LIBRARY
Michigan State
University

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuapBS—p. 1 5

DEPRESSION IN ASSISTED LIVING FACILITIES
By

Deborah Banazak Wagenaar

A THESIS

Submitted to
Michigan State University
In partial fulfillment of the requirements
For the degree of

MASTER OF SCIENCE
Department of Epidemiology

2002

ABSTRACT
DEPRESSION IN ASSISTED LIVING FACILITIES
By

Deborah Banazak Wagenaar

Assisted Living Facilities (ALFS) are a quickly growing component of the senior
residential housing market, with over 11,200 facilities nationally providing care for
elderly Americans. In 1999, the US Surgeon General reported that although nearly 20%
of all Americans have some sort of mental illness, only half will receive treatment.
Although the demographic, functional and psychiatric status of nursing home and foster
care patients has been well defined in psychiatric literature (3-6), little is known about
ALF residents. Given this growth, our proposal seeks to study the characteristics of
depression in a novel new setting (the ALF), as a basis for future interventions-based
research in ALF s. This project will: (1) determine the prevalence of depression in the
ALF; (2) describe the clinical characteristics of the depressed ALF population; (3)
delineate the current pattern of depression treatment in an ALF; and (4) highlight the
level of unmet clinical need for depression treatment in the ALF population. Using a
two-step process patterned after large community survey projects such as the ECA or
NDS, we will first screen for depression and then define diagnosis and treatment patterns
using three interviews. The first semi-structured interview will include data from the
MAI, HAM-DRS, PRIME-MD PHQ and demographics. The second interview will
consist of the SCID to determine a diagnosis of major or minor depression. Finally, the
third interview will elicit data on current depression treatment patterns as well as barriers

to service.

This thesis is dedicated to my grandmother Bernice Banaszak,

Whose life inspired my work.

iii

ACKNOWLEDGEMENTS

My sincere thanks goes out to my thesis committee, Drs. Paneth, Colenda and Gardiner

for their guidance, inspiration and helpful advice throughout my program of study.

I am grateful to the Departments of Epidemiology and Psychiatry, for providing both
time as well as financial resources to allow me to complete this degree. To the Colleges

of Osteopathic and Human Medicine, I thank you for your support.

My friends and family have encouraged this effort in incredible, unselfish ways over the
past two years. All of you are special people. My sincere gratitude goes to Lori, Nancy,

Linda, John, Sharon, Eric and the staff at Porter Hills, HHS and the Dept of Psychiatry.

To my husband Larry and our children, I am forever indebted. You have supported and

sheltered me through this process. I love you all ...... See, I really did need a master's

degree! ! .’

iv

TABLE OF CONTENTS

LIST OF TABLES ........................................................................................................ vii
LIST OF ABBREVIATIONS ...................................................................................... vii
I. SPECIFIC AIMS ........................................................................................................ 1
II. BACKGROUND AND SIGNIFICANCE ................................................................ 2
11a. Depression: Impact on Cost, Medical Illness ................................................... 3
11b. Depression: Community Based Surveys .......................................................... 5
He. Depression: Site-Specific Prevalence Studies .................................................. 7
IIcl. Primary Care Settings ...................................................................... 8

H02. Home Care Settings ......................................................................... 9

1103. Gaps in Our Knowledge Base ....................................................... 10

IId. Assisted Living: A Novel Setting for the Study of Depression ..................... 11
Me. What is an ALF and Why is it a Service Setting Worth Studying? ............... 12
He]. ALF Study Definition ................................................................... 12

IIe2. How do ALFs Differ from Nursing Homes? ............................... l4

IIe3. Who Lives in an ALF? ................................................................ 16

IIf. Depression Treatment for Older Adults: Adequacy ..................................... 18
fig. Rationale for this proposal ............................................................................ 19
III. PRELIMINARY STUDIES ................................................................................... 20
IIIa. Homes for the Aged: A Survey of Administrators ....................................... 20
[Kb Depression Treatment Patterns in ALF S: A Pilot ........................................ 23
IV. APPROACH AND METHODS ............................................................................ 27
Na. Overview ...................................................................................................... 27
M. Approach ...................................................................................................... 27
No. Data Collection Procedures ......................................................................... 31
N01. Study Design .............................................................................. 31

W02. Interviewers ............................................................................... 33

IV 03. Sampling Frame/Eligibility Criteria ......................................... 34

IVc4. Recruitment and Consent .......................................................... 38

IV c5. Interview #1 ............................................................................. 40

IV c6. Interview #2 ............................................................................. 42

IVc7. Interview #3 ............................................................................. 43

V. DATA ANALYSIS ................................................................................................... 46

Va. Primary Aim #1 ............................................................................................. 47
Vb. Primary Aim #2 ............................................................................................. 50
Vc. Primary Aim #3 ............................................................................................. 51
Vd. Secondary Aim #1 ......................................................................................... 51
Ve. Secondary Aim #2 ......................................................................................... 52
VI. IMPLICATIONS OF THIS STUDY .................................................................... 53
VII. REFERENCES ...................................................................................................... 57
APPENDIX I
RESEARCH CONSENT FORM ..................................................................................... 64
APPENDIX 11
MINI MENTAL STATE EXAMINATION .................................................................... 71
APPENDIX III
PRIME MD PHQ ............................................................................................................. 74
APPENDIX IV
MULTILEVEL ASSESSMENT INSTRUMENT ........................................................... 77

vi

Table I

Table II

Table HI

Table IV

Table V

Table VI

Table VII

Table VHI

Table D(

Table X

Table XI

LIST OF TABLES

Site-Specific Prevalence of Major Depression in Older Adults ......... 8
Most Common Mental Health Problems in the ALF ..................... 21
Mental Health Treatments Available in the ALF .......................... 21
Mental Health Service Providers in the ALF .............................. 22
Barriers that Prevent Residents from

Receiving Mental Health Services .......................................... 22
Pharmacotherapy Provided for NH and ALF Residents .................. 24
Project Scope and Timeline ................................................... 30
Project Flow Chart .............................................................. 33
Sample Strategy for ALF Depression Study ................................ 38
Expected Key Sample Demographics ........................................ 39
Sample Size Determinations based upon Prevalence ....................... 49

vii

LIST OF ABBREVIATIONS

ADLS ............................................................................................ Activities of Daily Living
ALFA ......................................................................................... Assisted Living Federation
ALFS ............................................................................................. Assisted Living Facilities
CES-D ........................................................................... Center for Epidemiological Studies
CIDI ................................................................................... Composite Diagnostic Interview
CMH ........................................................................................... Community Mental Health
DIS ....................................................................................... Diagnostic Interview Schedule
DSM ................................................................................ Diagnostic and Statistical Manual
ECA .................................................................................... Environmental Catchment Area
ER ............................................................................................................. Emergency Room
HAM DRS ..................................................................... Hamilton Depression Rating Scale
HMO ............................................................................... Health Maintenance Organization
MAI .................................................................................. Multilevel Assessment Inventory
MDCIS ...................................................................... Michigan Dept of Consumer Industry
MMSE ................................................................................ Mini Mental State Examination
MSU ............................................................................................ Michigan State University
NCS ........................................................................................ National Comorbidity Survey
PRIME MD PHQ ................................................ Prime MD Patient Health Questionnaire
SCID I/P .................................................................. Structured Clinical Interview for DSM

viii

1. Specific Aims

The primary aims of this proposed study are to:

1. Determine the prevalence of Major and Minor Depression in assisted living facilities

(ALF).

2. Describe the clinical and functional status of patients with Major and Minor

Depression in an ALF.

3. Describe the adequacy of current depression treatment as compared to “gold

standar ” depression treatment guidelines.

The secondary aims of this proposed study are to:

1. Identify potential barriers to care that make the ALF a unique setting to provide

depression treatment.

2. Establish the level of unmet clinical need for depression care in patients residing in

anALF.

ALFs are a quickly growing component of the senior residential housing market, with
over 11,200 facilities nationally providing care for elderly Americans. ALFS provide on-
site assistance with daily living activities, meals and domiciliary needs. Up to 7,000,000
older Americans currently need assistance with these activities with this number expected
to double by 2020 (1). The ALF industry is projected to continue its rapid expansion as
an alternative to nursing home or foster care placements.

In 1999, the US Surgeon General reported that although nearly 20% of all Americans
have some sort of mental illness, only half will receive treatment (2). Who is treated for
depression in an ALF, the types of treatment available and the nature of the provision of
services are largely unknown for the ALF population. Although the demographic,
functional and psychiatric status of nursing home and foster care patients has been well
defined in psychiatric literature (3-6), little is known about ALF residents.

Given this growth, our proposal seeks to study the characteristics of depression in a
novel new setting (the ALF), as a basis for future interventions-based research in ALFs.
It is a critical first step in understanding the nature of the problem of depression in ALF s

if we are to design meaningful and effective interventions for ALF residents.

11. Background, Significance and Rationale

Although depression is a major public health problem in this country (7), it is an
especially devastating illness for older adults. A frail elder living in an assisted living
facility (ALF) is vulnerable to the impact of weight loss, decreased mobility, fatigue and

impaired sleep that accompanies depression. Family involvement in care may be

limited for ALF residents, leaving only the staff to recognize and seek treatment for
depression. Unfortunately, without adequate mental health training, depression may go
unrecognized by ALF staff, misconstrued as a normal part of aging. As the ALF is a
growing housing market for older adults and is ofien seen as a better alternative to
nursing home treatment, it is imperative that we understand the problem of depression in

this setting as well as the degree of umnet clinical need present in the ALF environment.

