HillNIWIHIHWIHIWI

 

E llHililW“MINIMUM”!

143

THESIS

llllllllll‘llllllllllllllllll«llllllllllhlllllllllllllllllll

3 1293 0140

This is to certify that the
thesis entitled

SUBJECTIVE MEMORY COMPLAINTS
AND DEPRESSION IN THE
ABLE ELDERLY

presented by

Michael William Collins

has been accepted towards fulfillment
of the requirements for

MA PSYCHOLOGY

degree in

am

Major professor

Norman Abeles

 

 

 

March 21, 1995

 

MS U is an Affirmative Action/Equal Opportunity Institution

 

 

 

LIBRARY
Michigan State
University

 

 

 

. so

PLACE Ii RETURN BOXtomnwoMchockoutMywrn-cad.
TOAVOIDFINEsmmnonotbdmddoM

DATE DUE DATE DUE DATE DUE

    
  

 

 

 

 

 

 

 

   

mm? .1—{8 2003

Lv‘t ‘

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

L

M80 icAnNflmdivkooNEm-I Opportunityimlion
W1

 

SUBJECTIVE MEMORY COMPLAINTS
AND DEPRESSION IN THE
ABLE ELDERLY
By

Michael W. Collins

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

MASTER OF ARTS

Department of Psychology

1995

ABSTRACT
SUBJECTIVE MEMORY COMPLAINTS
AND DEPRESSION IN THE
ABLE ELDERLY
By

Michael William Collins

This study investigated the relationship between subjective memory complaints and
depression in the able elderly. Participants (11 = 90) for this study were community
dwelling elderly (Mean age = 70) who were offered periodic assessments of their mood
and memory, in addition to, a seven session workshop targeted to teach relaxation or
cognitive strategies for the relief of depression and/or memory difficulties. Level of
depressive symptoms were assessed with the Beck depression Inventory and the Geriatric
Depression Scale while memory complaints were measured with the Memory Assessment
Clinic Self-Report Scale. These test scores were then combined to form an afl‘ective and
somatic factor of depression as well as a sum total of subjective memory complaint.
Significant relationships were found between total subjective memory complaint and the
afl‘ective factor of depression (r = -.53, p < .01) and total subjective memory complaint
and the somatic factor of depression (r = -.39, p < .01). In addition, a profile analysis
revealed distinct depression groups as defined through cutoff scores on these measures.
One such group delineated was a "masked depressed" subgroup (8.8% of the sample).

Suggestions for fixture research as well as implications of these findings were discussed.

ACIGNIOWLEDGMENTS

I would like to express my sincere appreciation to the members of my thesis
committee, Dr. John Hurley and Dr. Robert Caldwell. Dr. Hurley spent a considerable
amount Of time helping me to prepare a quality manuscript as well as sharing insight into
the production of my proposal. Dr. Caldwell was extremely patient in helping me gain a
better understanding and grasp of the content of my thesis. In addition, he was always
accessible to share his statistical knowledge and provide clear descriptions of the edits that
needed to be made. I would especially like to thank Dr. Norman Abeles, my mentor and
committee chairperson. Through his sincere effort and in-depth participation I was able to
complete this work. In addition, Dr. Abeles has provided me with great clinical
Opportunities and experiences to reflect on in my three years of graduate work. He is
readily accessible and always willing to take the time to consult with clinical, research, and
coursework issues. I thank him for all that he has done.

I also want to thank my seven other classmates. We came here knowing very little
about one another and through the hours of studying, reflecting, and sharing that has
drastically changed. We will all reach that common goal.

Finally, Iwantto personallythankallofmyfamilyandmywife, Lynn. Allthe

support that has been given is immeasurable. You all mean so much to me.

TABLE OF CONTENTS

LIST OF TABLES ............................................................................................. vi
INTRODUCTION .............................................................................................. l
Subjective Memory Compliant and Cognitive Decline: A Closer Look 2
The Ubiquitous Role of Depression in Subjective Memory Appraisal ...... 4
Depression in the Elderly ........................................................................ 6
Aging and Masked Depression ................................................................ 8
Hypotheses ............................................................................................. 14
METHOD .......................................................................................................... 17
Participants ............................................................................................. 17
Measures ................................................................................................ 17
RESULTS ........................................................................................................... 22
Psychometric Properties .......................................................................... 22
Hypotheses .............................................................................................. 23
Profile Analysis to Delineate Depression Types ........................................ 24
Post-Hoe Analyses Of the Whole Sample .................................................. 29
DISCUSSION .................................................................................................... 32
Psychometric Properties ........................................................................... 32
Hypotheses .............................................................................................. 33
Profile Analysis to Delineate Depression Types ........................................ 36
The Relationship Between Age and Symptomatology Within the Groups 39
Effects of Sociodemographic Variables on the Depression Groups ............ 4]
Suggestions for Future Research ............................................................... 42
APPENDIX A Item-total Statistic for Affective Factor of Depression .................. 45
APPENDIX B Item-total Statistics for Somatic Factor of Depression ................... 47
APPENDIX C Total Scores on the BDI, GDS, and MAC-S (TOT-FOC-AB)
for Each Individual in the Sample .............................................. 48
LIST OF REFERENCES ...................................................................................... 51

iv

LIST OF TABLES

Table 1 Descriptive Statistics for Memory Complaints Score, Somatic
Factor Score, and Total Affective Score for the Four Depression
Groups in the Sample ........................................................................ 26

Table 2 Descriptive Statistics for Age, Education, and Health Complaints
for the Four Depression Groups in the Sample ................................... 28

Table 3 Frequency Percentage of Gender and Living Arrangements for
the Four Depression Groups in the Sample ......................................... 28

INTRODUCTION

There is considerable publicity in both the popular and scientific press concerning
memory loss in Older adults. Alzheimer's disease (AD), the cause of dementia in 50% of
today’s cases (McDonald & Nemerofl‘, 1991), has been referred to as the silent epidemic
and one of the most pervasive social health problems of our generation (Mclean, 1987).
With the fear Of such insidious diseases as AD and other progressive disorders, the
integrity of one's memory looms large for the elderly population. As Becker, Huff, Nebes,
Holland, and Boiler (1988) stated, memory is usually the initial cognitive fimction to be
altered by AD and other dementing progressive diseases, and is one of the cognitive
firnctions most severely afi‘ected. Intuitively, it is therefore expected that great fear
surrounds the issue of memory loss in the elderly population. One of the major
contributing reasons for this fear of cognitive decline is that older persons do, in fact,
complain more about their memory. This is stressful not only to the individuals concerned
but also to family members. Consequently, it is important to determine whether the
subjective impression of memory firnction might difl‘erentiate between those individuals
with and without a progressive dementing illness. The question posed is this: Are
subjective complaints about memory congruent with an actual decline in memory

firnctioning?

Subjective Memory CompLainLand Cognitive Decline: A Closer Look

This topic has recently received a considerable amount of attention in the scientific
press. Taylor, Miller, and Tinklenburg (1992) performed a longitudinal study in this area.
Using self-report questionnaires and cognitive tests, these investigators found that at the
individual leveL memory performance did not significantly correlate with change in
subjective self report. Brown, Dodrill, Clark, and Zych (1991) looked at whether reliable
relationships could be found between self-report and objective measures of memory in
younger patients with suspected or demonstrated neuropsychological dysfirnction, and
found that memory complaints were not indicative of specific memory dysfunction.
Devolder and Pressley ( 1991) looked at two samples of Older subjects and younger
subjects using a series of memory tasks and questionnaires. They found that memory
performance was usually better in the younger than the older subjects, however,
perceptions about memory varied little as a firnction of age, and the subjective reports
were unrelated to objective memory performance. Larrabee and Levin (1986) also studied
memory self-ratings and objective test performance in a normal elderly sample. These
researchers found no association between memory complaint and objective measures of
memory function. They concluded that subjective memory questionnaires should not be
used alone in the diagnosis of age related disorders when the presenting complaint is
memory loss. Kahn, Zarit, Hilbert, and Niederehe (1975) attempted to clarify the status of
memory impairment in the elderly by assessing the relationship between subjective
memory complaints and actual cognitive functioning and found a marked discrepancy
between complaints and memory performance. They stated that complaints can occur

with or without an actual deficit in memory, and that people who complained about

memory sometimes performed better than those who did not. Further evidence of this
lack of a relationship between memory complaints and memory decline was given by
Williams, Little, Scates, and Blockman (1987). Using a variety of clinical memory tests,
they found that memory complaints were not significantly related to objective memory test
performance. Chandler and Gemdt (1988) examined this relationship as well, and found
no difference in memory testing scores between subjects with and without memory
complaints. Derousene, Alperovitch, Arvay, and Migeon (1988) studied the
interrelationship between severity of memory complaints, performance in memory tests,
and afl‘ective status in 367 French 50-80 year olds. No relationship was found between
severity of memory complaints and age, sex, educational level, marital status, living alone
or in family, or memory test performance.

Given the above research, one might feel comfortable in concluding that subjective
memory complaints are not indicative of impaired memory functioning. However,
O'Brien, Beats, Hill, and Howard (1992) conducted a three year follow-up of 64 people
(aged 50 years) who complained of memory difficulties. These researchers did find that
cognitive performance of the nondemented subjects showed a significant but relatively
modest decline, which they related to normal aging. They concluded that memory
complaints must be taken seriously in the elderly and that it may sometimes indicate early
dementia, however, in most cases a finding of normality remained accurate.

Also, one must be careful in interpreting studies that did not control for
pre-morbid firnctioning. Christensen (1991) expanded on this point by examining the
relationship between complaint and objective test performance in 20 elderly persons who

identified themselves as having memory problems. Upon analysis, ”memory performance

proved poor in those memory complainers who: (a) considered their memory impairment
to be both worse than their peers; and (b) had a discrepancy between their current level of
memory fimctioning and that expected on the basis of premorbid intelligence" (p. 310).
However, with this in mind, analysis of variance did fail to find a relationship between the
report of failure of memory and objective test performance. It is possible that the
objective memory measures used in the above studies were not successfirl in delineating
those individuals who performed at a much higher cognitive level in their earlier adult
years and who experienced memory decline that landed above the cutoff scores for
”memory impairment. " However, the literature fails to report a significant relationship
between memory complaint and poor memory performance. This consensus is not
surprising since memory complaints are especially common in the elderly and since the
incidence and prevalence of actual progressive memory disorders is relatively low.

At this point a second question can be posed: Ifcomplaints about memory do not
reflect a cognitive decline than what are the correlates of memory complaints and why do
they occur?

