yield statistically significant
results. '

There were no hypotheses made about how the intentions
of elderly patients would affect medication taking behavior.
Nonetheless, it appears that intention is very important when
analyzing why the elderly patient may not comply with
medication regimens. As discussed above, when intent to
comply was controlled in analyzing memory impairment and
compliance percentages, significant results were achieved.
There was also a difference, although not significant,
between intent to comply and depression scores. Those who
are depressed may be intentional noncompliers--those patients
who do not take some or all of their medication for often
legitimate reasons. This may be the result of a more
negative mind-set that makes depressed individuals see the
world in realistic terms. If this is true, it is possible
that even mildly depressed individuals pay more attention to

extraneous factors such as the cost of medication, adverse

52

side effects, and drug interactions. It also is likely that
intentional noncompliance is underreported, since the
interviewers saw clear evidence that many participants in
this study intentionally noncomplied, yet when questioned
specifically did not admit to this.

There were also no specific hypotheses made about the
effect of using memory strategies on compliance. When this
variable was analyzed in conjunction with depression scores,
noncompliance percentages, and complexity ratings, results
did not reach statistical significance but are interesting.
Participants who were not depressed relied on memory
strategies more often than those who were depressed. Not
surprisingly, those who used memory aids of some type were
more compliant than those who tried to remember to take
medication without any aid. It also seems that those with
more complex dosing regimens rely on memory aids more often
than those taking less prescribed pills.

The results of this study indicate that there is much
more work to be done in the area of compliance with medical
treatment. Overall, the participants in this study were
noncompliant with prescribed medication regimens. It seems
more likely for depressed elderly patients to be at risk for
noncompliance. Elderly participants who are demented are
sometimes those with almost perfect compliance ratings.
Obviously these individuals are finding a way to remember to
take their medications.

It is possible to speculate about reasons for

53

noncompliance among elderly patients who are not cognitively
impaired. Older patients may be less likely to challenge the
authority of physicians and may lead them to believe they
will be compliant with prescribed drug therapy. All of the
participants questioned answered affirmatively when asked
whether they felt patients should always try to follow
instructions given by physicians. In reality, these patients
may not always have complete confidence in their treatment
protocol. Perhaps the more cognitively intact elderly
patient is more likely to question the appropriateness of
prescribed drug therapy and/or dosage and will noncomply more
often than those who are experiencing problems with cognitive
functioning. Depressed patients may also be more likely to
question drug therapy because of their more negative outlook
in general.

Therefore, one of the most important aspects of
improving compliance may be patient-physician communication.
Richardson (1986) suggests that social-psychological models
that contain cognitive variables such as motivation and
expectation may be more helpful in understanding reasons for
some forms of noncompliance. The study seems to support this
contention, as many of the participants who were noncompliant
in this study were not depressed or demented and did not seem
to be affected by age or complexity of dosing regimen.

The results of this study suggest that unintentional
noncompliance may be exacerbated by a depressed mood. Those

who are depressed do appear to be at risk for noncompliance.

54

Therefore, it may be helpful for physicians to utilize the
Beck Depression Inventory in order to assess levels of
depression, since even a mildly depressed mood appears to
have a clinically meaningful effect on compliance rates.

Late-life forgetting and mild forms of memory impairment
common among the able elderly living independently in the
community do not appear to be risk factors for unintentional
noncompliance. Those who have trouble remembering things
seek more assistance than those with normal recall.
Even so, it may be helpful to administer a mental status test
such as the Mini Mental State examination. Those who do
score in the demented range on this instrument should be
carefully counselled as to how important their drug therapy
is, since ability to comply was significantly affected by
intent to do so. The provider may also choose to assess
whether the patient is using reminder strategies, since this
study showed that elderly patients with memory impairment who
were successful in taking prescribed drugs relied on memory
aids. It is interesting that some studies have found that
patients cite "forgetfulness" when they are questioned about
reasons for noncompliance. This study found that this
"forgetfulness" is probably not due to any organic
deterioration or confused mental state but due instead to
psychological variables. It is possible that patients are
very selective in forgetting.