Ila. Depression: Impact on Cost, Medical Illness and Quality of Life

Depression has broad impact upon the financial, functional and health status of older
adults. The healthcare financing system is stressed as depressed elderly patients have
more emergency room and outpatient visits as well as twice the annual health care costs
(doctor, hospital, pharmacy, laboratory) of non-depressed elderly (8). Examining the
care of 13,653 enrollees of a Health Maintenance Organization (HMO), Fischer (9)
reported that having depressive symptoms was associated with a 19% increase in the
number of outpatient encounters and a 30% increase in total outpatient charges.

Depressive symptoms were not associated with increased inpatient utilization or charges.

Major Depression also increases morbidity and mortality fi'om medical illnesses (10).
The depressed are three times more likely to die within the fust year of treatment while
living in a nursing home than when compared to the non-depressed (6, 11-12). In a
prospective cohort study, Callahan (1998) followed 3,767 outpatients for 45 months.

Using a proportional hazards model he found that baseline depression score, age, gender

and serum albumin value were significant correlates to time of death. In a 6-year
prospective cohort study, Schoevers (14) obtained similar results in Britain, suggesting
that older persons with depression had a 1.676 fold higher mortality risk than those

without depressive symptoms.

An already frail depressed patient who suffers from weight loss may be at risk for
dehydration, malnutrition and impaired immunity leading to systemic infection (15). In
a sub acute-care facility, Thomas (16) discovered that only 8% of subjects could be
classified as well nourished and that elevated scores on the Geriatric Depression Scale

were associated with malnourishment for their patient sample (p=.05).

Depression can also impact the ability to walk, bathe, dress or feed oneself.
Callahan's 6-year prospective cohort study of over 3700 individuals concluded that
elderly patients with depressive symptoms had significantly worse physical and
psychosocial function scores on the Sickness Impact Profile than did those without
depression (p<.001) (13). Examining patients with heart failure, Vaccarino (17) noted
that individuals with greater than 11 depressive symptoms at baseline, when compared
with those with less than 6 depressive symptoms, had an 82% higher risk of either
fimctional decline or death, whereas the immediate levels of depressive symptoms

showed intermediate risk (p=.0003 for trend).

IIb. Depression: Community-Based Prevalence Surveys

Depression affects 5 million elderly in the United States (7). The prevalence and
treatment of depression has been well delineated in some settings (community, primary
care, nursing home) and virtually unstudied in other arenas (ALF).

Few community-based surveys have focused exclusively upon the elderly. Most often
data for persons age 65+ are found as a subset of a later general adult community survey.
One such large community survey is the National Institute of Mental Health’s
Epidemiological Catchments Area (ECA) project. The ECA was conducted from 1980 to
1985 in 5 sites providing 18571 households and 2290 residents’ age 18 or older. Alter
telephone screening using a structured interview designed to detect depressive symptoms,
two face-to-face interviews were conducted 12 months apart. These interviews used the
Diagnostic Interview Schedule (DIS) (17) to confirm a Diagnostic and Statistical Manual
11] specific diagnosis of depression. Blazer (18-19) was one of the first psychiatrists to
examine the ECA data for late life depression using 3,015 elderly individuals fi‘om the
ECA Piedmont site in North Carolina. He found 3.5% had Major Depression, 2.1% were

Dysthymic and 0.3% experienced Bereavement.

The National Comorbidity Survey (NCS) is another large community-based survey of
subjects age 15-75. The NCS was a cross-sectional survey of nationally representative
households of 8,098 adolescents and adults age 15-75 conducted from 1990 to 1992.
Kessler (20) examined NCS data collected by lay interviewers using a revised version of

the Composite International Diagnostic Interview (CIDI). Over 3500 subjects with multi-

system medical illness were first screened for depression using a structured telephone
interview followed by a definitive DSM IIIR diagnosis using the CIDI. Nearly 50% of
elderly reported at least one episode of depression and close to 30% reported at least one
12-month episode of depression. Among people with a lifetime history of three or more
comorbid medical illnesses and depression, the proportion who ever obtained specialty
sector mental health treatment was less than 50%. No data from ALF residents were

reported in the Kessler study.

Recently Narrow (21) reanalyzed ECA and NCS data using DSM-IV TR diagnostic
criteria for greater precision in estimates. To operationalize the definition of depression
and highlight disability, Narrow examined questions related to life interference,
professional treatment or using a medication for symptoms from the DIS for both ECA
and NCS community surveys. The focus upon these functional areas was designed to
address the disparity between the NCS and DIS report of mood disorder prevalence
(3.5% vs. 50%). Using the criteria for functional and clinical significance, 15.2% of
individuals age 54 and older had clinically significant depression in contrast to 17.2% of

the younger cohort in the combined analysis.

ECA data has also provided interesting gender differences in the study of Major
Depression. Using NIMH ECA data from New Haven CT, Weissman (22) obtained 6-
month prevalence based upon a screening phone interview followed by the DIS. Major

Depression was more frequent in women than in men (24:1) and sex ratios were fairly

consistent across generations, including the elderly. The cohort born before 1936 had an

earlier age of onset with higher rates of Major Depression.

Of note, as the ECA and NCS data were based upon older versions of DSM (III and
IIIR) using diagnostic measures appropriate for their time such as the CIDI and DIS . As
the American Psychiatric Association has completed a new updated manual, the
Diagnostic and Statistical Manual IV (DSM-IV TR), our study will use an updated
definition from DSM-IV TR. We will use two brief screening instruments for depression
(PRIME-MD PHQ)(47) and the Hamilton Depression Rating Scale (HAM-DRS) (49)
followed by the Schedule of Clinical and Diagnostic Interview (SCID-I/P)(48). The
SCID-I/P provides DSM-IV TR specific diagnosis of both Major and Minor Depression
while paralleling the two-step methodology used for the original large ECA and NCS

community surveys.

Ilc. Depression: Site-Specific Prevalence Studies

The prevalence of clinically significant depression has been described in a variety of

arenas. Table I delineates the site, assessment tool and prevalence rate Major Depression

for older adults evaluated in several different settings.

Table I: Site-Specific Prevalence of Major Depression in Older Adults

Age
Assessment Size

 

(GDS) &Zung Depression Scale (Zung) AZung Self-Acting Depression Scale
+Diagnostic Interview Schedule (DIS) %Structured Clinical Interview for Axis I
Diagnosis

IIcl. Primary Care Settings

Depression assessment in primary care has varied widely. Some studies (23-25) have
used screening instruments such as the Center for Epidemiological Studies-Depressed
(CBS-D) or the Hamilton Depression Rating Scale (HAM-DRS) as a definitive diagnostic
tool to identify subjects. False negatives, as well as false positives created by the
excessive endorsement of somatic symptoms, makes the use of these screening tools as

the definitive criteria for subject inclusion questionable at best. These studies report

prevalence figures 10-12% for Major Depressive disorder. When a more specific
screening tool is used that provides DSM-IV TR specific diagnoses such as the PRIME-
MD (26), the prevalence of Major Depression is 18.9%. Olfson used the PRIME-MD to
examine 1,255 primary care immigrant subjects age 45-83, finding that psychiatric

comorbidity was extensive.

Ilc2. Home Care Settings

As home care services grow in popularity as a means to decrease hospital stays and
avoid nursing home placement, the study of depression within the home environment is
increasingly relevant to clinical care. Using ECA data from New Haven, Connecticut
Bruce (27) examined depression in 2,553 non-institutionalized elders in New Haven.
Employing the DIS as the primary diagnostic tool, Bruce reported that 2.3% of elders had
Major Depression and 3.9% had Dysthymia. Depression was twice as prevalent among

elders confined to a bed or chair as among non-homebound elders.

In a large follow-up study of 2003 elderly visiting nurse clients residing in New York,
Bruce (28) evaluated homebound elders with co-morbid medical conditions. To obtain a
DSM-IV TR specific diagnosis of mood disorder, she used the SCID-I/P. Bruce found
that 13.5% of medically ill clients suffered from Major Depression while 10.5% had

Minor Depression.

lIc3. Gaps in Our Knowledge Base

Our search of the literature confirms the fact that there are no studies establishing the
prevalence of depression in an ALF environment. Primary care and homecare
populations are most likely to match the demographics and comorbidy profile of the
typical ALF resident. Using a two-tiered screening/diagnostic testing methodology well
established by the ECA and NCA, this study will first screen ALF residents for
depression using the PRIME-MD PHQ and then establish a DSM-IV TR diagnosis using
the SCID-I/P. We will further study those patients identified with Major and Minor
Depression, examining their functional characteristics, depression treatment patterns and
degree of adherence to “gold standar ” practice guidelines. The project is in preparation
for an intervention study, as characterizing the ALF environment is critical if a proposed

intervention is to be accepted and successful for its residents.