The Ubiquitous Role of I_)_emession in Subjective Memog Appraisal

Thereisamarkedincongruitybetweencomplaintaboutmemoryandactual

 

memory performance. Complaints can occur with or without an actual deficit in memory.
However, objective evidence of memory problems in an individual does typically
precipitate memory complaints (Thompson & Gallagher, 1990). Perhaps these difl‘erences
in performance between individuals are a result of other psychological processes,
including depression. Although we have seen that the relationship between increased

memory complaints and actual memory dysfirnction was less than well established (see

also, Lamberty & Bieliauskas, 1993 ), there was an observed relationship between
depression and memory complaints.

In the Larrabee and Levin (1986) study mentioned above, factor analysis indicated
that patients' memory self-ratings were primarily related to the afl‘ective state rather than
to objective memory performance. These researchers measured depression using the
lung Depression scale. In the Kahn et al. (1975) examination, it was found that while
performance varied with altered brain function, complaint was related to level of
depression, as measured by the Hamilton Rating Scale for Depression. Kahn et al.
concluded that complaints were found to be only a manifestation of depression, not of
cognitive performance. In the Williams et al. (1987) research, memory complaints had a
stronger association with depressed mood, as measured by the Beck Depression Inventory
(BDI), than with performance on memory tests. In the Derousene et al. (1988) study, a
strong correlation was found between severity of memory complaints and scores on a
self-reporting depression questionnaire, even among those with low depression scores.

Furthermore, Pettinati, Brown, and Mathisen (1985) showed that in depressed
geriatric inpatients, as measured by the Hamilton Depression Scale, memory complaints
were correlated with severity of depression and that it was this severity, rather than the
frequency of complaints, that may be an important dimension to consider. Additionally,
since a clinical lore has developed that emphasizes subjective memory complaints as a
diagnostic indicator for depression, it would make sense that subjective memory
complaints should decrease as depression lifts. Plotkin, Mintz, and Jarvik (1985) found
that improvements on the Hamilton Depression Scale were significantly related to a

decrease in subjective memory complaints, regardless of whether tricyclic antidepressant

or psychotherapeutic interventions were used. Furthermore, F eehan, Knight, and
Partridge (1991) assessed cognitive and test performance in elderly patients suffering
depression or dementia and concluded that depressed subjects, as measured by the
Hamilton Depression Scale, viewed themselves as more cognitively impaired than control
subjects and that demented subjects rated their level of cognitive functioning higher than
controls. These researchers stated that measurement of subjective complaint can provide
an indication of the degree of insight elderly patients have into the nature of their disorder.
Molinari (1991) reviewed mental health issues in the elderly and concluded that depressed
persons emphasize memory difficulties while those with probable Alzheimer's disease
attempt to minimize them.

Overall, researchers are in general agreement that memory complaints are
exacerbated by even low levels of depression. Furthermore, although eldedy persons'
assessments of their own abilities are of considerable importance, they have little validity
and should not be used to make even tentative diagnoses of dementia. However, from the
evidence above, it would be reasonable to assume that memory complaints are a
diagnostic sign for depression and that one may be able to judge the severity and type of
depression in the elderly population from the reliable and valid measurement of memory
complaints.
mpression in the Elderly

The general assumption in both the popular and scientific press is that aging is
associated with an increased risk of depression. For example, as Klerman ( 1983) stated,
"Mental illness, in general, appears to be more prevalent among the elderly than among

younger adults; but the incidence of depression is particularly high in persons 65 and

older- not only for depressive disorders, but also for transient symptoms of depression" (p.
3). Others do not firlly agree with this view. As Newmann (1989) stated, empirical
studies that have investigated the relationship between aging and depression do not Show
consistent support for this assumption. Epidemiological data about the incidence and
prevalence of mood disorders in the elderly do not yield a complete and consistent picture
(Blazer, 1983). Newmann's review attributed this inconsistency to diverse measurement
approaches and flaws in design and analysis which make it dificult to draw clear
conclusions regarding the relationship between depression and aging. Nonetheless,
”depression is among the most prevalent health problems of the elderly, occurring for the
first time in about 10 to 20 percent of the population 60 years or older" (p. 87) (Ruegg,
Swerdlow, & Zisook, 1988). The occurrence of depressive symptoms seems to be high
in the elderly ranging from 5 to 40 percent (Lamberty & Bieliauskas, 1993). Although
traditional views regard depression as having a good prognosis in the general population, a
prospective study of 124 elderly patients showed that only one-third actually had a good
outcome (Murphy, 1983). This study suggested that poor outcome is most clearly
associated with severity of depression, physical health problems, and strenuous life events
such as bereavement and separation. Therefore, as Ruegg et al. (1988) stated, it is critical
that clinicians appreciate the importance of depression in the elderly and be fully aware of
the diagnostic indicators and the standard course of pathogenesis. Elderly individuals
often have much to be sad about, as people around them become sick and die, social and
economic limitations often arise, health may be compromised, and commonly prescribed

medications are ofien associated with sadness and fatigue.

AS Lamberty and Bieliauskas (1993) pointed out, depression in the elderly is
difi'erent from the classical depressions which occur in younger individuals. According to
the Beck (1974) model of depression, the basic syndrome of depression is classically
described by a cluster of five symptoms: (a) A specific alteration in mood, sadness,
loneliness, or apathy; (b) a negative self-concept associated with self reproaches and
self-blame; (c) regressive and self-punitive wishes; desires to escape, hide, or die; (d)
vegetative changes; anorexia, insomnia, loss of libido; and (e) change in activity level;
retardation or agitation. Lamberty and Bieliauskas (1993) stated that these symptoms may
be inherent in the aging process and that ”they are easily confounded by the effects of age
and illness states cormnon to the elderly as well as of changes due to physiological aging"
(p. 151).
mg' and Masked Depression

Another variation of depression seen most commonly in the elderly is that of
masked depression. A Masked depression is a disorder with significant subjective and
firnctional disability marked by a cluster of vegetative symptoms but without prominent
dysphoria or guilt (Weiss, Nagel, & Aronson, 1986). As Ruegg et al. (1988) stated, the
elderly frequently do not complain of sadness or dysphoria but instead mask this afl‘ect by
"prominent somatic complaints such as gastrointestinal upsets, complaints of memory or
concentration disturbances, or decreased energy or drive" (p. 91).

Neskes and Jarvik (1987), in describing masked depression, stated that patients
may complain of insomnia, physical problems, pain, and constipation and deny being
depressed but yet respond to antidepressant treatment. Goldstein (1979) claimed that

these somatic complaints may be an attempt on the part of the depressed elderly patient to

combat feelings of helplessness, to avoid fear of failure, and to restore some measure of
control. The exact relationship between physical health and depression is complex.
Growing old predisposes oneself to a variety of debilitating illnesses. Physical illness can
precipitate depression, and in the face of illness, depression is more common (N eskes &
Jarvik, 1987). Therefore, it is important to take these complaints seriously. However,
these somatic complaints may be masking an underlying affective disorder and the elderly
may not properly attribute true indices of physical distress. Therefore, since the elderly
may be more likely to mask their depression than younger individuals (Salzman & Shader,
1978), it is suggested that such somatic complaints are an important variable to consider in
the delineation of depression and physical illness in the elderly.

On the other hand, as Rapp and Vrana (1989) pointed out, some somatic signs and
symptoms of depression- for example, appetite or weight change, hypo- or hypersomnia,
fatigue or loss of energy, and diminished ability to concentrate also can be symptoms of
various medical illnesses and/or drug side effects and are correlates of normal aging.
These researchers claimed that somatic symptoms currently used to diagnose major
depression in the elderly may be the result of normal aging and are highly unspecific in the
diagnosis of depression. Rapp and Vrana (1989) suggested that it would be more usefirl
to substitute nonsomatic for somatic symptoms in the diagnosis of depression in the
elderly. In a study of 150 elderly male medical inpatients, they examined the sensitivity
and specificity of a modified (substituting nonsomatic for somatic symptoms) version of
the Research Diagnostic Criteria (RDC) for major, minor, and intermittent depression.
These researchers provided empirical support to conclude that it may be easier and more

accurate to assess nonsomatic symptoms than to decide when somatic symptoms are

10

secondary to depression and when they are secondary to the host of other potential
causes.

Bourge, Blanchard, and Saulnici (1992) looked at the impact of somatic symptoms
in the evaluation of depression among a geriatric population. These researchers found a
significant correlation between physical health and the degree of depression, supporting
the view of Rapp and Vrana (1989). However, as Bourge et a1. (1992) stated, the higher
level of depression Observed in the elderly with poor physical health is caused by the
inflation of somatic items as well as the mood related items on the BDI, which taps both
somatic and mood related symptoms of depression. Furthermore, Bella and Bleecker
(1989), in examining the efi‘ects of age and sex on the BDI, the Minnesota Multiphasic
Personality Inventory (MMPI), and the Geriatric Depression Scale (GDS) found that
greater physical malfirnctioning was significantly associated with higher scores on all the
depression scales addressing somatic symptoms. Stewart, Blashfieid, Hale, and Moore
(1991), assessing the correlates of BDI scores in an ambulatory elderly population,
indicated depression is underdiagnosed in older patients and that multiple somatic
symptoms are the best indicators of depression in this population.

As Waxmen and Carner (1984) suggested, physicians have fiequently reported
that elderly patients present with a variety of somatic complaints that lack apparent
physical bases. After unsuccessful treatment, physicians have concluded that these
complaints are a result of the somaticization of an underlying afl‘ective disturbance. Thus,
the delineation of those complaints which indicated a medical disorder and those that
represented an afl‘ective disorder is extremely important . Unfortunately, it is generally not

known which somatic complaints are most indicative of depression. Therefore, as these

11

researchers stated, the relationship between chronic mental illness, depression, and somatic
complaints is interactive and complex. Their study of 227 community elderly concluded
that a wide variety of somatic complaints were associated with depression and that
depression scores were potent predictors of somatic complaints. Therefore, the role of
somatic complaints in the diagnosis of depression is important to consider. Rapp, Walsh,
Parisi, and Wallace (1988) used RDC criteria to assess the base rate of depression in a
random sample of 150 elderiy medical inpatients and found that prevalence rates for
depression indicated that this population is at high risk for major depression. Self-report
screening measures were especially sensitive to mq'or depression and that even among the
frail elderly, somatic features should be used as a criteria for diagnosing depression. They
found that the BDI appears better suited for measuring RDC-defined symptoms of
depression than the GDS, which entirely omits the somatic/vegetative domain.