The results of this study also suggest that those

elderly patients who are cognitively intact and not depressed

55

are actually more at risk for noncompliance to prescribed
therapy. It seems likely these are the patients who are
intentionally noncomplying, even though they self-report that
they intend to comply.
Summary of Findings

In summary, our major findings are as follows:
(a) There was no direct relationship between memory loss and
compliance until there was statistical control for intent to
comply. Then there was a statistically significant
relationship between overall functioning and compliance.
Specifically, participants without memory problems were less
compliant. The relationship between scores on the Mini
Mental State and compliance approached statistical
significance suggesting that participants without memory
problems were less compliant.
(b) The relationship between compliance and depression
approached statistical significance indicating that depressed
participants are more likely to be noncompliant.
(c) The expected relationship between both dementia and
depression and the ability to comply was not supported.
Specifically, those who were not depressed or demented were
just as likely to be noncompliant as those experiencing
depression and/or dementia.
(d) The intent to comply with medication instructions is
related to actual compliance. Specifically, those who
admitted to intentional noncompliance were not compliant with

prescribed regimens.

56

(e) The hypothesized relationships between age and compliance

and complexity and compliance were not supported.

REFERENCES

REFERENCES

American Geriatrics Society Public Policy Committee. (1989).

Comprehensive geriatric assessment. Journal of the
American Geriatrics Society, 31, 473-474.

 

W.B. (1987). Use of assessment instruments in

Applegate,
Journal of the American Geriatrics

clinical settings.
Society, 33, 45-50.

Ascione, F.J., & Shimp. L.A. (1984). The effectiveness of
four education strategies in the elderly. Drug
Intelligence_§pd Clinical Pharmacy, 13, 926-931.

Bayer, A.J., Pathy, J., & Twining, C. (1987). The memory
clinic. Drugs, 33, 84-89.

Beck, A.T., Steer, R.A., & Garbin, M.G. (1988). Psychometric
properties of the Beck depression inventory: Twenty-five
years of evaluation. Clinical Psychology Review, 3, 77-

100.

Becker, M.H. (1985). Patient adherence to prescribed
therapies. Medical Care, 33, 539-555.

Beller, L.A., & Evans, E.R. (1987). Drug therapy in the
elderly: Effects on mental status. American Family

Physicians, 33, 149-152.
Bensen, D.F. (1982). Clinical aspects of dementia, p. 231-

232. In J.C. Beck (moderator). Dementia in the elderly:
The silent epidemic. Annals of Internal Medicine, 31, 231-

241.
Bergmann, K., Foster, E.M., Justice, A.W., & Matthews, V.

(1978). Management of the demented elderly patient in
the community. British Journal of Psychigtry, 132, 441-

449.

Black, K.S., & Hughes, P.L. (1987). Alzheimer’s disease:
Making the diagnosis. American Family Publishers, 33,
198-202.

Blazer, D., Hugher, D.C., & George, L.K. (1987). The

epidemiology of depression in an elderly community
population. The Geronotologiscal Society of America, 31,

281-287.
58

59

Bleecker, M.L., Bolla—Wilson, K., Kawas, C., & Agnew, J.

(1988). Age-specific norms for the Mini-mental state exam.
Neurology, 33, 1565-1568.

Cavanaugh, J.C., Grady, J.C., & Perlmutter, M. (1983).
Forgetting and use of memory aids in 20 to 70 year olds

everyday life. International Journal of Agingyand Human
Develgpment, 31, 113-122.

Cavanaugh, S., & Wettstein, R.M. (1983). The relationship
between severity of depreSsion, cognitive dysfunction, and

age in medical inpatients. American Journal of Psychiatry,

Ciocon, J.O., & Potter, J.F. (1988). Age-related changes in
human memory: Normal and abnormal. Geriatrics, 33, 43-48.

Cohen, J., & Cohen, P. (1975). Applied multiple regression/
correlational analysis for the behavioral sciences. New
York: Halsted Press.

Currie, C.T., Moore, J .T., Friedman, S.W., & Warshaw, G.A.
(1981). Assessment of elderly patients at home: A report
of fifty cases. Journal of the American Geriatrica
Societ , 33,398-401.

Darnell, J.C., Murray, M.D., Martz, B.L., & Weinberger, M.
(1986). Medication use by ambulatory elderly. Journal
of the American Geriatrics Society, 33, 1-4.

Davis, M.S., & Eichhorn, R.L. (1960). Compliance with medical

regimens: A panel study. Journal of Health and Human
Behavior, 241-249.

Davis, P.B., & Robins, L.N. (1989). History-taking in the
elderly with and without cognitive impairment. Journal of
the American Geriatrics Society, 31, 249-255.