Given the critical nature of depression and the impact that effective treatment can
have upon function, medical illness and quality of life, we believe the assisted living
setting is ready for study, because the area is growing and little literature has been
published describing mental health issues in the ALF setting. A systematic literature
review of Medline and Psych Info search databases using the terms “quality of life”
“mental health” “psychiatric illness”, “depression”, “mood disorder” or “assisted living”
“foster care” and “functional status” “homes for the aged” or “domicile” provided no

studies evaluating depression and quality of life in the ALF setting

10

Ild. Assisted Living: A novel setting for the study of depression

There are currently 35 million people age 65 and older in America, accounting for
13% of the total US population. The population over age 85 is expected to grow flour 4
million in 2000 to 19 million in 2050 (29). While most elderly reside in their own
homes, as many as 6.5 million older people need assistance with activities of daily living
(ADL’s). Nursing homes have historically been the primary place where this type of care

has been provided (1).

Assisted living facilities are growing rapidly as an alternative to the high labor costs
and turnover, heavy regulation and stigrnatization faced in the nursing home industry,
while nursing home age-adjusted admission rates have declined over time. Forty-five
persons per 1,000 in 2000 compared to 54 per 1,000 in 1985 were admitted to a nursing
home. Because the nation has aged since 1985, were these data age-standardized and
age-specific, this decline would be even more dramatic. A recent national survey found
11,472 assisted-living facilities nationwide, accommodating 5,558,400 residents (30).
Other sources suggest this number is actually much higher. The Assisted Living
Federation of America (ALFA), the national for-profit assisted living organization,
suggests that ADL care can be provided in a number of for-profit or private assisted
living settings (1). About 55% of all ALFs are freestanding, 35% are attached to a
campus (usually nursing facility) and 10% are associated with an independent apartment

complex (3).

11

Ile. What is an ALF and Why is it a Service Setting Worth Studying?

IIel. ALF Study Definition

There is no single national definition for an ALF. Each organization associated with
assisted living espouses its own definition of an ALF. The Assisted Living Federation of
America (ALFA) (1) defines assisted living as “the combining of housing, personalized
supportive services and health care designed to meet the individual needs of persons who
need help with activities of daily living (ADL), but do not need skilled medical care
provided in a nursing home”. The lack of skilled medical care differentiates the ALF
environment from a skilled care nursing home setting. Skilled care may include
intravenous medication, tube feeding, physical/occupational therapy and

cardiopulmonary treatment.

The US Department of Health and Human Services sponsored a survey in 2000 that
detailed the characteristics of ALF s and their residents. Hawes and Phillips (30)
surveyed 1500 ALF s across the United States to obtain information on the policies,
practices, residents and staff as well as service configurations available in these facilities
(32). This is the largest national survey of ALFs available. The research group found
that ALFS are almost equally likely to be operated by for-profit as by not-for-profit
entities (50% vs. 49%). Two thirds of ALFs were engaged in some type of additional

long-term care service (i.e., nursing home, other ALF s) at the time of the survey.

12

In addition to a variety of definitions available for an ALF, a number of terms have
been used interchangeable to describe an ALF environment including “supportive care”,
”residential care” “homes for the aged”, “foster care”, “board and care homes” and
“group homes”. Each state may use a different term to describe these facilities within
their local. No single term has been adopted as a “gold standard” by the United States
government or legislature. The United States Department of Health and Human services
simplifies the debate by referring to all facilities as “assisted living”, referring to the fact
that these facilities provide residents with either routine general protective oversight or

assistance with activities necessary for independent living (33).

To add to the confusion, state or federal licensure for ALFs is voluntary in the United
States. Unlike nursing homes or hospitals that must pass licensing and accreditation
standards to function and receive federal support, an ALF may choose to be licensed or
unlicensed. The licensing process is costly and involves a review of the facility and
procedures. This licensing dilemma makes it very difficult to track facilities for resident

status, medical care and resident outcome.

Given these issues, what ALF definition should we adopt for this study? As the
Michigan Department of Consumer and Industry Services (MDCIS) Bureau of
Regulatory Systems is charged with regulating and licensing ALFS in the state of

Michigan, we will use the MDCIS definition of an ALF. Under this criteria, an ALF is

defined as “a supervised personal care facility, other than a hotel, foster care, hospital,

13

nursing home or coring medical facilig, that provides room, board, and supervised

 

personal care to 21 or more unrelated non-transient individuals 60 years of age or older”

 

(34). The two ALFS we will use for this study, Porter Hills Presbyterian Village and
Independence village are licensed by the state of Michigan and meet the study definition
criteria. Given this fact, we believe that results obtained by these two facilities will be

generalizable to other ALF communities.

IIe2. How Do ALFs Differ From Nursing Homes?

ALF s differ from nursing homes in structure, financing and philosophy of care. The
Center for Disease Control and Prevention (35) defines nursing homes as "skilled
facilities providing 24-hour nursing care to sick or disabled patients who do not require
hospitalization, but cannot be cared for at home or through other community options."
Often nursing home patients are admitted either for brief rehabilitation services not able
to be provided in the ALF or for end-of-life hospice comfort measures. According to the
National Nursing Home Survey (NNHS) (36), the average nursing home resident is 82.6
years old with multi-system medical disease. The mean length of stay is 45 days with
30% recuperated back to the community, 36% returned to the hospital and 25% dying.
Unlike the ALF where residents may live for many years, the nursing home has become a
“short-term” stay facility with high medical acuity and medical service needs. These
services (e. g. 24 hour nursing, rehabilitative services, face-to-face physician intervention)

are not generally provided in an ALF setting.

14

ALF s are regulated less intensely than are nursing homes. Unlike nursing homes in
which licensing is mandated, an ALF may choose to be licensed or unlicensed within the
state of Michigan. This creates a diverse set of admission, treatment and discharge
criteria among not-for-profit and entrepreneurial assisted living homes. In 1987 Congress
passed the Omnibus Reconciliation Act (OBRA-87), which regulated restraints,
psychotropic treatment and mental health monitoring in American nursing homes.
OBRA-87 legislation caused reduced psychotropic prescribing practices and aggressively
monitored nursing home admission practices to filter out mentally ill patients warehoused
in nursing homes. However, ALF 5 do not have comparable regulations in place, leaving

the study of how depression is recognized and treated even more important (3 7).

Philosophical and financial issues also help to distinguish the ALF environment from
that of a nursing home. Unlike nursing homes that operate on a medical model of care,
the ALF care model falls into one of three categories: 1) the hospitality model offers a
“hotel-like” atmosphere; 2) the personal care model offers care for the frail elderly and 3)
the aging-in model offers onsite care for chronic conditions (38). Assisted living
facilities have espoused an “aging in place” philosophy that provides services in a
graduated manner dependent upon resident needs (39), relieving families of the burden of
multiple moves at the end of life. In addition, ALFS cost money and are not reimbursed
by Medicare or Medicaid. The ALF may contain a growing but unique subgroup of
Americans, i.e., those middle-upper class individuals who are able to support a high cost
lifestyle (i.e., the typical ALF monthly rental fee is $2,000 or more) based upon savings

or other assets. This financial requirement may exclude individuals of color, those with

15

low socioeconomic status or those with chronic persistent mental illness such as
schizophrenia, who may often find themselves the recipients of care provided by their

fiiends and relatives within a home environment.

lle3 Who lives in an ALF?

Over 16% of all Michigan residents are age 60 or older (9,759,306 x 0.25 = 1,561,489
persons) (40). Of note, 130,591 of these individuals are over age 85. In Michigan, there
are nearly two times more females than males age 65 and older. With respect to race,
80% are white, 14.2% are black, 4% are Hispanic and 1.8% are Asian. Given the fact
that of this 85 yr + population only 37% own their own homes (130,591 x 0.37=48,318)
and 4.4% are in nursing homes (130,591 x 0.04:5,223), it is possible that the remainder
of elderly over age 85 (n=77,050) may reside in some type of "other” supervised housing
situation (family, fee-for-service caregivers or ALF s). Michigan has both licensed and

unlicensed ALFS that have a total of 47,761 beds available for residents (41).

According to the US Department of Commerce’s Census 2000 (40) the population
age 65 or older is approximately 12%. Although the population age 65 and over grew
slower than the total population, among the older population, those age 85 years and
older showed the highest percentage increase (40). This “old-old” cohort comprises 53%
of the older population who may require more intense residential and medical care than
their own homes provide. There were fewer people age 85+ living in nursing homes in

2000 than 1990 (18.2% vs. 24.5%), suggesting that other alternatives such as home care

16

services where meals or therapies are brought to the patient in their own home or ALFS
may be growing to meet the increased desire of the elderly to live out their last days in a
home-like environment. National Center for Health Statistics data suggest that homecare
is growing, with up to 25% of seniors age 65 and older receiving services (42).
Interestingly, the NCHS database does not offer demographic data on seniors living in

ALF s to serve as a comparison for those in home health care.

In Hawes and Phillip’s national survey of ALFS, over half of the residents living in
ALFS were age 85 or older (30). Seventy-eight percent (SE=1.11) of the 184,558 ALF
residents surveyed were female and 98.7% (SE=0.55) were white. With respect to
marital status, 70.8% (SE=1.64) were widowed, 12.1% (SE=1.00) were married, 9.9%
(SE=1.23) were never married and 7.2% (SE=0.70) were divorced. Three quarters had
living children and 85.9% (SE=1.90) described relatives that lived within an hour. One
quarter saw relatives twice a week, another 26.7% (SE=1.44) reported more than one visit
weekly and 6.3% (SE=0.91) had daily family visits. The sample was well educated, only
26.8% (SE=2.23) had not graduated from high school while 20.3% (SE=1.80) had a
college degree. In terms of financial status, one quarter reported income less than
$14,000, one quarter endorsed income less than $24,000 and one quarter had less than

$50,000 annual income.