The present study proposes to investigate the relationship between subjective
memory complaints and depression in greater detail. The literature suggested that
subjective memory complaints are a diagnostic indicator for depression. However, when
one takes a closer look at the measurement of depression in the above studies, each scale
had somatic complaints as a scoring criterion for the delineation of depression.

According to Weiss, Nagel, and Aronson (1986), the lung Depression Scale,
Hamilton Rating Scale for Depression, and BDI all use somatic complaints as a
characteristic in scoring depression. The self-reporting depression questionnaire used by
Derousene et al. (1987) also contained somatic complaints as a criterion. It should be
noted that no consensus has been reached as to whether subjective memory complaints are

equally valid as a diagnostic indicator for somatic/vegetative types of depression and for

12

affective/mood related types of depression or whether they are more diagnostic for one
than the other. To find out, this study will look at whether elderly individuals who have
more subjective memory complaints, as measured by the Memory Assessment Clinic
Self-Report Scale (MAC-S), will respond differently on a factor of a depression measure
which is loaded with somatic/neurovegetative items (BDI-somatic factor only) than on
measures that are solely mood related (GDS and BDI-affective factor only).

Several studies have suggested that the BDI measures two constructs, designated
somatic complaints and non-somatic, or affective complaints (Volk, Pace, & Parchman,
1993). One study which delineated this factor structure was a study by Cavanaugh, Clark,
and Gibbons (1983). These researchers stated that the BDI can be demarcated into a
dominant first factor (affective) and a second factor (somatic). Cavanaugh et al. also
determined which BDI standard form items discriminated depressive severity in Older
hospitalized medical patients. The first 14 items were shown to be grouped as affective
items and the last seven items as somatic criteria. Therefore, for the purpose of this study,
the somatic factor of depression will be defined as the last seven items of the BDI.
According to Yesavage, Brink, and Rose (1983), the GDS is a unifactoral scale,
measuring afl‘ective aspects of depression. Therefore, according to these researchers,
none of the GDS items specifically measure somatic concerns. Therefore, this study will
define the afl‘ective factor of depression as the first 14 items of the BDI and all 30 items of
the GDS.

Previous research (Lan'abee, West, & Cook, 1991) assessing the association of
memory complaints with computer-stimulated everyday memory performance, determined

there was not a high correlation between depression and the MAC-S factors

13

(correlation's between 06-26). They suggested that the MAC-S avoids confounding
depression with memory self evaluation. However, the relatively small correlations with
depression and subjective memory complaint were based only on relationships between
the MAC-S and GDS. When one addresses somatic indicators of depression it would
appear that a greater relationship should exist between subjective memory complaint and
scores of depression. The discrepancies between the above studies may be attributed to
how one measures depressive symptoms. Subjective memory complaints are predicted to
have a stronger link with somatic aspects of depression rather than mood-only related
characteristics of depression.

Furthermore, the relationship between somatic symptoms and memory complaints
may signal the presence of hidden or "masked” depression. "Masked,” or hidden
depression, by the definition used, requires that there be a high number of memory
complaints, a high number of somatic complaints, a low number of mood complaints, and
little or no objective evidence of physical impairment that could account for the somatic
complaints. If individuals score high on memory complaints and have comparable scores
on somatic and affective aspects of depression then this could be taken to suggest that
memory complaints may be a diagnostic indicator for "regular depression", and that the
elderly subjects are giving a fair subjective appraisal of their depressive symptomatology.
However, if individuals score high on memory complaints, low on affective complaints,
high on somatic complaints, and do have objective evidence of physical impairment of the
sort that could account for the somatic complaints, then it would be reasonable to assume
that these individuals can be classified as ”ill, but not depressed." It is proposed that a

complex interaction exists between subjective memory complaints and the nature of the

14

symptoms of depression. The present study is designed to document whether such an
interaction is at work within a normal elderly population and whether one may delineate
the presence of hidden or ”masked” depression with the criteria used. Since it has been
shown that individuals who present with such somatic complaints, and have no Objective
physical impairment, do in fact respond in a positive manner to antidepressant therapy
(Makanjuola & Olaifa, 1987; Neskes & Jarvik, 1987), than it would be quite helpfirl to
propose diagnostic criteria for those who are "masked” depressed.

Hypptheses

In light of the preceding discussion, the purpose of this study is to assess whether
subjective memory complaints are a diagnostic indicator of depression and to delineate the
role of these complaints as they relate to somatic and mood related aspects of depression.

It is hypothesized that:

1. A significant positive relationship will exist between subjective memory
complaint and mood-related depression. Operationally, this will be tested by assessing the
relationship between the MAC-S and the 30 items of the GDS plus the first 14 items of the
BDI. Together, these 44 items address an affective factor of depression (Cavanaugh et
al., 1983, Yesavage et al., 1983).

2. A significant positive relationship will exist between subjective memory
complaint and somatic aspects of depression. Operationally, this will be tested by studying
the relationship between the MAC-S and the last seven BDI items which address the
somatic factor of depression (Cavanaugh et al., 1983).

3. A significantly stronger relationship will exist between subjective memory

complaint and somatic aspects of depression than between subjective memory complaint

15

and mood-only aspects of depression. Operationally, this will be tested by studying the
difi‘erence in the relationship between the MAC-S and the somatic factor of depression as
compared to the MAC-S and the afl‘ective factor of depression.

On a descriptive leveL I expect to isolate three groups of able elderly. The first
two groups will consist of individuals who evidence subjective memory complaint as
measured by the MAC- S, but evince vegetative symptoms Of depression (somatic factor)
rather than mood related aspects (afi‘ective factor). Group A will comprise of individuals
who have a significant medical history and Group B will comprise of individuals whose
physical health is essentially noncontributory. Group A, who will be called ”ill, but not
depressed," will be those individuals whose memory complaint is a firnction of genuine
physical impairment. Group B will be those considered to have "masked”, or hidden
depression, since there is a lack of contributing physical impairment. Operationally,
physical impairment will be measured by the Memory Assessment Inventory (MAI) and
will take account of major exclusion criteria as suggested by Blackford and Larue in
delineating Age Associated Memory impairment (Crook et al., 1986). The criteria will
include: (a) Any neurologic disorder that could produce cognitive deterioration. Such
disorders include AD, Parkinson's Disease, stroke, intracranial hemorrhage, local brain
lesions including tumors, and normal pressure hydrocephalus. (b) History of any infective
or inflammatory brain disease including those of viral, firngal, or syphilitic etiologies. (c)
History of repeated minor head injury or single injury resulting in a period of
unconsciousness for one hour or more. (d) Any medical disorder that could produce

cognitive deterioration including renal, respiratory, cardiac, and hepatic disease; diabetes

16

mellitus; endocrine, metabolic, or hematologic disturbances; and malignancy not in
remission for more than two years.

Additionally, a third group; Group C, may be isolated as well. Group C, who will
be called "regular depressed", will consist of individuals who evince high subjective
memory complaints (MAC-S), and also endorse both mood-related aspects of depression

(afi'ective factor) and vegetative aspects of depression (somatic factor).

METHOD

Participants

The participants were chosen from an ongoing Michigan State University (MSU)
Psychological Clinic Aging Research Preject. They were a subset of community dwelling
elderly recruited through advertisement. Each individual was offered two assessments of
their mood and memory, as well as a seven session workshop targeted to teach relaxation
or cognitive strategies for the relief of depression and/or memory difficulties. This pool of
90 participants contained protocols of older individuals ranging from 55 to 91 years old
(M = 70.40; §I_) =2.97). Of the 90 participants, 67 were women and 23 were men. The
group had a mean education of 14.76 years (SD = 2.97).

Measures
1. Beck Depression Inventog (BDI)

This self-rating instrument addresses depression. It consists of 21 items with four
graded statements relating to the severity of the symptom. Recent literature (Karanci,
1988) indicates that the BDI taps affective, motivational, cognitive, and somatic
symptomatology of depression. Karanci stated that although the BDI total score has
shown good sensitivity and specificity in detecting depression in medical patients, several
researchers have raised an objection to the utilization of the total BDI score for medical

patients. Karanci noted that the somatic/vegetative symptomatology tapped by the BDI

17

18

may be by-products of the physical illness. Karanci‘s study of patterns of depression in
medical patients, found that the somatic/vegetative symptomatology Of depression can be
partially accounted for by greater age, illness severity, and attributions to
nonpsychological causes.

A cutting score between 14 and 15 points is recommended to identify the
presence of significantly depressed mood by Beck and Beamesderfer (1974). Results of
studies suggest that the BDI has respectable internal consistency and stability for use with
the elderly. For example, there was a positive relationship between conventional cut off
scores on the BDI and selected diagnostic classifications of the Research Diagnostic
Criteria (Spitzer, Endicott, & Robins, 197 8) for detection of major and minor depressive
disorders. Only 16.7% were misclassified by customary BDI cutoff scores (Gallagher,
Nies, & Thompson, 1983). These results suggested that the BDI can be used as a
screening instrument for identification of clinically depressed elders.

2. er'atric Depression Scale (GDS)

Yesavage, Brink, and Rose's GDS (1983) is composed of 30 true/false
self-referent statements and was designed specifically for rating depression in the elderly.
According to Bieliauskas (1993), "the GDS appears to be quite sensitive to symptoms of
distress in this population but there are some questions as to whether or not it may be
overdiagnostic of clinical depression” (p. 123). None of the GDS items specifically
measure somatic concerns. Yesavage et al. (1983) suggested that the GDS is a reliable
scale. Test-retest reliability was calculated for the GDS by having 20 subjects complete
the questionnaire twice, one week apart. A correlation of .85 was obtained (p< .01),

suggesting that, at least within the time frame considered , scores on the GDS reflected

l9

stable individual difl‘erences. Convergent validity was shown to exist between the GDS,
the lung Self-Rating Scale for Depression (SDS), and the Hamilton Rating Scale for
Depression (HRS-D). According to Yesavage et al. (1983), the obtained correlation
between the GDS and the SDS was found to be .84 while a correlation of .83 was found
between the GDS and the HRS-D. Furthermore, the GDS appeared to have respectable
internal consistency and stability for use with the elderiy. According to Yesavage et al.
(1983 ), a positive relationship between conventional cut-ofi‘ scores on the GDS and
selected diagnostic classification of the Research Diagnostic Criteria was elucidated.
T-tests conducted showed that subjects classified as normal scored significantly lower on
each of the scales compared to the mildly and severely depressed subjects while the
severely depressed group scored higher than each of the other two groups (all p< .01).
According to Yesavage et al. (1983) this scale is especially suitable for detecting
depression in the elderly because it avoids classification errors due to bad physical health.
3. Memegz Assessment Climb Self-Report ScaLle (MAC-S)

The MAC-S (Larrabee, West, & Crook, 1991) seems to be a promising new scale
tint assesses the subjective domain of memory complaints. This revised memory
questionnaire contains 21 ability items and 24 fiequency of occurrence of memory
problem items. Factor analysis of this scale demonstrated the usefirlness of MAC-S
factors based on a sample of 1106 participants. It has a large normative base that covers
the adult range of 18-92 years. Data has bear provided by the above researchas showing
the concurrent validity of this new self-report scale. Data suggested that significant

associations can be demonstrated between self-rated and objectively measured memory.