 

Dunn, V.K., & Sacco, W.P. (1988). Effect of instructional set

on responses to the Geriatric depression scale. Psychology
and Aging, 3, 315-316.

Falvo, D.R. (1985). Effective patient education. Rockville:
Aspen Publishers.

Farber, J.F., Schmitt, F.A., & Logue, P.E. (1988). Predicting
intellectual level from the Mini—mental state examination.
Journal of the Aaerican Geriatrics Society, §§: 509-510.

Fedder, D.O. (1984). Drug use in the elderly: Issues of

noncompliance. Drug_Intelligence and Clinical Pharmac ,
33, 158-162.

Feinberg, T., & Goodman, B. (1984). Affective illness,

60

dementia, and pseudodementia. Journal of Clinical
Psychiatry, 33, 99-103.

Fitten, L.J., Morley, J.E., Gross, P.L., Petry, S.D., & Cole,
K.D. (1989). Depression. Journal of the American
Geriatrics Society, 31, 459-472.

Folkman, S., Bernstein, L., & Lazarus, R.S. (1987). Stress
processes and the misuse of drugs in older adults.
Psychology and Aging, 3, 366-374.

Folstein, M., Anthony, J.C., Parhad, I., Duffy, 8., &
Gruenberg, E.M. (1985). The meaning of cognitive
impairment in the elderly. Journal of the Americam
Germatrics Society, 33, 228-235.

Folstein, M.F., Folstein, S.E., & McHugh, P.E. (1975). Mini-

mental state. Journal of Psychiatric Research, 13, 189—
198.

Freedman, N., Bucci. W., & Elkowitz, E. (1982). Depression in
a family practice elderly population. Journal of the
American Geriatrics Society, 33, 372-377.

Gabriel, M., Gagnon, J.F., & Bryan, C.K. (1977). Improved
patient compliance through use of a daily drug reminder
chart. American Journal of Public Health, 31, 968-969.

German, P.S., Klein, L.E., McPhee, S.J., & Smith, C.R.
(1982). Knowledge of and compliance with drug regimens in

the elderly. Journal of the American Geriatricaasociety,
33, 568-571.

Good, W.R., Vlachonikolis, I., Griffiths, P., & Griffiths,
R.A. (1987). The structure of depressive symptoms in the
elderly. Britiah Journal of Paychiatry, 150, 463-470.

Griffths, R.A., Good, W.R., Watson, N.P., O’Donnell, H.F.,
Fell, P.J., & Shakepspeare, J.M. (1987). Depression,

dementia and disability in the elderly. Britiah Journal of
Psychiatry, 150, 482-493.

Haggerty, J.J., Golden, R.N., Dvans, D.L., & Janowsky, D.S.

(1988). Differential deagnosis of pseudodementia in the
elderly. Geriatrics, 13, 61—74.

Haynes, R.B., Taylor, D.W., & Sackett, D.L. (1979).

Compliance in haalth care. Baltimore: Johns Hopkins
University Press.

Hays, R.B., & DiMatteo, M.R. (1987). Key issues and
suggestions for patient compliance assessment: Sources of
information, focus of measures, and nature of response
options. The Journal of Compliance in Health Cama, 3, 37-

 

61

53.

Heston. L.L., Mastri, A.R., Anderson, E., & White, J. (1981).
Dementia of the alzheimer type. Archives of Genera1

Hurd, P.D., & Butkovich, S.L. (1986). Compliance problems and
the older patient: Assessing functional limitations. Drug
Intelligence_and Clinical Pharmacy, 33, 228-231.

Ives, R.J., Bentz, E.J., & Gwyther, R.E. (1987). Drug-related
admissions to a family medicine inpatient service.
Archives of Internal Medicine, 147, 1117-1120.

Johnston, M., Clarki, A., Mundy, K., Cromarty, E., & Ridout,
K. (1986). Facilitating comprehension of descharge

medication in elderly patients. Age and Ageing, 13, 304-
306.

Kafonek, S., Ettinger, W.J., Roca, R., Kittner, S., Taylor,
N., & German, P.S. (1989). Instruments for screening for
depression and dementia in a long-term care facility.
Journal of the AmerigamfiGer1atrics Society, 31, 29-34.

Kazis, L.E., & Friedman, R.H. (1988). Improving medication
compliance in the elderly. Journal of the Ameriqam
Geriatrics Society, 33, 1161-1162.