17

IIf. Depression Treatment for Older Adults: Adequacy and Effectiveness

Older adults respond well to treatment for depression. This treatment may
include phannacotherapy, psychotherapy and/or electroconvulsive therapy (ECT). In a
metanalysis of 40 clinical trials using only older adults, McCusker found that both
tricyclic and selective serotonin reuptake inhibitor antidepressants were effective in
reducing HAM-DRS scores (43). The efficacy of psychotherapy has also been
demonstrated in a number of studies. In a meta-analysis of 14 geriatric-specific
psychotherapy vs. placebo studies, cognitive-behavior treatment was found more
effective than interpersonal therapy or placebo (43). Similarly, ECT remains an effective
treatment for older adults with severe Major Depression with psychotic depression. It is
also helpful for patients with an inability to tolerate or respond to antidepressant

treatment (44).

Undertreatrnent of older adults with depression is a common problem, especially in
primary care (44). To evaluate the adequacy of antidepressant treatment, an Expert
Consensus Guideline on the Pharmacotherapy of Depressive Disorders in Older Patients
is now available (4546). We will use this guideline’s antidepressant treatment duration
and dosage parameters to compare with that of our ALF subjects. This is a critical first

step in understanding the adequacy of depression treatment in the ALF.

18

llg. Rationale for this Proposal

Since ALF s may become the fastest growing service sector for longitudinal care of
older adults, we must first understand the nature of the problem of depression, the status
of available depression treatment in an ALF as well as the level of unmet clinical need
within this setting. These are foundational components key to developing effective future
intervention studies. Implementing an intervention without understanding the ALF

milieu and patient population will put the intervention at risk for failure.

We propose a thorough study of Major and Minor Depression in the ALF in order to
establish the prevalence of each condition. These methods include a rigorous
methodological study of the prevalence of Major and Minor Depression. Establishing
baseline prevalence is essential as a first step to developing effective intervention
strategies to appropriately deal with the problem. Using a local ALF we will implement
DSM-IV TR specific case identification strategies to carefully identify residents with
Major and Minor Depression. These data will then help direct future intervention studies
by helping us understand how prevalent depression is in the ALF when compared to other

settings such as the primary care office or nursing home.

This project will also examine the nature and adequacy of current Major and Minor
Depression treatment for ALF residents. Using a structured clinical interview
corroborated with a family and staff interview, we will describe the types of depression

treatment provided, as well as frequency of intervention and duration of treatment. We

19

will then use the Pharmacotherapy of Depressive Disorders in Older Patients Guidelines

to access the adequacy of antidepressant dosage as well as duration (45). This
information is critical to inform our next intervention, as it will help focus future

interventions studies to be well suited and accepted by both ALF residents and staff.

Ill. Preliminagy Studies and Any Results

Illa. Wagenaar DB, Mickus MA, Luz C, Sawade J. Homes for the Aged: A

Survey of Mental Health Needs

Objective: To describe the types of psychiatric disorders found in the assisted living

setting as well as mental health service availability.

Desigp: A Mental Health Needs survey was sent to 176 administrators for Michigan

Homes for the Aged.

Participants: The questionnaire was mailed to 176 administrators in Michigan in July-
September 2001. Because staffing levels may be variable in an assisted living setting, we
chose to survey administrators because of their overall knowledgebase of the home and

its residents.

20

Rim: Ninety-four surveys were returned for a response rate of 53%. Ownership was
evenly divided between profit and non-profits. Mean resident age was 80.1 (SD=2.4).
These data suggest that administrators believe depression is the second most common
mental health problem in the ALF. Medications and activity therapy were the most
fiequently identified treatment modalities available for ALF residents. The primary care
physician, followed by the psychiatrist, provided most of the mental health care the ALF
residents received. Patient and family resistance to consultation accounted for the most

common barriers to obtaining formal mental health services.

Table 11: Most common mental health problems in the ALF
(Ranked by Percent endorsing)

 

 

 

 

 

 

 

 

Dementia 57.4% (n=54)
Depression 24.5% (n=23)
Hallucinations/Delusions 4.3% (n=4)
Anxiety 3.2 % (n=3)
Alcohol 1.1% (n=1)

 

Table III: Mental health treatments available in the ALF
(More than one response may be endorsed)

Medications 94.7% n=89

Activi 64.9% n=6l

Individual 62.8 % n=59
' ° ' 38.3% n=

F ' 30.9% n=29
' ° ' 29.8% n=28
Electroconvulsive l . 1% n=1

 

21

Table IV: Mental Health Service Providers in the ALF
(More than one response may be endorsed)

Care ' ' 59.6%
' 47.9%

Communi Mental Health 42.6%

Private ' 28.7 %
Staff member 20.2%
'est 12.8%

 

Table V: Barriers that Prevent Residents from Receiving Mental Health Services
(More than one response may be endorsed)

 

 

 

 

 

 

 

Resident refuses treatment 66% (n=62)
Family refirses treatment 56.4% (n=53)
Stigma of mental illness 26.6% (n=25)
Treatment costs too much 24.5% (n=23)
Transportation not available 17% (n=l6)
Treatment not available 13.8% (n=13)
Confidentiality concems 3.2% (n=3)

 

 

 

 

Implications for this proposal: The results of this pilot suggest that ALF administrators
believe depression is a clinically significant problem for their residents. The project we
propose will take the next step in assessing ALF residents for depression and determining
the prevalence of Major and Minor Depression. This will serve to test the accuracy of
this pilot survey, helping us compare the ALF administrators’ perceptions to objective
data obtained through subject assessment with strict diagnostic instruments. In addition,
this project will assess the types of mental health services and providers available to ALF
residents. The research we propose will serve to validate these the results of our earlier
survey in an effort to understand how to most effectively target interventions designed to

improve mental health care delivery in an ALF setting. We will see whether the barriers

22

to depression treatment endorsed by administrators actually match barriers that the ALF

resident experiences in obtaining needed mental health services.

IIIb. Wagenaar DB, Colenda CC, Kreft M, Sawade J, Merges L. Depression
treatment patterns in Nursing Homes and Assisted Living Facilities: A pilot study

(in submission)

Objective: To study the types and course of depression treatment in two Michigan
ALF s. We piloted the methods found in this ROI submission at a local level to

determine feasibility and barriers to completion.

stjgp: A prospective follow-up study identifying nursing home and ALF residents
with depression and following these individuals over the course of 6 months for the
purposes of describing treatment characteristics and clinical outcomes. We evaluated
mood (Geriatric Depression Scale), firnctional status (SF-36 (PCS and MCS
components) at study onset, month 3 and month 6. In addition, we characterized the

amount and types of mental health services provided for these residents.

Participants: Seventeen nursing home and eighteen ALF residents were identified for

this study. Resident mean age was 85.3 (SD=85.3), 88 % were female, 77% were

widowed and 12% were single in the combined cohort.

23

m: Nursing Home and ALF residents did not vary significantly with respect to GDS
scores at month 0 (6.65 vs. 7.11, p<. 08) or month 6 (4.82 vs. 5.47, p<. 11) however an
overall decrease in GDS scores between groups was noted between month 0 and 6 (p<.
04). Mean SF-36 subscale scores (MCS and PCS) did not vary significantly between the

groups at month 0, 3 or 6.

Table VI: Pharmacotherapy provided for NH and ALF residents

 

 

 

 

 

 

 

Nursing Home Resident: Assisted Livig Residents
Antidepressants 73.2% (n=12) 55% (n=9)*
Anxiolytics 11.8% (n=2) 24% (n=5)"""
Antipsychotics 10% (n=2) 24.5% (n=3)"
Hypnotics 7.7% (n=1) 20.7% (n=3)*
Mood Stabilizers O 6% (n=1)

 

 

 

 

* Significant at the p < .05 level ** significant at the p < .01 level

Overall we found that ALF residents received fewer antidepressants and more multi-
drug combination therapy residents that did NH residents. This may be due the absence
of psychotropic regulation in the ALF when compared to highly controlled psychotropic
OBRA standards enforced in the nursing home, decentralized medical care or inadequate
depression assessment. Each patient may receive medication fi'om more than one drug
class. Group means are presented, as small cell size prevents more detailed statistical

comparisons. Provider type was not defined.

ALF residents received a wider variety of psychotropic medications in this pilot
study. Many of these agents (antipsychotics, hypnotics) may be inappropriately used to

address depression in the ALF setting. Of those who received antidepressant therapy,

24

87% of Nursing Home residents were prescribed an Selective Serotonin Reuptake
Inhibitor (SSRI) while only 72% of ALF residents received an SSRI (p<. 05). ALF
residents were more likely to be treated with tricyclic compounds (9% n=2) or non-
tricyclic antidepressant drugs (9% n=2). Of note, antipsychotics were used 24.5% of
ALF patients vs. 10% of Nursing Home patients. These medications can have serious
permanent neurological side effects (tardive dyskinesia) and may be misused in the ALF

setting, as they do not treat depression.

Polypharmacy was more likely in the ALF resident group as 18% (n=3) were treated
with four psychotropic agents and 37% (n=5) were treated with two agents. This is in
contrast to the nursing home sample where no resident received four drugs at one time.
From these pilot data we conclude that further study of prescribing practices within the
ALF group is warranted. It appears that depression is less appropriately treated in this
setting, as the frequency of antidepressants is much less than that of a nursing home
sample. In addition, poly pharmacy is more widespread, suggesting the need for more

active study of how antidepressants are used in the ALF.