20

Overall, the percentage of shared variance between MAC-S and computer-simulated
objective memory tasks ranged from 27.0% to 29.4% (Larrabee et al., 1991).

Youngjohn, Larrabee, and Crook (1992) examined the test-retest reliabilities and
practice efl‘ect magnitudes comprising the MAC-S battery and five traditional
neuropsychological tests in 115 subjects and reported significant practice effects on
reevaluation. The test-retest reliabilities were equal or superior to the other traditional
neuropsychological measures but the traditional measures were superior in measures of
attention and concentration.

4. Multilevel Assessment Instrument (M)

Lawton, Moss, Fulcomer, and Kleban‘s MAI (1982) measures the well-being of the
aged in the areas of behavioral competence (health, cognition, time use, social interaction,
etc), psychological well-being, and perceived environmental quality. Reliability was
reported by Lawton, et a1, as follows: Summary rating scales in seven domains were
completed by using an interviewer and a ”reader-rater" for 484 of the 590 respondents and
by an interviewer and interview observer for the remaining 106. In the sample of 484,
interviews and reader-raters agreed with either a 0 or 1 point discrepancy in 95% of all
instances. Intraclass correlations ranged from .88 (activities of daily living) to a low of .58
(social interaction). Reliability of the final MAI scales were all .61 or above except for
health behavior (.39). Retest reliability done at a three week interval on 22 cases was
acceptable, the majority receiving a perfect score (p. 95). Lawton et al. also determined
the validity of this scale by doing summary ratings, multiple correlations, and by
constructing a "durmny variable" representing independent versus dependent groups.

Except for the social domain category, Lawton et al. concluded that the reliabilities and

21

validities of the MAI indices and subindices were affirmed by several different approaches.
Because analysis had been done separately for each domain, portions of the MAI may be
used if desired. Lawton et al. (1982) maintained that physical health, cognition, and
activities of daily living were all strong domains.
Procedure

Participants were assessed on both depression measures, the memory assessment
scale, and multilevel assessment instrument on two occasions at approximately a three
month interval. These tests were administered as part of a larger study designed to assess
mood and memory firnctioning. All testing was carried out by clinicians enrolled in MSU's
clinical psychology programs. Examiners had been previously trained in the administration
of the questionnaires. Participants were not paid for their participation. However, each
individual was told they would be receiving feedback on their mood and memory status, as
well as receiving the seven session workshops targeted at improving depression and
memory difficulties. The entire assessment required between one-and-one-half and two
hours. All tests were initially scored by the trained clinician administering the battery.

However, the tests were re-scored by the author to ensure accuracy.

RESULTS

Data analyses were subdivided into four components. First, the psychometric
properties of the two depressive indices and resulting somatic and affective factors were
assessed. Second, the three hypotheses were evaluated. Third, a profile analysis was
conducted to determine if any individuals, with the measures used, could be classified as
”masked depressed", "ill, but not depressed", or ”regular depressed”. Fourth, post-hoc
analyses of the overall sample were conducted.

Psychometric Properties

First, an inter-item reliability check, using Cronbach's (1951) alpha, was conducted
to determine if adequate reliability existed to justify combining the first 14 items of the
BDI (Cavanaugh, 1983) and 30 items of the GDS (Y esavage et al. 1983) to produce a
total affective factor of depression. The item total statistics and stability coefficients for
this factor (elphe = .91) and resulting 44 items can be found in Appendix A. In summary,
it appears that there was adequate reliability to support combining these items to produce
a unifactoral affective factor. The only items in question were four BDI items (#‘s 6, 10,
11, & 14). However, these items were included in the factor since deleting the items
would only minimally increase the reliability of the factor. Furthermore, the items in

qUeStion were retained to provide comparability for past and firture studies done in this

22

23

subject area. The individual raw data totals for the BDI, GDS, and MAC-S can be found
in Appendix C.

Second, an inter-item reliability check was conducted using Cronbach's (1951)
alpha, to justify the use of the last seven items on the BDI to form a somatic factor of
depression (Cavanaugh, 1983). The item-total statistics and stability coefficients for these
seven items can be found in Appendix B. Again, in summary, it appears that adequate
reliability (M: .44) was achieved to support previous research in delineating the last
seven items of the BDI as a separate somatic factor of depression. One item (BDI item
19) was somewhat doubtfirl for inclusion in the factor. Since deleting the item would only
minimally increase the reliability of the factor and retaining the item would assure
comparability for firture and past studies, this item was included in the factor. Again, the
individual raw score totals on the BDI, GDS, and MAC-S can be found in Appendix C.
Hypetheses
Hypothesis 1. It was predicted that a significant relationship would exist between
subjective memory complaints, as measured by the total MAC-S score, and the affective
factor of depression, as measured by the 30 items of the GDS and first 14 items of the
BDI. Hypothesis I was supported. A Pearson product-moment correlation revealed a
significant relationship between these measures, 1' (90) = -.53, p < .01. That is, people
who scored lower on the MAC-S, indicating a higher number of subjective memory

complaints, scored significantly higher on the afl‘ective factor of depression.

Hypothesis II. The second hypothesis predicted that a significant relationship would exist

between subjective memory complaint, as measured by total MAC-S score, and the

24

somatic factor of depression, as measured by the last seven items of the BDI. Hypothesis
II was supported. A Pearson product-moment correlation revealed a significant
relationship between these measures, 1 (90) = -.39, p < 01. That is, people who scored
lower on the MAC-S, indicating subjective difficulties with their memory, scored

significantly higher on the somatic factor of depression.

Hypothesis IH. The third hypothesis predicted that a significantly stronger relationship
would exist between subjective memory complaint and the somatic factor of depression
than between subjective memory complaint and the afl‘ective factor of depression.
Specifically, it was predicted that the total score on the MAC-S would have a significantly
stronger relationship with the last seven items on the BDI (somatic factor) than with the
30 items on the GDS combined with the first 14 items on the BDI (afi‘ective factor). This
hypothesis must be rejected. Pearson product-moment correlations did reveal significant
relationships in both instances, supporting hypotheses I and 11. However, results were in a
direction opposite to that of the hypothesis. The strength of correlation was geater
between the MAC-S and afi‘ective factor of depression (1 (90) = -.53, p < .01) than
between the MAC-S and somatic factor of depression (r (90) = -.39, p < 01). An r to 2
conversion revealed that the difi‘erence between these correlations was not significant (r; to
g = 1.64).
Profile Analysis to Delineate Depression Tm

A profile analysis was designed to reveal distinct depression groups as defined
through cutofl‘ scores on the affective factor of depression, somatic factor of depression,

and MAC-S. As suggested by Crook and Larrabee (1992), individuals were defined as

25

having considerable subjective memory complaints if they scored one standard deviation
below the mean on the Total Ability scale summed with the Frequency of Occurrence
(Tot-Foc-Ab) scale on the MAC-S. Therefore, considerable subjective problems with
memory was defined as a total score of 119 or below on these two scales of the MAC-S
(Crook & Larrabee, 1992). Out of a total sample of 90 participants, 38 were defined as
having considerable subjective memory difficulties, falling at or below this cutoff.

These 38 individuals who had considerable subjective dificulties with their
memory were then placed into depression types using cutofl‘ scores on the somatic and
afi‘ective factors of depression. To keep the cutoffs uniform across scales, participants
were defined as scoring high on the somatic aspects of depression if they scored one
standard deviation above the mean for total somatic factor (TotSorna) score (M_=3.77;
SQ=2.88). Similarly, participants were defined as scoring high on affective aspects of
depression if they scored one standard deviation above the mean for total affective factor
(TotAfi) score M=10.49; SE88?) Cut-ofl‘ scores delineated four distinct depression
groups. The goups were labeled as follows: (a) Masked depressed, (b) regular
depressed, (c) nondepressed memory complainers, and (d) afi‘ectively depressed. The
means and standard deviations of memory complaints score, somatic factor score, and
affective factor score for the 38 individuals, and resulting four depression goups, can be
found in Table 1.

In addition, there was a search for individuals whose physical symptoms might
account for their memory difficulties. Physical impairment symptomatology was measured
by the MAI and took into account major exclusion criteria as suggested by Blackford and

Larue (Crook et al., 1986). None of the 38 individuals in the subjective memory

26

Table 1: Descriptive Statistics for Memory Complaints Score, Somatic Factor Score, and
Total Affective Score for the Four Depression Groups in the Sample

 

 

Group N TOT-FOC-AB TOT-SOMA TOT-AFF
Mean SD Mean SD Mean SD

Masked Depressed 8 98.63 17.59 7.38 2.45 12.13 3.56

Regular Depressed 11 99.45 11.64 7.00 2.14 25.73 7.40

Non-Depressed

Memory Complainers 15 109.00 8.71 2.80 1.37 8.40 3.22

Afi‘ectivelyDepressed 4 111.25 9.64 2.50 1.29 22.75 2.50

 

27

complaint goup revealed such physical symptomatology. Therefore, no participants in

this study were found who met the "ill, but not depressed" classification.

Group A "Masked Depressed"
A total of eight individuals were labeled as "masked depressed" (8.8% of the total

sample) with the use of the diagnostic criteria proposed. As Table 1 shows, each
individual in this goup was found to evince considerable subjective difficulties with their
memory, a high number of somatic depression complaints, and a low number of mood or
affective complaints. Table 2 shows the descriptive statistics for age, education, and total
health complaints for this goup and the other three subgoups in the sample. Total health
complaints were measured by the MAI. Each health complaint was equally weighted and
total health complaints were tabulated. Additionally, Table 3 shows the frequency
percentage of men and women found in the four depression goups as well as the
frequency percentage of those living with someone or living alone. The "masked
depressed" goup contained more women (five) than men (three) and five of the eight
individuals lived with someone else.