Kinzie, J.D., Lewinsohn, P., Maricle, R., & Teri, L. (1986).
The relationship of depression to medical illness in an

older community population. Campyehensive Psychiatyy,
31, 241-246.

Klein, L.E., roca, R.P., McArthur, J., Vogelsang, G., Klein,
G.B., Kirby, S.M., & Folstein, M. (1985). Diagnosing

dementia. Journal of the Amerigan Geriatrics Society, 33,
483-488.

Koenig, H.G., Meador, K.G., Cohen, H.J., & Blazer, D.G.
(1988). Self-rated depression scales and screening for
major depression in the older hospitalized patient with
medical illness. Journal of the American Gey1atriqa
Society, 33, 699-706.

Landry, J.A., Smyer, M.A., Tubman, J.G., Lago, d.J., Roberts,
J., & Simonson, W. (1988). Validation of two methods of
data collection of self-reported medicine use among the

elderly. The Gerontological Society of America, 33, 672—
676.

Larson, E.B., Kukull, W.A., Buchner, D., & Reifler, B.V.
(1987). Adverse drug reactions associated with global
cognitive impairment in elderly persons. Anna1a19f
Internal Medicine, 131, 169-173.

 

62

Larson, E.B., Reifler, B.V., Featherstone, H.J., & English,
D.R. (1984). Dementia in elderly outpatients: A
prospective study. Annals of Internal Medicine, 100, 417-
423.

Law, R., & Chalmers, C. (1976). Medicines and elderly people:

A general practice survey. British Medical Journa1, 3,
565-568.

Lay, P. (1986). Cognitive variables and noncompliance. The
Journal of Compliance in Health Care, 1, 171-188.

Lazarus, L.W., Newton, N., Cohler, B., Lesser, J., & Schweon,
C. (1987). Frequency and presentation of depression
symptoms in patients with primary degenerative dementia.
American Journal of Psyahiatr , 133, 41-45.

Magi, G., Palazzolo, 0., & Bianchin, G.(1988). The course of
depression in elderly outpatients. Canadian Journal of
Psychiatry, 33, 21-24.

Malahy, B. (1966). The effect of instruction and labeling on
the number of medication errors made by patients at home.
American Journal of Hospital Pharmacy, 33, 283-292.

Marston, M. (1970). Compliance with medical regimens: A
review of the literature. Nursing_Research, 13, 312-323.

Martin, D.C., & Mead, K. (1982). Reducing medication errors
in a geriatric population. Journal of the American

McCartney, J.R., & Palmateer, L.M. (1985). Assessment of
cognitive deficit in geriatric patients. Journa1 of
the American Geriatrics Society, 33, 467—471.

McInytre, L., & Frank, J. (1987). Evaluation of the demented
patient. The Journal of Family Practice, 33, 399-404.

Meichenbaum, D., & Turk, D.G. (1987). Facilitating treatment
adherence. New York: Plenum Press.

Meyer, M.E., & Schuna, A.A. (1989). Assessment of geriatric
patients’ functional ability of take medication. Drug
Intelligence and Clinical Pharmac , 33, 171-174.

Morgan, T.O., Nowson, c., Murphy, J., & Snowden, R. (1986).
Compliance and the elderly hypertensive. Dru s, 31, 174-
183.

Morrell, R.W., Park. D.C., & Poon. L.W. (1989). Quality of
instructions on prescription drug labels: Effects on
memory and comphehension in young and old adults. 133
Gerontologica1 Societyyof Amarica, 33, 345-354.

63

Morrow, D., Leirer, V., & Sheikh, J. (1988). Adherence and
medication instructions. Journal of the American
Geriatrics Society, 33, 1147-1160.

Mulley, G.P. (1986). Differential diagnosis of dementia.
Brit1§h Medical Journal, 292, 1416-1418.

National Institute on Aging. (1980). Senility reconsidered.
Journal of the American Medical Association, 244, 259-263.

Neely, E., & Patrick, M.L. (1968).

Problems of aged persons
taking medications at home.

Nursing Research, 11, 52-55.

Pace, W.D. (1989). Geriatric assessment in the office
setting. Geriatrics, AA, 29-35.

Parmelii, P.A., Lawton, M.P., & Katz, I.R. (1989).

Psychometric properties of the geriatric depression scale
among the institutionalized aged. Journal of Consulting
and Clinical Psychology, 1, 331-338.