Implications for this proposal: This study demonstrates that depression remains fairly
consistent over time for both nursing home as well as ALF residents. Our results show
that depression is not simply a transient “adjustment reaction” to living in an ALF, but
rather a persistent medical problem. This finding makes a more compelling argument for

the need to identify effective treatment options for ALF resident.

25

In addition, this pilot work suggests that ALF resident who did receive treatment for
depression experienced more poly pharmacy than did their cohorts in a nursing home.
The addition of several classes of psychotropic agents may cause additional medication
side effects such as confusion or sedation as well as aggravate pre-existing medical
conditions. However, our sample size for this pilot was small (n=9). This proposed
study will use a much larger sample size to study psychotropic drug usage patterns in
ALF residents, to determine whether excessive use of psychotropic medications or
inappropriate dosing is seen in an ALF. This will inform the focus for firture research
projects toward educating staff and primary care physicians on appropriate depression

treatment strategies.

26

IV. The Epidemiology of Major and Minor Depression in ALFs: Study and

Approach

lVa. Overview

This project will: (1) determine the prevalence of depression in the ALF; (2) describe
the clinical characteristics of the depressed ALF population; (3) delineate the current
pattern of depression treatment in an ALF; and (4) highlight the level of unmet clinical
need for depression treatment in the ALF population. This project will sample older
adults (age 60+) living in Porter Hills Presbyterian Village in Grand Rapids, Michigan
and Independence Village in Lansing, Michigan. These facilities were chosen for the
study because they demonstrate a wide spectrum of residents with mild to significant
firnctional impairment and medical problems. They are among the largest ALF s in
western and mid-Michigan. As we anticipate that the demographic characteristics of our
ALF sample will not be dissimilar from the elderly population in Michigan or the US, we

anticipate that our results will generalize to a broader ALF population.

IVb. Approach

To identify depression within ALF residents we will use a two-step methodology to
first screen and then definitively identify subjects with Major or Minor Depression. This
two-step process is patterned after large community survey projects using ECA or NDS

data (18,20-21). Because diagnostic testing is time-consuming and potentially

27

burdensome for subjects, we wish to avoid excessive burden to our subjects, many of
which may be frail elderly with medical and functional issues that limit endurance. A
screening methodology using the PRIME-MD PHQ will enable us to identify subjects for
further study using the SCID-I/P, thereby minimizing both subject burden as well as

study cost (47-48).

The entire community of ALF residents will be invited to participate in a depression-
screening project for initial case-identification. Each resident will receive information
about the screening program and his or her written consent will be obtained. Those who
agree to the study will complete a semi-structured interview to obtain demographic data
and clinical history. Information on the degree, onset, duration of current episode of
depression, cognitive impairment and co-morbid anxiety features will be obtained during
this initial in-person interview. During this time, participants will complete the PRIME-
MD PHQ and the Hamilton Depression Rating Scale, the screening tools used to initially

identify Major and Minor Depression during Interview #1 (49).

Participants who screen positive for either Major or Minor Depression will undergo a
second 2-hour SCID-I/P structured interview to obtain a definitive diagnosis. The SCID-
I/P offers a definitive depression diagnosis as well as helps to rule out other co-morbid
conditions such as anxiety or dementia. Additional measures of function and overall
well-being will be completed including the Multivariable Assessment Instrument (MAI)
and Mini Mental State Examination (MMSE) (50-51). These measures will accomplish

one of the project’s aims, i.e. to describe the clinical and functional status of depressed

28

ALF residents. The study is expected to take 24 months to obtain an adequate sample

size and complete data collection and analysis.

Subjects whose SCID-I/P ratings suggest a diagnosis of Major or Minor Depression
will be notified of their diagnosis. In addition, a letter will be sent to their primary care
physician explaining this diagnosis as well as the need for additional evaluation and
possible treatment. If the subject endorses active suicidal, homicidal or intensive
depressive ideation, they will be referred for immediate psychiatric treatment via

Community Mental Health at Emergency Services (Lansing, MI) or Cornerstone (Grand

Rapids. MI)

During the structured interview process, we will also review current psychiatric
treatments the subject is receiving and describe the type and frequency of these therapies.
These will include psychotherapy (individual, marital, group), psychotropic medications,
activity therapy, spiritual counseling and electroconvulsive therapy. We will inquire
about antidepressant dose, length of treatment, fi'equency of medication administration,
side effect profile and mechanism of administration (self-administered vs. staff
administered). This project will use the Dgpression Expert Consensus Guidelines on

Pharmacotherapy of Dgpressive Disorders in Older Patients to assess the quality of

pharmacotherapy and adequacy of antidepressant dosing (45).

Based upon data collected during this structured interview, we will calculate the

prevalence of Major and Minor Depression. Because the SCID-I/P also provides

29

information on other DSM-IV TR diagnostic groups (i.e., anxiety, thought disorder,
somatoforrn disorder) we will report the prevalence of these conditions as well in the
ALF patient population. Subjects with co-morbid depression plus another Axis I

diagnosis will be considered separately in our analyses.

Table VII: Project Scope and Timeline

 

Project Aims Activities Timeline
Interview 1.Detennine prevalence of l.Recruit all residents Month 1-24
and #2 Major Depression in ALF of a local ALF to participate in
depression screening

 

2.Determine prevalence of
Minor Depression in ALF 2. Screen residents using the
PRIME-MD PHQ, HAM-DRS,
3.Describe the clinical MMSE and MAI.
characteristics of ALF
patients with depression 3. Structured SCID-I/P clinical
interview for subjects who

4. Describe the prevalence screen positive for depression
of comorbid psychiatric
disorders in ALF patients
with Major or Minor
Depression including
anxiety, dementia and I

 

 

 

psychosis.
Project Aims Activities Timeline
Interview 1. Describe the adequacy o 1. Develop and implement a Month 3-24
#3 current depression third structured clinical interview

treatment as compared with each subject to document

to “gold standard” depression treatment including

depression treatment psychotherapy, antidepressants,

guidelines. group therapy, ECT, activity

therapy or other modalities.
2.Establish the level

of unmet clinical need
for depression care

 

 

 

 

 

 

 

30

V. Data Collection Procedures

Va. Study Design

This project will employ a cross-sectional study design used to determine the
prevalence of Major and Minor Depression in the ALF. This project will accomplish the
following primary aim to: (1) determine the prevalence of Major and Minor Depression
in ALF; (2) describe the clinical and functional characteristics of ALF patients with
depression and (3) describe the adequacy of current depression treatment in the ALF.
Secondary aims include the ability to describe potential barriers to effective depression
treatment encountered during this project in the ALF environment and establish the level

of unmet clinical need for depression care in patients residing in the ALF environment.

Because we are unable to project the prevalence rate for Major or Minor Depression
in this setting given the lack of epidemiological studies of the ALF environment and
given the financial burden involved with SCHD-I/P administration to every ALF resident,
we will use a depression screening strategy for our first stage, with definitive diagnostic
testing for those assessed positive for depression. To screen for depression we will use a
15-item self-report instrument, the Primary Care Evaluation of Mental Disorders Patient
Health Questionnaire (PRIME-MD PHQ). Because the PRIME-MD PHQ has a
sensitivity of 75%, we will use a second screening measure (the HAM-DRS) to increase
our case finding abilities, as the HAM-DRS has a sensitivity of 88%. A two-phase

design has been recommended in community-based settings because the prevalence of

31

Major Depression is low (52). PRIME-MD PHQ scores will also be used to assess the
severity of depression in our ALF subjects. The Multilevel Assessment Instrument
(MAI) will be given to each subject during this first interview as well (50). The MAI
assesses quality of life and health status issues and is a “gold standar ” research tool for
geriatric psychiatry. The Mini Mental State Examination will also be administered during
the second interview, as a screening measure of cognitive impairment (51). The first

interview will occur within the first month that the subject is enrolled within the study.

Those patients who screen positive for depression using the PRIME-MD PHQ and
HAM-DRS will undergo a complete SCID-I/P assessment. This SCID-I/P assessment
will occur in the second month of enrollment and will serve as a reliability screen for the
initial PRIME-MD PHQ screening results. The SCID-I/P interview each subject
undergoes will also identify other co-morbid DSM-IV TR disorders such as anxiety,
psychosis and somatization. Subjects with a confirmed diagnosis of depression will be
notified of this finding and a letter with these study findings will be sent to their primary

care physician.

As a part of the third session, subjects will participate in a semi-structured depression
treatment interview. In the first two months of the project, the PI will develop and pilot a
treatment questionnaire based upon previous in-home health care depression prevalence
studies (28). This semi-structured interview will cover onset of treatment, treatment type,
frequency of administration, drug doses and duration, medication side effects. Therapy

questions will focus on type, frequency of intervention, mental health professional

32

educational background and the subject’s belief about efficacy and current quality of life.

Table VIII: Project Flow Chart

Interview #1: Interview #2: Interview #3:
Screening Diagnosis Tx Patterns

PHQ +
(depression)

*SCID-I/P ’Depression
Structured Treatment
Clinical Interview
Interview

Month 1 Month 3

 

Vb. Interviewers

We will use master’s trained mental health clinical specialists to recruit subjects,
screen them for depression and perform the structured SCID-I/P interviews on those
subjects meeting study criteria. The cost of this study prohibits psychiatrists or
psychologists from doing much of the direct subject interviews. Our mental health
specialists will have extensive training in dealing with mental health problems as part of

their graduate training.