Group B, "Regelar Depressed"
A total of 11 individuals (12.2% of the total samme) were labeled as "regular

 

depressed. " As Table 1 shows, each individual was found to score high on subjective
memory complaints and endorsed increased levels of somatic as well as affective aspects
of depression. Table 2 shows that this was the oldest subgoup of the four subgoups in
the sarnpie. Also, members of this goup endorsed more health complaints when

compared to the other tlnree depressed goups. Table 3 describes that more women

28

Table 2: Descriptive Statistics for Age, Education, and Health Complaints for the Four
Depression Groups in the Sample

 

 

Group N Age Educ Health
Mean SD Mean SD Mean SD
Masked Depressed 8 75.38 11.62 14.50 2.14 3.25 2.60
Regular Depressed 11 76.64 10.67 13.36 4.18 3.36 1.57
Non-Depressed
Memory Complainers 15 66.60 7.77 14.67 2.92 2.00 1.51
Afi‘ectively Depressed 4 69.75 13.60 16.00 4.69 2.75 1.26 !

 

 

Table 3: Frequency Percentage of Gender and Living Arrangements for the Four
Depression Groups in the Sample

 

 

Group N Gender Living Arrangements
Male Female Alone W/ Someone

Masked Depressed 8 37.5% 62.5% 37.5% 62.5%

Regular Depressed 1 1 36.4% 63.6% 72. .7% 27.3%

Non-Depressed

Memory Complainers 15 20.0% 80.0% 26.7% 73.3%

Afl‘ectively Depressed 4 25.0% 75.0% 0.00% 100.0%

 

29

(seven) than men (four) made up the subgoup and more of the individuals lived alone
(eight) than with someone else (three).
Additional Depressed Groups

Two additional subgoups of individuals who endorsed high subjective memory
complaints were isolated.

The first additional subgoup, goup C, accounted for 15 of the 38 individuals
(16.7% of the total sample) and was labeled "nondepressed memory complainers. " As
Table 1 shows, participants in this subgoup scored high on subjective memory complaint
but did not endorse high levels of either afi‘ective or somatic aspects of depression. As
Table 2 shows, this subgoup was the youngest of the four goups and endorsed the
fewest health complaints. Table 3 shows that more women (12) than men (tlnree) made up
the subgoup and the majority of the 15 individuals lived with someone else (11).

The final subgoup of individuals delineated, goup D, was labeled "affectively
depressed." As Table 1 shows, these participants scored high on subjective memory
complaints, high on the affective factor of depression, and low on somatic aspects of
depression. Four individuals comprise this goup (3.8% of the total sample). Table 2
shows that this goup was the most educated of the four depression goups. Table 3
shows that three of the four individuals in this goup were women and all four individuals
lived with someone else.

Moe Anehges of the Whole Sa_rr_rpl_e

First, an analysis of the whole sample (p = 90) was conducted to delineate the

relationship between chronological age and memory complaints, as measured by the

MAC-S. A Pearson product-moment correlation matrix revealed a sigrificant relationship

30

between these two variables, I (90) = -.23, p < .03. That is, as the able elderly increase in
age they endorse more subjective memory complaints. Next, analyses for the whole
sample were conducted to determine the relationships between chronological age and
somatic aspects of depression (somatic factor) and chronological age and affective aspects
of depression (affective factor). A Pearson product-moment correlation revealed a
significant relationship between chronological age and the somatic factor of depression, 1
(90) = .28, p < .01. That is, as the able elderly increase in age they endorse higher levels r
of somatic aspects of depression. However, a post-hoc analysis revealed that when age

was controlled for, a significant relationship still existed between subjective memory

 

‘1—

complaints and somatic aspects of depression, ; (90) = -.33, p < .01. Additionally, no
significant relationship was found between chronological age and the afi‘ective factor of
depression. Therefore, according to the findings, there is a true significant relationship
between subjective memory complaint and both affective and somatic aspects of
depression

Third, analyses were conducted looking at the relationships between gender and
subjective memory complaints. A point by serial correlation revealed a statistically
nonsignificant relationship between gender and memory complaints (r_= -.1). Tlnat is,
males were slightly more likely to endorse subjective problems with their memory.

Finally, analyses were conducted to determine the degee of relationship between
gender and afi‘ective aspMs of depression and gender and somatic aspects of depression.
Again, a point by serial correlation revealed a very slight difference in favor of males
endorsing somatic aspects of depression (1 = .02). Additionally, almost no relationship

was found between gender and afi'ective aspects of depression. A very small difference in

31

favor of females endorsing affective aspects of depression was found (; = -.01). None of

these differences were statistically significant.

 

DISCUSSION

The analyses for this study were subdivided into four components. First, the
psychometric properties of the two depressive indices and resulting somatic and affective
factors were assessed. Second, the tlnree hypotheses were evaluated. Third, a profile
analysis was conducted to delineate four depression goups in our sample: (a) masked
depressed, (b) regular depressed, (c) just memory complainers, and (d) afl‘ectively
depressed. Fourth, post-hoe analyses were conducted. This same format will be used to
discuss the findings of these analyses.

Psychometric Properties

Results indicated that there was reasonable evidence of inter-item reliability to
support combining the 44 items of the GDS and first 14 items of the BDI to form an
afl‘ective factor of depression. Additionally, adequate reliability existed to support the use
of the last seven items of the BDI as a somatic factor of depression. These results support
previous research (Cavanaugln, 1983) which demarcated the BDI into a dominant first
factor (affective) and second factor (somatic). In addition, the results supported
Yesavage's (1983) research which found the GDS to be a unifactoral afi‘ective scale.

One possibility for the low internal consistency coefficient on the somatic factor of
depression as compared to the afl'ective factor of depression is the inclusion of only seven

items in the factor. The affective factor, which had a higher reliability coeficient,

32

 

 

33

contained 44 items. However, the most likely result of reduced reliability in the somatic
factor may be the measurement of the construct itself. It is apparent that the somatic
factor is defined less clearly. The high fiequency of physical illness among the elderly
makes the differentiation between true illness and somatic symptoms both diffith and
challenging. The alpha coefficient certainly gives credibility to the association between
depression and somatic complaints, but it also raises a question about the extent to which
somaticizers' complaints may result fiom physical illness and/or physical illness interacting I
with depression. Somatic signs and syrnptorrns of depression; including insomnia, weight i

change, appetite change, and reduced concentration, also can be symptoms of certain

 

medical ailments. It is also true that the elderly are more fi'equently medicated (Shultz, y,
1991) and these somatic synnptoms may partially be attributed to medication side-effects.
Therefore, these symptoms are correlates of normal aging and measurement of these
symptoms in terms of depression may be very difficult. In this regard the measurement of
somatic symptoms becomes less specific than the measurement of affective symptoms of
depression.
Hypetheses

In this study we attempted to investigate the relationship between subjective
memory complaint and depression in geater detail. Previous literature suggested that a
link between subjective memory complaint and depression existed. However, some
studies found a much stronger relationship between the two variables than others. In
reviewing the literature, it became apparernt that those studies which included somatic
complaints as a scoring criterion for the delineation of depression (Larrabee & Levin

( 1986), Kahn et al. (1975), Williams et al. (1987), Derousene (1988), Pettinati et al.

34

(1985), Feehan et al. (1991), and Molinari (1991)) found a geater link between subjective
memory complaint and depression. On the other hand, a study (Larrabee et al., 1991)
assessing the relationship between subjective memory complaint and depression with the
use of an afl‘ective measure of depression found low correlations (correlation between .06
and .26) between the two variables.

Therefore, in this study, we attempted to measure depression in a novel way,
separating the construct into an afi‘ective factor and somatic factor. It was predicted that P
separate analyses of these factors would elucidate the relationship between depression and

subjective memory complaint.

 

The first analysis revealed a significant correlation between the afl‘ective factor of
depression and total subjective memory complaint score. This finding contradicts previous
research (Larrabee et al., 1991) which determined there was not a high correlation
between afi‘ective aspects of depression (GDS) and the MAC-S factors. According to
Larrabee et al. (1991) several factors may contribute to the variability in observed
associations between memory self-report and depression. Such factors include differences
related to the subject goup, or patients being investigated, and factors related to the
design of the questionnaires used. These two factors may help explain the differences in
the correlation currerntly found. The Larrabee et al. investigation looked at this
relationship in younger adults ( Age; M_= 48.83, SI); 14.66). The present study
investigated the relationship in older adults (Age; M; 70.40, SD = 2.97). Furthermore,
although the MAC-S was used in both studies, the present study used the additional 14

affective items fi'om the BDI plus the 30 items fi'om the GDS to form an affective factor of

35

depression. These additional 14 items may have accounted in part for the higher
correlations found in the present study.

The second analysis revealed a significant correlation between the somatic factor
of depression and total subjective memory complaints score. This result parallels previous
research findings which found relationships between depression, measured with somatic
complaints as a scoring criteriorn, and subjective memory complairnt. This finding provides
further support to the notion that self-rating of memory disturbance by older adults is k
related to depressed mood.

The present study suggests that subjective memory complaints may be a diagnostic

 

indicator for depression. Furthermore, separating depression into two separate factors l-ir
does not change the clinical picture. Both affective and somatic aspects of depression
were found to be significantly related to subjective memory complaint. In fact, the first
surprising finding was that affective aspects of depression correlated more strongly with
subjective memory complaint than did somatic aspects of depression. As discussed
above, one possible reason for this finding may be the measurement of the somatic factor
of depression. Not only did the factor consist of only seven items, but the construct is less
defined and may be confounded by correlates of normal aging. Ifthe somatic construct
were measured with a geater number of more specific items then somatic aspects of
depression may have correlated more strongly with subjective memory complaint.