Pfeiffer, E. (1975). A short portable mental status

questionnaire for the assessment of organic brain deficit

in elderly patients. Journal of the American Geriatrics
Society, 33, 433-441.

Reding, M., Haycox. J., & Blass, J. (1985).
patients referred to a dementia clinic.
Neurology, A3, 894-896.

Depression in
Archives of

Reifler, B.V., Larson, E., & Hanley, R. (1982). Coexistence
of cognitive impairment and depression in geriatric
outpatients. American Journal of Psychiatry, 139, 623-626.

Reifler, B.V., Larson, E., Teri. L., & Poulsen, M. (1986).

Dementia of the alzheimer’s type and depression. Journal
of the American Geriatrics Society, 31, 855-859.

Richardson, J.L. (1986). Perspectives on compliance with drug

regimens among the elderly. The Journal of Compliance in
Health Care, 1, 33-45.

Rodin, F., & Voshart, K.

ill: An overview.
696-705.

(1986). Depression in the medically
American Journal of Psychiatry, 143,

Rost, K., & Roter, D. (1987). Predictors of recall of
medication regimens and recommendations for lifestyle

change in elderly patients. The Gerontological Society

Salzman, C., & Shader, R.I. (1979). Clinical evaluation of
depression in the elderly. In A. Kaskin, & L.F. Jarvik
(Eds.), Psychiamric Symptoms and Cognitive Losa in the

 

64

Elderly. New York: Wiley.

Schwartz, D., Wang, M., Zeltz, L., & Goss, M. (1962).
Medication errors made by elderly, chronically ill

patients. American Journal of Public Health, 33, 2018-
2029.

Small, G.W., & Jarvik, L.F. (1982).

The dementia syndrome.

Spagnoli, A., Ostino, g., Borga, A.D., D’Ambrosio, R.,
Maggiorotti, P., Todisco, E., Prattichizzo, W., Pia, L., &
Comelli, M. (1989). Drug compliance and unreported drugs

in the elderly. Journal of the American Geriatrics
Societ , 31, 619-624.

Spar, J.E. (1982). Affective disorders and dementia, p. 237.
In J.C. Beck (moderator). Dementia in the elderly: The
silent epidemic. Annals of Internal Medicine, 31, 231-241.

Stern, L.Z., & Bernick, C. (1987). Mental disorders in the
elderly. Comprehensive Therapy, 13, 43-50.

Stewart, R.B., & Cluff, L.E. (1972). A review of medication
errors and compliance in ambulant patients. Clinical
Pharmacology and Therapeutics, 13, 463-468.

Storandt, M., Botwinick, J., Danziger, W.L., Gerg, L., &

Hughes, G.P. (1984). Psychometric differentation of mild

senile dementia of the alzheimer type. Archives of
Neurology, 31, 497-499.

Swift, C.G. (1988). Prescribing in old age.
Journal, 296, 913-915.

British Medical

U’ren, R.C. (1987). Testing older patients’ mental status:
Practical office-based approach. Geriatrics, A3, 49-56.

Weingartner, H., Cohen, R.M., Murphy, D.L., Martello, J., &
Gerdt, C. (1981). Cognitive processes in depression.
Archiveg of General Psyphiatry, 33, 42-47.

Wandless, I., & Davie, J.W. (1977). Can drug compliance in

the elderly be improved? British Medical Journal, 3, 359-
361.

Wiebert, R.T., & Dee, D.A. (1980). 1mproving1patient
madication compliance. New Jersey: Medical Economics Co.

Williamson, J., Stokoe, I.R., Gray, 8., Fisher, M., Smith,

A., McGhee, A., & Stephenson, I. (1964). Old people at
home. Lancet, 1117-1120.

Winograd, C. (1984). Mental status tests and the capacity for

65

self—care.

Journal of the American Geriatrics Society,
33, 49-55.

Wong, E., & Norman, D.C. (1987).

Evaluation of a novel
medication aid,

the calendar blister-pak, and its effect
on drug compliance in a geriatric outpatient population.
Journal of the American Geriatrics Society, 33, 21-26.

Yesavage, J.A., Brink, T.L., Rose, R.L., Lum, 0., Huang, V.,

Adley, M., & Von Leirer, O. (1983). Development and

validation of a geriatric depression screening scale:

A
preliminary report. Journal of Psychiatric Research, 11,
37-49.

   

"11111111111111