Prior to the study we will provide intensive training in interpersonal skills and subject

recruitment as well as the assessment measures used in this study. Each mental health

33

clinical specialist will attain a SCID-I/P training workshop including training tapes and
supervised training interviews. Using both training tapes and live patient interviews, we
will assess inter-rater reliability of the SCID-I/P using a test-retest design until 95% inter-

rater agreement is obtained among raters.

Vc. Sampling Frame/Eligibility Criteria

The ALF is a unique living environment in that most people who live there are from
a higher socioeconomic class in order to be able to afford the costs associated with
residence. Major insurance carriers such as Blue Cross, Medicare or Medicaid do not
reimburse patients for ALF monthly rental. Expenses are usually paid out of pocket by
residents or supportive families. Because of this financial bias, the ALF resident is
usually a middle or upper class individual with retirement savings. This phenomenon

alone makes the ALF a unique environment to study depression.

Criteria for inclusion in this study are: (1) age 60 or older; (2) ability to speak, read
and understand English; (3) permanent resident of Independence Village or Porter Hills
Presbyterian Village; (4) ability to give informed consent (determined by the mental

health clinical specialist and the patient’s primary care physician).

Exclusion criteria include (1) presence of psychosis, dementia or other psychiatric

condition that prohibits the subject from understanding the study or giving consent.

These conditions may include coma, delirium, schizophrenia etc; (2) severe hearing or

34

visual disturbance that prevents comprehension of the assessment measures and

structured clinical interview.

Given the fact that dementia is present in up to 50% of the population at age 80,
cognitive disorders may limit the ability of subjects to participate (53). However, this
should be a less pressing issue in the ALF where residents must theoretically be
cognitively intact enough to provide their basic activities of daily living and managing
their personal affairs. To assess for the presence of dementia that may impair the
subject's ability to participate in the study, we will use the Mini Mental State Exam (51).
This 30-item screening instrument detects cognitive impairment in persons with an 8th
grade or higher education. The MMSE is sensitive (0.85) and specific (0.87) within this
age range and setting. We anticipate that 15% of potential subjects will be ineligible for

the study because of moderate to severe dementia (MMSE score <18).

35

Vd. The Study Sites

Two ALF facilities will be used for this study. Porter Hills Presbyterian Village is a
large continuum of care facility located in Grand Rapids, Michigan. On 34 acres of land,
it offers residential living, assisted living, dementia care and basic as well as skilled
nursing care. Porter Hills owns six other assisted living facilities within the greater
Grand Rapids area. There are over 800 residents on the main campus and 175 living
within in the assisted living portion of the residence. The average resident age is 81.1
years old (SD=3.9), Caucasian (97.5%), Female (77%) and Presbyterian (50%). The
length of stay in assisted living is 20.1 months with residents discharged to both home as

well as to the nursing center.

Independence Village is a 200 resident assisted living facility in Lansing Michigan.
Lansing and Grand Rapids, Michigan are 55 miles apart. The geographic areas for
Independence Village and Porter Hills Presbyterian Village are similar in that both are
suburban, middle-class, white-collar neighborhoods. The average resident at
Independence Village is 78.9 years old (SD=6.3), Caucasian (95%) and Female (70%).

Religious preference is unknown.

36

Ve. Sampling Procedures

Residents of ALF Independence Village and Porter Hills Presbyterian Village will be
invited to participate in this study. The ALF Independence Village and Porter Hills’
admissions office will provide a roster of potential participants. An informational flyer
will be delivered to each resident and orientation coffees will be set up to help recruit
subjects. The administration of both ALF s will also provide a letter of support for the
study that each resident will receive. Enrollment activities will occur at convenient times
for the residents (i.e., mealtimes, early evening) to facilitate exposure to the program.
The enrollment phase of this study will occur over the first year of the project, in order to
provide adequate time for the mental health clinical specialists to contact all potential
subjects residing in each ALF. Residents who have not consented to the program will be

individually contacted and encouraged to participate by the clinical psychiatric specialist.

We anticipate that 85% of residents contacted will be eligible for the study. During
the first interview we will screen potential subjects for dementia using the Mini Mental
State Examination (51). We anticipate that 15% will be disqualified due to cognitive
disorders that prohibit them from understanding the nature of the study and giving clear
consent. In addition, we expect that up to 20% of all subjects will refuse to participate in
the study for other reasons. Table IX delineates the sampling strategy and estimated

sample size for this study.

37

To evaluate the possibility of false negatives, we will use the SCID-I/P interview with
10% of the sample without depression using the PRIME-MD PHQ. This will help us
evaluate false negatives as well as bolster validation of this instrument as a depression

screener for the ALF setting.

Table IX: Sampling Strategy for ALF Depression Study

 

 

 

 

 

 

 

 

 

 

 

 

 

Facility # of ALI # # with Minor # with Majo Total # over a
resident: with Depression Depression 2 year period
dementia (11%) (14%)
(15%)
Independence 200 30 19 24 86
Village
Porter Hills 175 25 17 21 76
Presbyterian
Village
Totals 375 55 36 45 162
Vf. Recruitment and Consent

Before the study begins we will distribute information about the project via several
avenues. First, each ALF resident will receive an informational flyer about the study.
The flyer will encourage participation as well as ny to provide facts about the nature and
scope of the study. Each ALF resident will be invited to attend an informational coffee
held in the ALF lounge. Several sessions will be offered to assure that ALF residents
may select from convenient times throughout the day. A member of the research team

will conduct the coffee and will offer time for questions and answers during this meeting.

38

The mental health clinical specialist will then telephone each ALF resident to invite
them to participate and Sign a consent form. The consent form, approved by the MSU
Institutional Review Board, will be provided during an initial appointment made with a
member of the research team. During this appointment, we will confirm eligibility for
the study and review the consent form, answering questions about its content and

obtaining appropriate Signatures.

The expected demographics of this sample are listed on Table X. Seventy four percent of
individuals age 65+ are female while 26% are male. Fourteen percent of the sample is
expected to be black, 5.5% Hispanic and 1.8% “Other” with the remaining 78.7%

Caucasian. (40)

Table X: Expected Key Sample Demographics

 

 

 

 

 

 

Black White Hispanic Other Total
Male 6 33 2 1 42
Female 17 94 7 2 120
Total 23 130 9 3 162

 

 

 

 

 

 

 

To evaluate differences in responders and non-responders, we will use demographic
information obtained from the ALF administration such as age, gender and date of
admission to compare individuals to decline to participate in the study with the
demographics of those who agree to participate. This will inform our understanding of
potential differences in our sample and the overall ALF population as well as help

delineate any type of volunteer bias present within our sample.

39

Vg. INTERVIEW #1: Depression Screening and Functional Assessment

Interview #1 will serve as a depression-screening interview. The mental health clinical
specialist will schedule this interview via telephone. A convenient time will be arranged
with the subject and the specialist will come to the subject’s room in the ALF campus.

During this visit the study consent will be explained, signed and witnessed.

Subjects will then participate in a brief interview in which the PRIME-MD PHQ, MAI,

MMSE and clinical features checklist will be completed:

i. Demogpaphic Profile will include age, gender, race, marital status, educational level

and socioeconomic status (SE8). SE8 will be determined using the Multilevel

Assessment Instrument (MAI) (50).

ii. The Prima_ry Care Evaluation of Mental Disorders--Patient Health Qpestionnaire

 

(PRIME-MD PHQ) will serve as our depression-screening instrument. Its screens for the
presence of depression and provides a severity measure as well. The PRIME-MD PHQ is
a 15-item patient-administered questionnaire that will reduce the clinician time needed to
complete the screening interview. When compared to “gold standard” psychiatric
assessments, the PHQ shows good reliability (kappa =0.65, overall accuracy 85%,
sensitivity 75%, specificity 90%) for elderly patients. The average time required to
complete the assessment is 10 minutes for patients and 3 minutes for clinician review

(47).

40

 

 

iii.

ser

1hr

0°.

iii. The Hamilton Depression Rating Scale (HAM-DRS) will serve as an addition
sensitivity measure for case identification. This will complement the low sensitivity of
the PRIME-MD PHQ as the HAM-DRS has a sensitivity of 88%, specificity of 90% and
an overall accuracy rate of 83% (54). The HAM-DRS is a 22-item measure of the
severity of depression used commonly in geriatrics and clinical trials literature (refs).
The HAM-D assesses both neurovegatative items (sleep, appetite, weight loss, energy

level, etc) as well as mood symptoms (severity of depression, apathy, irritability) (49).

iv. _M_ultilevel Assessment Instrument (MA!) is a 14-item clinician-rated instrument we
will use to assess functional disability. The MAI consists of several subscales including
the self-maintenance subscale, the instrumental activities of daily living subscale and a
global disability score. Unlike an instrument such as the SF-36 that has floor effect with
respect to functional disability, the MAI is sensitive to changes in functional level in an
already disabled population. The MAI has excellent inter-rater reliability (>.91 intra-
class correlations) in its assessment of functional disability and social demographics such
as SES. The MAI has been validated in long-term and home-care settings using older

adults (50).

iv. Mini Mental State Examination (MMSE) will assess cognition. This is a 30-item

screening tool evaluates orientation, attention, short-term memory and higher cortical
functions (visual spatial ability, sequencing, language). The MMSE is considered the
gold standard tool for dementia screening; a score of >24 suggests cognitive dysfunction.