However, the possibility does exist that the findings are accurate of the relationship
between subjective memory complaint and somatic and affective aspects of depression. It
may be that the link between the variables does not depend on how one measures

depression. Subjective memory complaint may be equally related to affective and somatic

36

aspects of depression. It could be that illness, including both depressive symptomatology
and physical distress, is best measured along a continuum. Thus, it may be diflicult, if not
impossible, to separate physical illness, affective aspects of depression, and somatic
aspects of depression into separate categories.
Profile is to Delineate ressi n T es

This study was undertaken, in part, to take a closer look at the measurement of
depression in the elderly. By distinguishing between the afi'ective and somatic factors of E
depression, it was hypothesized that the presence of masked depression might be

identified. By definition, masked depression is a condition in which the affective aspects

 

of depression are concealed by somatic symptoms. Therefore, such individuals who are y
masked depressed evince increased somatic concerns such as memory complaints, endorse
an increased amount of somatic or vegetative depressive symptomatology, and endorse
fewer afi‘ective items of depression. Thus, with the measures used in this study (GDS,
BDI, and MAC-S) we hypothesized that a profile of masked depressed individuals, by
definition, could be delineated. Furthermore, additional distinct profiles of depressive
symptomatology could be defined. In this study, four distinct depression subgoups were
iderntified. Analyses of age, education, health, gender, and social support provided
additional dernogaplnic information.

The masked depressed goup represented 8.8% of our total sample (p = 90). In a
nrral primary care practice using RDC defined diagnostic criteria, Barret et al. (1988)
found the estimated prevalence of masked depression to be 7.9% in individuals aged 65
and over. Barret et al. (1988) applied a two-step case identification process, a screening

inventory, and a structured clinical interview to estimate the prevalence of masked

37

depression. He defined masked depressed individuals as those who evidenced depressive
symptomatology (somatic signs) but denied experiencing depressed mood. Barret et al.
(1988) and Waxman et al. (1985) also stated that masked depression is more likely to
occur in Older women than older men. NO other demogaphic data was available on
masked depression from these researchers.

This data is quantitatively similar to our findings and may suggest that our cutoff
criteria was reasonably successfirl in identifying masked depressed individuals. However,
without conducting a treatment study for this subgoup, we cannot be certain these
individuals are actually presenting as masked depressed. Future research needs to assess
how these individuals respond to psychotherapy and/or antidepressant medication before
determining if the masked depression profile used in this study is valid.

As F isch (1987) stated, masked depression is a major problem within the health
care system. Most often, the disorder proceeds without being diagnosed. In fact, as
Katon (1982) stated, at least halfof somaticized depressions go unrecognized or
misdiagnosed and not properly treated by primary care physicians. In addition, Lesse
(1983) found dust the duration of illness prior to diagnosis of masked depression is more
than one year in nearly 80% of patients, and more than five years in 30% of patients.

The potential reasons for this diagnostic neglect are many. First, the disorder has
received little attention in the popular and scientific press and has not been well taught in
the medical fields. Second, the primary care physician and patient may both choose to
focus on the somatic symptoms of illness rather than afl‘ective symptoms. As Fisch (1987)
stated, many physicians may be "somaticizers" who would rather focus on physical

ailments than psychological ones. Third, the syndrome remains elusive. Not only are

 

 

38

primary care workers unfamiliar with the syndrome but there are no specific diagnostic
criteria or screerning measures available to tap the construct. Such a screening measure
could prove usefirl in medical, psychiatric, and psychological settings. Research shows
that masked depression is quite prevalernt in the geriatric population (Barret et al., 1988)
and may fester for quite some time before being diagnosed (Lesse, 1983). Furthermore,
untreated depression has been shown to increase mortality and morbidity in the geriatric
population (Waxrnan et al., 1985).

This study attempted to delineate a goup of masked depressed individuals with
the use of self-report neuropsychological questionnaires. Although this study did not
confirm the presence of masked depression in the goup, it would appear that eight of the
90 individuals in the sample, by definition, presented with masked depression. If this were
the case, and treatment studies proved efl‘ective, then a screerning measure of similar forrn
could begin to be refined. Such an instrument could prove invaluable in sorting through
the dificult and challenging constructs of masked depression and genuine physical illness.
Such a screening measure could lead to quick and effective testing, treatment, and proper
care. Treating depression early would reduce cost of care, and more importantly, the
morbidity and mortality of prolonged depression. Furthermore, the early recognition of a
true physical disorder might also reduce cost and curb the dangers associated with an
untreated somatic ailment. It is noted that in the use of the above language there is no
attempt to denigate psychological sypmptomatology. It is neither suggested nor implied
that being depressed is not a "true" or "genuine" concern. The issue is whether there are
objectively determined physical causes of the distress, thus suggesting somatic concerns

secondary to physical ailment rather than somatic concerns secondary to depression.

 

39

The Relationship Between Age and Smptometology Within the Groups

Although the primary purpose of this portion of the study was to delineate a
masked depressed profile, other interesting profiles did emerge. Additionally, within these
profiles, descriptive statistics for age and health complaints revealed an interesting trend in
the data. The largest goup of individuals found were those who endorsed high numbers
of memory complaint but failed to endorse high levels of either somatic or affective
aspects of depression. This goup ("non-depressed memory complainers") was found to
be the youngest and healthiest relative to the other depression subgoups. In general, the
possibility exists that concerns about memory precede health problems and depressive
symptomatology seen more commornly at a later age. These memory complaints may be
the first reflection of the patients' awareness of real decline or they could be stereotyped
expectations of intellectual deterioration in the elderly (Plotkin et al., 198 5). Perceived
memory complaints may potentially start the spiraling process of increased health
complaints and depression.

The second to the youngest defined goup were those who endorsed memory
complaints and affective aspects of depression. These "afl‘ectively depressed" individuals
endorsed more health complaints than the "non-depressed memory complainers" but fewer
health complaints than the "regular depressed" and the ”masked depressed. " Although
only four members comprise this afi‘ectively depressed goup, the data indicate that
affective aspects of depression may follow memory complaints. Additionally, within the
”afi‘ectively depressed" goup one sees an increased amount of health complaint.

The data indicate that the next stage in the process may be endorsing affective

and/or somatic aspects of depression, as well as increased memory complaint. In the

 

40

"regular depressed" and "masked depressed" goups we also find the most health
complaints relative to the other goups.

It may be that aging progesses in a step-wise faslnion. As Scogin, Storandt, and
Lott (1987) stated, people seeking treatment for memory problems may have special
attitudes and expectations about their cognitive ability. It is possible that these individuals
may have preconceived notions about the aging process and are more sensitive to age
related changes within themselves. When changes in memory do occur, this may P
eventually lead to feelings of sadness and hopelessness, which ultimately progess to

somatic concerns and then the overall package of depression and health complaints.

 

Post-hoe analyses revealed that a significant relationship did exist between chronological F1"
age and memory complaints as well as chronological age and somatic aspects of
depression. This provides evidence to suggest that as the elderly increase in age they
worry more about their cognitive functioning and experience a geater number of somatic
complaints.

On the contrary, low memory complainers may be more accepting of age related
changes in their own abilities and beliefs, leading to a healthier mental and somatic
outlook. These hypotheses should be tested to give us a better perspective on irndividual
adaptation to the aging process. As Scogin et al. (1985) stated, research that looks at
individual adaptation to aging may guide the development of appropriate treatrnernt
procedures. From the present study, it appears a trend between age and symptomatology
does exist. However larger goup sizes are needed and a more specific evaluation of the

process is warranted.

41

Effects of Sociodemogaphic Variables on the Depression Groups

For every depression subgoup examined we obtained higher rates for women
when compared to men. This difference is consistent with virtually every epidemiologic
study in the literature (Barret et al., 1988). Furthermore, this finding is not surprising
since the majority of our sample was composed of women. An interesting note is that the
goup with the highest percentage of men was the masked depressed goup. Barret et a1.
( 1988) found a much higher percentage of women than men in their masked depressed
goup. However, only eight members comprise our masked depressed goup, which may
explain the difference in findings. Furthermore, we still found nearly a two to one ratio of
women to men in our masked depressed goup.

In terms of social support, the only goup consisting of more individuals living
alone than living with someone else was the "regular depressed" goup. This may help
explain the severity of symptomatology seen within the goup, since individuals who live
alone are more likely to lack a close confiding relationship and live in an environment that
lacks needed stimulation (Henderson, Scott, & Kay, 1986). The 11 individuals in the
”regular depressed" goup endorsed considerable amounts of memory difficulties, affective
aspects of depression, and somatic symptoms of depression. Additionally, this goup was
found to have more health complaints than any other subgoup. These results are
consistent with previous research that found more physical and psychological
symptomatology in elderly that live alone (Henderson et al. 1986, Weissman & Klerman,
1977). Furthermore, the "non-depressed memory complainers" and "afi‘ectively
depressed" subgoups had the geatest frequency of people living with them and also the

fewest health complaints, providing firrther support for the previous findings. From our

 

 

42

study, it does appear social support acts a potential buffer against worsened health and
increased depression witlnin the elderly.

Previous literature on the relationship between overall years of education and
degee of depression showed that fewer years of formal education was independently
related to a geater degee of depressive symptomatology and geater cognitive decline

(Fuhrer, et al. 1992). Carpirniello, Carta, and Rudas (1989) also found that lower levels of

 

education in both men and women were related to increased levels of depression. In our "t.
study, the "regular depressed" goup had the fewest years of formal education, providing :
support for these findings. This relationship may be true since individuals who are less :
educated tend to have less prestigious and less satisfying jobs and may have more L .—

dificulty with finances. This may ultimately lead to individual dissatisfaction which is
highly correlated with depression (Carpirniello et al., 1989).
S_ugge_stion for Future Res_egeh

While this study is a preliminary investigation, it has provided us with many
suggestions for future research.

First, if possible, it is recommended that a better measure of somatic aspects of
depression in the elderly be constructed. As mentioned previously, this may be difficult
since many somatic symptoms of depression are correlates of normal aging. The issue of
how well somatic items measure depression in the elderly and how well they discriminate
the depressed form nondepressed is an issue which requires firrther research.

F urtlnermore, one potential reason for this study failing to add specific information to the
relationship between subjective memory complaint and depression was the use of a

seven-item scale that had relatively low inter-item reliability. If firture research could

 

43

devise a somatic scale with more items that were more specific, than the association
between subjective memory complaint and type of depression could be better elucidated.

This study was able to add credence to the belief that subjective memory
complaints are a diagnostic indicator for depression. However, much more research needs
to be done in the area. This is true since subjective memory complaint may also be an
early symptom of a degenerative disorder such as Alzheimer's disease. Information
relating to the accuracy of these complaints and how they specifically involve different
aspects of depression could aid health care workers in formulating more accurate
diagnoses. Future studies should carefillly assess this relationship to help clear up the
diagnostic picture. Iftlnis was accomplished people's self-reports of memory could
actually help serve in describing the nature of the disorder and help with firture treatment
planrning.

Another interesting question evolving from this study involves the independent
appraisal of one’s memory dificulties and how memory complaints appear to precede
depressive and health related symptomatology. Elderly who experience subjective
difficulties with their memory may have special attitudes and expectations about their
cognitive abilities. This may help raise into consciousness the fears associated with "losing
one's mind. " These individuals may be more concerned about the aging process and may
be more sensitive to age-related changes. This may ultimately lead to worry and a
decrease in self-esteem which may lead to depression and other health related
symptomatology. However, people who feel fine with their memory may interpret
age-related changes differently, resulting in a more positive outlook and more healthy

aging experience. Individual adaptation to the aging process is an area that could lead to

 

44

more effective treatment strategies. Furthermore, therapy in a cognitive-behavioral
fi'amework could help facilitate changes in attitudes, thoughts, and expectations about the
aging process. This study was only able to touch on some of these issues. Future studies
would need much larger goup sizes and a design specific to the topic.

The most logical extension of this study would be focusing on the issue of masked

depression and devising a masked depressed neuropsychological questionnaire. The first

 

step might be conducting a treatment study on the eight individuals who fit the profile of '.
”masked depressed. " Ifthese individuals showed improvement secondary to

antidepressant and/or psychotherapy treatment then a questionnaire of similar form could

be devised. One possible change in the questionnaire would be including more somatic in:

items that were more specific to the somatic aspects of depression. Next, similar cutoff
points on the measures could be defined. However, much more validity and reliability
research would need to be done on the scale. The clinical utility of such a scale would be
geat. Masked depression is a prevalent syndrome that often goes undiagnosed. The
development of a masked depressed neuropsychological questionnaire could be the first

step of many in gaining a better understanding of this elusive syndrome.

APPENDIX A

 

APPENDIX A

Table A-1: Item-total Statistics for Affective Factor of Depression

Reliability Coefficients: 44 items

Alpha=.9051
Mean=10.49
Variancfi 8.72
Std Dev=8.87

Affective Factor
Item

B]
B2
B3
B4
B5
B6
B7
B8
B9
B10
B11
B12
813
314
61
G2
G3

G5

G7

Scale Mean if
Item Deleted

10.3
10.26
10.37

10.1
10.32
10.31
10.22
10.17
10.46
10.19

9.96
10.32

9.98
10.09

10.4
10.27
10.39
10.27
10.33
10.23
10.41

Scale Variance Corrected Item

Alpha if

if Item Deleted Total Correlation Item Deleted

45

75.2

74.1
75.38
73.17
74.96
75.07
74.58
74.23
77.64
73.75
73.82
75.32
70.54
76.76
76.65
74.18
76.62
75.84
76.16

74.9
77.23

0.46
0.51
0.47

0.6
0.48
0.29
0.46
0.48
0.33
0.32
0.31
0.43
0.62
0.11

0.4
0.61
0.38
0.37
0.38
0.48

0.3

0.9025
0.9016
0.9024
0.9003
0.9022
0.9055
0.9024
0.9021
0.9044
0.9062
0.9068
0.9028
0.8998
0.9097
0.9035
0.9008
0.9036
0.9034
0.9034
0.9022
0.9043

“7"?

 

G8

G10
G11
G12
G13
G14
G15
G16
G17
G18
G19
G20
G21
G22
G23
G24
G25
G26
G27
G28
G29
G30

10.4
10.36
10.37
10.23
10.23

10.2
10.06

10.4
10.31
10.39
10.39
10.18

10
10.07
10.42
10.43
10.18
10.38
10.01
10.36
10.37
10.13

9.83

46

77.61
74.73
76.77
74.83
76.88
74.16
75.27
76.42
74.15
76.04
75.95
75.56
74.18
74.36
77.06
77.69
74.55
76.53
74.33
75.15
76.77

74.7
75.74

0.2
0.66
0.32
0.49
0.22
0.55
0.37
0.44
0.67
0.49
0.51
0.36

0.5
0.48
0.36
0.24
0.49
0.38
0.48
0.58

0.321
0.46
0.33

0.905
0.9009

0.904
0.9021
0.9053
0.9012
0.9035
0.9031
0.9003
0.9026
0.9025
0.9036
0.9018

0.902
0.9039
0.9047
0.9019
0.9035
0.9021
0.9016

0.904
0.9024

0.904

 

APPENDIX B

Table B-1: Item-total Statistics for Somatic Factor of Depression

Reliability Coefficients: 7 items

Alpha=.4351

Mean= 3.77

Variance=8.32

Std Dev=2.89

Somatic Factor Scale Mean if Scale Variance Corrected Item Alpha if Item
Item Item Deleted if Item Deleted Total Correlation Deleted
B15 3.09 6.94 .33 .359
816 2.93 5.03 .24 .391
317 2.89 7.13 .36 .364
B18 3.69 7.71 .30 .402
B19 3.32 5.71 .15 .453
BZO 3.56 7.61 .22 .407
B21 3.12 6.71 .17 .415

47

 

APPENDIX C

 

APPENDIX C

Table C-l: Total Scores on the BDI, GDS, and MAC-S (TOT-FOC-AB) for Each

Individual in the Sample
Subject Number Total BDI Score
1 2
2 2
3 4
4 5
5 15
6 18
7 3
8 1
9 13
10 10
11 4
12 7
13 12
14 10
15 1
16 4
17 10
18 27
19 6
2O 9
21 2
22 7
23 4
24 9
25 1
26 2
27 0
28 7
29 6
30 9
31 3
32 13
33 10
34 1

48

Total GDS Score
1
2
4
3
17
13
2
2
8
17
1
18
17
14
1
1
16
16

t—I
cc

oasmnogqmorO—n—o‘wt—os

TOT-FOC-AB
162
138
142
141

93
105
146
134

88
114
159
128
116
116
142
140

97
100
118
105
136
117
119
138
132
145
174
100
118
187
133
121
146
141

 

 

35
36
37
38
39
40
41
42
43

45
46
47
48
49
50
51
52
53
54
55
56
57
58
59

61
62
63

65

67
68
69
70
71
72
73
74
75
76
77
78
79

I—d h—lI—i h—d I—i

OZSooecmmg’mbqwquamq—rosqaqawmufi-hwm;

49

Scouawccbficoo

:&

NZ'EAoohmqoS-h-ooQAowwamwBOOSNBNN—‘GHNQOO

146
115
121
133
102
128

63
132
100

95
143

98
116
121
162
167

73
142
100
123

113

93
109

97
114
113
124
154
129
116
141
120
129
100
121
126
142

97
126

93
109
135
107
127

 

‘1'

 

80
81
82
83
84
85
86
87
88
89

N

:A

Azmmwoo

50

m—cchloomccw

H
Nw

139
120
126
127
126
120
118
142
117
122
151

 

LIST OF REFERENCES

LIST OF REFERENCES

Barret, J.E., Oxrnan, J.A., Gerber, T.E. & Paul, D. (1988). The prevalence of
psychiatric disorders in a primary care practice, Journel of Psychjetric
Research. 45. 1100-1106.

Beck, A & Beamesderfer, A (1974). Assessment of depression: The depression
inverntory. Modl_fynp' ' g Problems in Pharmacopgchiatg, 7, 151-169.

Beck, A & Beck, R (1974). Screening depressed patients in family practice: A
rapid technique. Pestgflate Mfl'cine, 52, 81-85.

Beck A, Ward T., Mendelson, J., & Mock, A (1961). An inventory for measuring
depression. Archives pf (fieral Research, 40, 569-573.

Becker, 1., Hufl‘, J., Nebes, R, Holland, A & Boiler, F. (1988).
Neuropsychological function in Alzheimer’s disease. Archivesef
Neurology, 4;, 263-268.

Bieliauskas, L. A (1993). Cognitive and psychopathologicai changes in

degenerative disease in the elderly. leurnal pf Cliniefl QdE Wm
Neuropgchelpgy, 15, 421-438.

Blazer, D. (1983). The epidemiology of depression in late life. In Breslau LO,
Haig MR (Eds): ression and ' : C Car an Con uences.
New York, Springer Publishing Company, 30-50.

Bolla, 1., Hufi; J., Nebes, R, Holland, A., & Boiler, F. (1988). Neuropsychological
fimction in Alzheimer's disease. flehives pf Neurology, 45, 263-268.

Bolla, W., & Bleecker, ML. (1989). Absence of depression in elderly adults.
Journel of Gerontology, 44, 53-55.

Bourge, P., Blanchard, L. & Saulnici, J. (1992). The impact of somatic symptoms
in the evaluation of depression among a geriatric population. Canadian
Journal efBehaviem Science, 24, 118-128.

Bower, G. (1981). Mood and memory. Ameriegp Psyehelpgig, 36, 129-148.

51

52

Brown, F .H., Dodrill, C.B., Clark, T. & Zych, K. (1991). An investigation of the
relationship between self report of memory functioning and memory test
performance. Journel of Clini_cal Psychology, 47, 772-777.

Carpiniello, B., Carta, M.G., & Rudas, N. (1989). Depression among elderly
people. Acta Psychiatnig Scandinavia, 80, 445-450.

Cavanaugh, S., Clark, D. & Gibbons, R (1983). Diagnosing depression in the
hospitalized medically ill. Psychosomatics, 24, 809-815.

Chandler, J.D. & Gemdt, J. (1988). Memory complaints and memory deficits in
young and old psychiatric inpatients. Journal of Geriatrics and Neurology,
_1_, 84—88.

Christensen, H. (1991). The validity of memory complaints by elderly persons.
MW of Gerietric chhier, 6, 307-312.

Clark, D.C., Cavanaugh, SA & Gibbons, RD. (1983). The symptoms of
depression ill medical and psychiatric patients. lpurnal of Nmous and
Mm Disease, 171, 705-713.

Cohen, J. (1992). A power primer. Psychological Bulletm,’ 112, 155-159.

Craik, F.I. & Lockharte, RS. (1972). Levels of processing: A framework for
memory research. 19ml of Verbal Learning and Verbal Behavior, 11,
671-684.

Cronbach, L. (1951). Coefl'lcient alpha and the irntemal structure of tests.
WE. 297-334.

Crook, T., Bartus, RT., Ferriss, S.H., Winitehouse, R, Cohen, G.D., & Gershorn, S.
(1986). Age Associated Memory Impairment: Proposed diagnostic criteria
and measures of clinical change: Report of a NIMH work goup.
Wt; 261-276.

Crook, T.H. & Larrabee, GJ. (1990). A self-rating scale for evaluating memory in
everyday life. Psycholegy and Agm' g, 5, 48-57.