The MMSE is reliable and valid for English speaking subjects with at least an 8th grade

41

 

education. For purposes of this study we will use a cut-off of <18 to eliminate subjects
with moderate to severe dementia, as their cognitive impairment may preclude them from
accurately completing the SCID-I/P and other self-report measures of mood or

functioning (5 1).

Vh. INTERVIEW #2: Definitive Diagnosis (SCID-I/P)

Interview #2 will be entirely devoted to completing the Structured Clinical Interview for
DSM-IV TR (SCID-I/P). Eligible subjects who have scored positive for Major or Minor
depressive disorder using the PRIME-MD PHQ will be eligible for the second diagnostic

interview.

i. Diagostic Interview (SCID-I/P). The SCID-I/P incorporates operational criteria
from the DSM-IV TR into a categorical system for rating symptoms and diagnosing
Major and Minor Depression. Ventura found a sensitivity of 87% and specificity of 81%
for 20 case vignettes (55). The SCID-I/P has been validated for use in elderly
populations. We have found that the SCID-I/P is the only validated diagnostic instrument
used in long-term care (56). We anticipate that the SCID-I/P may have significant
respondent burden in the frail ALF resident as completion may require several hours. We
therefore may be required to split Interview #2 into two one-hour long sessions to

increase compliance and prevent fatigue.

42

To accurately assess the prevalence of Major and Minor Depression while also
capturing other comorbid diagnostic groups such as anxiety, somatoform and psychotic
disorders, we will administer the complete SCID-I/P to each subject. This will help us to
accurately identify depressed patients and will facilitate the identification of comorbid
conditions (e.g., Anxiety disorders, Somatoforrn disorders) as well as other types of

depression (Dysthymic disorder, Adj ustrnent disorder).

Vi. INTERVIEW #3: Depression Treatment

The third interview will focus upon current and past depression treatment experiences of
the ALF resident. A member of the research team will conduct a 1-hour session with
those subjects determined to have Major or Minor Depression via the SCID-I/P during
interview #2. Based upon our experiences during our pilot work, information about the
nature of current and past depression treatment may be sketchy. The ALF facility is not
required by state or federal law to maintain extensive medical records. Thus, a
medication list may not be readily available if medications are self-administered by the
ALF-resident. In addition, some ALF residents may be unaware of the types of
medications they take or their doses. Thus, relying simply on ALF chart information or a

sole ALF resident interview may provide incomplete information.

Instead of relying upon chart review in a system beset with inadequate medical records,

we will directly interview the ALF resident to obtain treatment information. Interview #3

43

will be completed with the ALF resident as well as family member to corroborate
information given by the resident. Requesting that family members join this interview
will help these individuals offer additional data about the subject’s treatment, provide
memory prompts for the subject and serve as a secondary source to validate the specific
details of information given by the subject. Interview #3 will consist of a semi-structured

interview focusing upon the following domains of mental health treatment:

1. Current Treatment for Dgpression: Data obtained in this section will include:
psychotropic medications, doses and frequency of administration, date of onset of
treatment, monitoring physician, medication side effects, self—report compliance with
psychotropic medications. In addition, formal mental health professional contact will be
elicited including specialty, frequency of contact, perceived effectiveness and potential
barriers to further treatment. Other types of potential ongoing mental health interventions
will be covered as well including exposure to group or marital therapy, activity therapy,

spiritual counseling, psychiatric hospitalizations and electroconvulsive therapy.

2. Past Treatment for De_pression: Similar domains for past depression treatment will
next be elicited from each subject including previous history of pharmacotherapy,

psychotherapy or other psychosocial therapies.

3. Mental Healthcare Utilization: To obtain this information we will ask each subject to
estimate the following activities over the past year: visits to primary care physicians,

visits to mental health professionals, ER visits, number of laboratory tests, number of

prescription medications used over the past one year and frequency of ancillary medical

services.

4. Expert Consensus Guidelines on Pharmacotherapy of Depressive Disorders in Older

 

Patients: In 2001, an nationally known panel of experts on the treatment of late life
depression developed pharmacotherapy practice guidelines for the treatment of depressed
elderly. Using literature review and a Delphi-style process, they surveyed 50 national
experts on geriatric psychiatry and obtained consensus on 64 questions related to
pharmacotherapy principles with the elderly depressed patient. The group achieved
consensus on 89% of items and developed “Adequate Dose of Antidepressant Guidelines
for the Suggested Treatment of Depression in the Elderly”. These guidelines include
recommendations for (1) average starting antidepressant dose; (2) average target
antidepressant dose afier 6 weeks of treatment and (3) usual highest final acute

antidepressant dose (45).

Using data obtained from Interview #3, we will compare and contrast the mean
starting dose, average target dose after 6 weeks and usual highest final acute dose for
ALF residents with the Expert Consensus Guideline values. Using descriptive statistics,
we will detail the adequacy the drug treatment for ALF residents when compared to a
nationally developed gold standard. This will also facilitate exploration of the
relationship between a variety of factors (e.g., demographics, diagnostic groups,

functional status) and treatment adequacy in the ALF setting. These data will help us

45

describe the level of “unmet clinical need” for treatment for depressed ALF patients,

critical information that is currently lacking in clinical practice and the literature.

VI. Data Analysis

The specific aims of this project include both descriptive and comparative analysis.
This is a two-year cross-sectional study of two Michigan ALF s. It is cross-sectional in
nature in that we will use descriptive statistics to capture the prevalence of both Major
and Minor Depression in the ALF. This will enable us to describe the clinical and
functional characteristics of a sample of ALF residents. We hypothesize that the
prevalence of Major and Minor Depression will be higher than a home-care based
sample, i.e. that Major Depression will exceed 13.5% and Minor Depression will exceed

10.5% (28).

This study is also comparative in nature in that we will collect data about depression
treatment patterns (antidepressant use, psychotherapy, etc) that we will be able to contrast
with national practice standards using the Depression Expert Consensus Guidelines on
Pharmacothegpy of Depressive Disorders in Older Patients (45). We hypothesize that
antidepressant usage and dosing ranges will fall below recommended national practice

standards for the ALF subjects in this study.

We will use a number of categorical and continuous variables in this database.

Summary statistics will be reported as proportions for categorical variables, and means or

46

medians for continuous measures. Categorical variables will be analyzed using x2 tests,
t-tests and correlational methods will be used to assess continuous variables. The
outcome measures (PRIME-MD PHQ, HAM-DRS, MAI) will be analyzed as continuous
variables in the dataset, based upon the assumption of a normally distributed response.
During these analyses, if this assumption is incorrect, transformations of the measures by
the logarithmic or square root will be employed to achieve a normally distributed
response. Subjects who do not complete the study will be dropped from the analysis and

not included in our results.

All statistical analysis will be completed using the Statistical Analfiis Sfitem (57). A
significance level of 5% will be employed on all tests of hypothesis on regression

parameters.

VIa. Primag Aim #1: Determine the prevalence of Major and Minor Depression
in the ALF.

We will use descriptive analyses to determine the prevalence of Major and Minor
Depression, once subjects are defined using the SCID-I/P. In order to accurately depict
this proportion, thorough screening and case identification must occur. In addition, the
calculation of confidence intervals and sample size will help guide subject recruitment

and assure adequate power for our analyses.

47

 

‘1.

 

i. Improving Case Identificetion: The u_se of two screening measures

Although the PRIME-MD PHQ is an excellent mechanism to delineate the
difference between major and minor depression and provide a measure of depression
severity, its sensitivity is low (75%). Using this instrument alone as our depression
screener may result in up to 25% of all potential cases being missed. To address this
concern and assure we will not omit potential cases in our study, we will introduce a
second screening instrument, the Hamilton Depression Rating Scale (HAM-DRS).
The HAM-DRS has a sensitivity of 88% for major depression and should assist in
identifying cases that the PRIME-MD could miss. It has been validated for use in the
home care population and will provide adequate specificity 90% and overall accuracy

(83%). (28,54).

ii. The Calculation of Confidence Intervals and Sample Size

To project the 95% confidence intervals we will use Major and Minor Depression
prevalence figures from Bruce’s homecare study (28), as this population of elderly living
in their homes is likely to most closely match our ALF study sample. To determine the
upper and lower confidence interval we will employ the “Exact Two-sided 100% x (1-
or) Confidence Limits for Binomial Proportions (or =.05) Table” (58). Using Bruce’s
homecare data, we estimate prevalence (p) of Major Depression at 13.5% and Minor
Depression at 10.5% (28). Given the binomial formula npq, where n=anticipated
number of subjects enrolled, p=prevalence of subjects with depression and q=prevalence
of subjects who are depression-free, npq=l62 (22/162)(140/162). The “Exact Two-Sided

100% x (l-or) Confidence Limits for Binomial Proportions (or=.05) Table” suggests a

48

 

95% confidence interval of (007,019) for Major Depression and (0.06, 0.16) for Minor

Depression in this ALF study.