Crook, T. & Larrabee, J. (1992). Norrnnative data on a self-rating scale for
evaluating memory in everyday life. Archiyes Lf Clinical Neuropsychelogy,
1, 41-51.

Derousene, C. Alperovitcln, A, Arvay, N. & Migeon, P. (1988). Memory
complaints in the elderly: A study of 367 community dwelling individuals
fiom 50 to 80 years old. Symposium: Memory and Aging, Lausanne,
Switzerland. (From MW, 1989, Abstract No. 76-39701.)

53

Devolder, P.A, & Pressley, M. (1991). Memory complairnts in younger and older
adults. Applied Cogmp' 've Psycholegy, 5, 443-454.

Feeharn, M., Knight, RG., & Partridge, F. (1991). Coglitive complaint and test
performance in elderly patients sufl‘ering depression or dementia.
Intematiorml Joun_n_a,l of Geriatric Pachiatg, 6, 287-293.

Feinberg, T., & Goodman, B. (1984). Afi‘ective illness, dementia, and
pseudodernentia. Journal of Clinieal Psychiatry, 45, 99-103.

 

Fisch, RZ. (1987). Masked depression: Its' interrelations with somatization,
hypochondriasis, and conversion. Internatiogl J ournpl of Psychiatry in
Medicine 17 367-379.

 

Fuhrer, R, Antonucci, C., Gagnon, M., Dartigues, F., Barber-Gateau, P., &
Alperovitch, A ( 1992). Depressive symptomatology and cognitive
finnctioning: An epidemiological survey in an elderly community sample in
France. Psychologieel Megh'cine, 22, 159-172.

Gallagher, D., Nies, G., & Thompsorn, L.W. (1982). Reliability of the Beck

Depression Inventory with older adults. Journal of Censultmg' and Clinicg
Psycholegy, 50, 152-153.

Glbbons, RD., Clark, D.C., Cavanaugh, S.A. & Davis, J.M. (1987). Application of
modern psychosomatic theory in psychiatric research. Jeurnal of Psychiatric

Reseamhfls 43-55.

Goldstein, SE. (1979). Depression in the elderly. Journal of the Amep'fl
gm 'cs Socieg, 17, 27-38.

Hendersorn, AS, Scott, R, & Kay, D. W. K. (1986). The elderly who live alone:
Their mental health and social relationships. WW

f P chi 20 202-209.

Kahrn, RL., Zarit, S.H., Hilbert, NM. & Niederehe, G. (1975). Memory complaint
and impairment in the aged. Archives pf @eral Psyclniepy, 32, 1569-1573.

Karanci, N. (1988). Patterns of depression in medical patients and their relationship

with causal attributions for illness. Pachotherepy gd thosomatjes, 50,
207-215.

Katorn, W. (1982). Depression: Somatic symptoms and medical disorders in
Pfimal'y care. MW, 23, 274-287.

54

Klerman, G.L. (1988). Problems in the definition and diagnosis of depression in

the elderly. In Breslau LO, Haig (Eds): mum and Agp' g: Cases,

W New York: Springer Publishing Company, 1983,
3-19.

Kral, V. A. (1962). Senescent forgetfirlness: Benign and malignant. Journ_al of the
i M 86 257-260.

Lamberty, G.J. & Bieliauskas, LA (1993). Distinguishing between depression and
dementia in the elderly: A review of neuropsychological findings. Archives
of Clinical Neuropgchoiogy, 8, 149-170.

Larrabee, G. J., & Levirn, H. S. (1986). Memory self-ratings and objective test
performance rn a normal elderly sample. mm
W _mhelmé, 275-284

Larrabee, G. J., West, RL., & Crook, T. (1991). The association ofmemory
complaint with computer-simulated everyday memory performance. J oumal

 

Lawton, M.P., Moss, M.A, Fulcomer, M & Klebarn, MB. (1982). A research and
service oriented multilevel assessrnernt instrument. Jorgpsl of (fiscntology,
17, 91-99.

Lesse, S. (1983). The masked depression syndrome-results of a seventeen year
clinical study. Amep'cep Jeugfl of Psychothgepy, 37, 456-475.

Louks, J., Hayne, C., & Smith, J. ( 1989). Replicated factor structure of the Beck
Depression Inventory. f Nervo s d Men Di 5 177
473-479.

Makanjuola, J .D., & Olaifa, EA (1987). Masked depression in ngerians treated
at the Neuro-Psychiatric Hospital Aro, Abeokuta. Age Psyehietrica
mm 480-485.

McDonald, W. M., & Nemerofl', CB. (1991). Neurotransmitters and
neuropeptides in Alzheimer's disease. Psychiatric Clinics of North Americe,
2, 421-442.

McEntee, W.J., & Crook, TH. (1990). Age associated memory impairment, a role
for catecholarrnines. Nflolegy, 40, 526-530.

Mclean, S. (1987). Assessing dementia: Dificulties, definition and differential
diagnosis. 'an N Z fP hi 21 142-174.

55

Miller, E., & Lewis, P. (1977). Recognition memory in elderly patients with
depression and dementia: A signal detection analysis. Ipprpal of Abnennal
Psycholpgy, 86, 84-86.

Molinari, VA. (1991). Mental Health issues in the elderly. Special issue. The
mentally impaired elderly: Strategies and interventions to maintain function.
Pusical and Qc_cu_petipna1 Therepy in Geriatp'cs, 9, 23-30.

Murphy, E. (1983). The prognosis of depression in old age. Brain Journal of
Psychiatsy, 142, 111-119.

Neskes, RE, & Jarvik, LP. (1987). Affective disorders in the elderly. Annual
Review of Medicine. 38. 445-456.

Newman, J .P. (1989). Aging and depression. chhology and Aggg,’ 4, 150-165.

Nussbaum, P.D., Kaszrniak, AW., Allender, J., & Rapcsak, $1. (1991).
Depression and cognitive deterioration in the elderly: A follow up study.

Jeurnal pf Clirnieal epd Em‘ entel Psychelpgy, 13, 100.

O'Briern, J.T., Beats, 8., Hill, K., & Howard, R (1992). Do subjective memory
complaints precede dementia? A three year follow-up of patients with
supposed "benign senescent forgetfirlness. “ Integm'onal J oumel of Gen'_a_t_rl_je
Psychiatn, 7, 481-486.

Pettinati, H.M., Brown, M., & Mathisern, KS. (1985). Memory complaints in
depressed geriatric inpatients. Annals of New York Academy of Sciences,
4444, 528-530.

Plotkin, D.A, Mintz, J. & Jarvik, LP. (1985). Subjective memory in geriatric
depression. American Journal of Psychim, 142, 1103-1105.

 

Rapp, S.R, & Vrana, CS. (1989). Substituting nonsomatic for somatic symptoms
in the diagnosis of depression in elderly male medical patients. m
qurnal efPsychiagz, 146, 1197-1200.

Rapp, S.R, Walsh, D.A, Parisi, SA, & Wallace, CE. (1988). Detecting
depression in elderly medical inpatients. 10211131 of Consulpp' g epd Clinical
Pachoiogy, 56, 509-513.

Reugg, RG., Swerdlow, NR, & Zisook, S. (1988). Depression in the aged.
Pachiatric Clinics of North Americe,_11, 83-99.

Salzman, C., & Shader, RI. (1978). Depression in the elderly: Relationship
between depression psychologic defense mechanisms and physical illness.
19m efthe Amerigp @‘etnies Socigy, 26, 253.

56

Scogin, F., Storandt, M., & Lott, L. ( 1985). Memory-skills training, memory
complaints, and depression in older adults. J ournel of Gerontology, 40,
562-568.

Spitzer, R, Endicott, J. & Robins, E. (1978). Research diagnostic criteria:
Rationale and reliability. Archives of Cmgfl chhim, 35, 773-782.

Shultz, BM. (1991). Minimizing the use of psychoactive medications in the elderly.
Clinical fientolegy, 11, 80-83.

Sternberg, D.E., & Jarvik, ME. (1976). Memory firnction in depression.
Arehives of @eral Psyehiapry, 33, 219-224.

Stewart, RB., Blashfield, R, Hale, W.E., & Moore, MT. (1991). Correlates of
Beck Depression Inventory scores in an ambulatory elderly population:
Symptoms, diseases, laboratory values, and medications. Journal of Femily
Practice 32, 497-502.

 

Taylor, J.L., Miller, T.P., & Tinklenberg, JR (1992). Correlates of memory
decline: A 4 year longitudinal study of older adults with memory complaints.
Psychology Ed Am 7, 185-193.

Teeharn, M., Knight, RG., & Partridge, FM. (1991). Cognitive complaint and test
performance in elderly patients sufl‘ering depression or dementia.
_ImMQpel Journel of Geriatjc Psychology, 6, 287-293.

Thompson B.A & Gallagher J .L., ( 1990). Testing and mnemonic strategies. In
J .E. Birren & K.W. Schaie, Handbook of the Psychology of Ag'pg (3'rd ed)
(pp.362-375). NY. Academic Press Inc.

 

Volk, RJ., Pace, M. & Parchman, ML. (1993). Screening for depression in
primary patients: Dimensionaiity of the short form of the Beck Depression

Inventory. Pachological Assessment, 5, 173-181.

Waxman, H.M., & Camer, EA (1984). Physicians recognition, diagnosis, and
treatment of mental disorders in elderly medical patients. The mmelogist,
24, 593-597.

Waxrnan, H.M., McCreary, G.M., Weinrit, RM. & Carner, EA (1985). A
comparison of somatic complaints among depressed and non-depressed

elders. The (fiemoleg'st, 25, 501-507.

Weiss, I.K., Nagel, C.L. & Aronson, MK. (1986). Applicability of depression

scales to the old old person. Journal of the America (filiapjcs Socim, 34,
21 5-218.

57

Weissman, H.M. & Myers, J.K. (1977). Rates and risks of depressive symptoms in

a United States urban community. Acta chhiatric Scandinavie, 57,
219-231.

Williams, J .M., Little, M.M., Scates, S. & Blockman, N. (1987). Memory complaints and
abilities among depressed older adults. Jom of Counseling and Clinical
chholegy, 55, 595-598.

Yesavage, J.A, Brink, R, & Rose, TL. (1983). Development and validation of a
Geriatric Depression Scale. Journal of Psychiatric Research 17. 31-49.

Youngjohrn, JR, Larrabee, G.J., & Crook, TH. (1992). Discriminating age
associated memory impairment fiom Alzheimer's disease. chholog’cal
Mment, 4, 54-59.