We have also calculated the minimum sample size based upon a range of prevalence
figures for our study. We will use an estimate for the prevalence of major depression
based upon home healthcare and nursing home figures (27, 4). pr=sarnple estimate of P
(prevalence of major depression), e=desired error limit (% above or below true
prevalence), Z=two tailed critical normal value, n'=sarnple estimate and n=fina1 sample
size, then we may use the formula n'= p(1-p)(Z)2 divided by e2 to estimate our sample
size for a given prevalence estimate. For example, if we estimate that 20% of our sample
will have major depression, then p=20% or 0.2 and (1-p)=80% or 0.8, and if e=0.5 so that
Z=1.96, then n'=(0.2)(0.8)(1.96)2 divided by (0.05)2 then n' = 245 ALF residents required
for our sample. Using this power calculation, Table XI describes sample size needed for

a range of major depression prevalence figures.

Table XI: Sample Size Determinations based upon Depression Prevalence

 

 

 

 

 

 

Estimated Prevalence of Maj or Sample Size Required
Depression in Assisted Living

10% 138

12% 162

13% 173

15% 195

20% 245

 

 

 

 

49

 

Based upon the work of Bruce in the home healthcare setting, we anticipate that
the prevalence of depression in an ALF may be similar, i.e. 12-13% (28). Thus, our
current sample size of 162 should be adequate to assess for major depression in this
sample. If necessary, we will recruit other local facilities to bolster our sample if

recruitment proves inadequate for these calculations.

VIb. Primayy Aim #2:. Describe the clinical and functional status of patients with

Major and Minor Depression in the ALF setting.

In this project we will use several functional and disease measures as clinical
outcome indicators. The MAI will be used to measure functional status in our ALF
sample. The PRIME-MD PHQ serves as a measure of depression severity in this study.
To correlate these clinical measures with depression type, depression treatment modality
and adequacy of depression treatment we will employ logistic regression, multinomial

logit models and least squares regression.

All models will include sociodemographic elements including age, gender, marital
status, race and socioeconomic status (as determined by the MAI). These data will be
collected during Interview #1. Models will be constructed to examine the impact of each
sociodemographic/economic element upon type of depression (Major vs. Minor) as well

as clinical features (MAI, PRIME MD PHQ).

50

 

Vlc. Primagy Aim #3: Describe the adequacy of current depression treatment as

compared to “gold standard” depression treatment guidelines.

We will use the Expert Consensus Guideline values as the “gold standard” for this
ALF study. Mean antidepressant dose and duration of treatment will be calculated for
each ALF resident and summary scores will be obtained for each antidepressant type.
Using these guidelines we will compare ALF residents with the gold standard
recommendations, in terms of average starting dose, average target dose after 6 weeks of
treatment and usual highest final acute dose. This analysis will also provide an
opportunity to examine the relationship between key demographic, diagnostic and
functional factors in the ALF sample with adequacy of treatment (dose and duration of

66

treatment). Considering subjects as either antidepressant “therapeutic” or sub
therapeutic”, we will use logistic regression or multinomial logit models to examine each
group’s relationship to key clinical factors (MAI and PRIME MD PHQ score) as
appropriate. This analysis will help us to describe the level of “unmet clinical need” in

the ALF, information that is lacking in the literature and in clinical practice.

VId. Secondagy Aim #1: Identify potential barriers to care that make the ALF a

unique setting to provide depression treatment.

This is a descriptive aim. As the study begins in each ALF, the master's trained
mental health clinical specialists will maintain a journal of observations about the

implementation of methods. Using this journal, the master's trained psychiatric

51

specialist will construct a list of barriers encountered during subject recruitment,
interview process, collateral contacts and staff interactions. In addition, we will ask ALF
residents who did not consent for study participation to describe the reasons for their

refusal.

VIe. Secondagy Aim #2: Establish the level of unmet clinical services need in

depressed ALF residents.

During Interview #3 we will collect data on the frequency of formal mental health
intervention (individual/family/marital therapy sessions, psychiatric evaluation,
neuropsychological testing, etc) in ALF subjects with Major or Minor Depression.
Descriptive analysis will include frequencies of treatment, treatment type and mean
treatment length. Ninety-five percent confidence intervals will be determined for each
proportion. We will use logistic regression to compare residents receiving formal mental
health services to subjects who remain untreated, examining key variables such as race,

gender, age, functional status and severity of depression.

52

VII. Implications of this Study for the Treatment of Depression in ALFS

Assisted living facilities are the vanguard of future service settings for older adults.
We need to know the prevalence of key psychiatric conditions in these facilities to ensure

that patients have access to and receive effective treatments for depression.

This proposal to study the prevalence of depression will inform our understanding of
the problem of depression in the ALF setting. This study will focus on research questions
yet unanswered including: (1) the prevalence of Major and Minor Depression; (2) the
clinical and functional characteristics of depressed ALF residents; (3) how "depression
care as usual" is provided for ALF residents and (4) the degree of adherence of current
treatment to national depression practice standards. In addition, we will explore unique
potential barriers to the delivery of mental health services in the ALF setting, delineating

barriers on the part of ALF staff, residents, families and healthcare providers.

Why is determining the prevalence of Major and Minor Depression an important first
step in this research trajectory? Although we already know that depression is devastating
in late life, the ways that this disease uniquely manifests itself in an ALF are unstudied.
Establishing the baseline prevalence for Major and Minor Depression in an ALF will lend
credence to the clinical observation that this condition occurs frequently in an ALF. It
will also help us focus future research on identifying ALF residents at risk for depression
as well as how to best provide treatment for those individuals already experiencing

depression.

53

Our data on the functional status of ALF residents will also be useful for public health
policy and licensing. If our data suggests that our ALF subjects are as frail as the
typical nursing home patient, this finding alone may bolster the argument that ALFS need
better state and federal regulation in order to assure that adequate care is being provided
for these residents. Developing a "profile" of the typical clinical and firnctional features
of patients who are depressed may also help clinicians to target new ALF residents at risk

for depression during their stay in the facility.

The results of this study are critical to inform us about mental health resources and
treatment in the ALF. If we determine that depression is problematic in the ALF, we
must also learn something about how depression "care as usual" is delivered, in order to
find key elements of current treatment that require improvement. If we can study ALF
patients currently treated for depression, we will be able to evaluate whether current
antidepressant treatment received by these individuals meets national depression practice
guideline standards (Pharmacotherapy of Depree_sive Disorders in Older Patients,
Alexpololous, 2001). These data will then help us design interventions that specifically
deal with the mental health treatment delivery issues we identify during this study. To
develop an intervention without knowing the current availability of treatment options or
the receptivity of the ALF environment to mental health services could doom an

intervention to failure.

This is also a study of barriers. Can effective mental health care be delivered to this

segment of our population? Will current staffing patterns and education allow for the

54

 

expedient identification of a depressed ALF resident? How do ALF residents perceive
the adequacy of their depression care? This proposal will allow us to identify key
obstacles in subject recruitment, data collection and baseline mental health service
delivery so that we can set the stage to answer a number of important questions. This
study will also help us prepare to outline strategies for interventional research focused

upon improving access to mental health care and depression treatment.

Given that the ALF industry is growing and may house elderly residents who in
earlier years would have been placed in a nursing home, the federal government may be
interested in the results of this study. In 1987, Congress passed the OBRA-87 Act to
regulate the misuse of psychotropic medications in nursing homes and provide a
mechanism for psychiatric assessment and mental health service delivery (37). This
program has demonstrated effectiveness with the reduction of chemical and physical
restraints (59). If data from our ALF study suggests either polyphannacy or inadequate
dosing of antidepressant medications, this may focus more federal attention upon
implementing psychotropic regulatory mechanisms such as OBRA for the ALF

environment.

The second potential policy impact is on the relationship between the ALF industry
and the community public mental health sector. Providing care for the mentally ill with
minimal insurance resources, state-funded community mental health (CMH) centers are
charged with providing care for all ages. Often in a CMH center, staffing resources are

severely stretched, allowing staff to only provide services for nursing home or

55

ambulatory patients. ALF residents may not be receiving in-house CMH services. If this
study finds these counseling services are not easily available for ALF residents, it may
convince CMH to allocate their staff resources toward services the ALF instead of only
the nursing home. Redirecting clinician manpower fiom the nursing home to the assisted
living environment could impact the quality of life for the average ALF resident for a

longer period of time than if the resident lived in a nursing home.

56

 

REFERENCES

57

 

References
www.alfa.org ”Facts about Assisted Living” website page

Colenda CC, Bartels SC, Gottlieb GL: Chapter 127. The North American
System of Care. In Copeland J, Abou-Saleh M, Blazer D. Principles and
Pr_actice of Geriatric Psychim 2"d Edition. Wiley and Son, Ltd. London.
2002, pp.689-696.

Callahan CM, Hui SL, Nienaber, NA, Musick BS, Tierney WM,.
Longitudinal study of depression and health services use among elderly
primary care patients. J Am Geriatrics Sociey 1997;42:833-838.

Katz IR, Pannelee PA. Depression in elderly patients in residential care
settings. In Schneider LS, Reynolds CF [[1, Lebowitz BD, Frieidhoff AJ, eds.

Diagposis and Treatment of Depression in Late Life. Washington DC:
American Psychiatric Press, 1994:437

Katz IR, Lesher E, Kleban M. Clinical features of depression in the nursing
home. Int Psychogeriatr 1:5, 1989.

Rovner BW, German PS, Brant LJ. Depression and mortality in nursing
homes. JAMA 1991;265:993-99.

Lebowitz BD. Diagnosis and treatment of depression in late life: an overview
of the NIH consensus statement. Am J Geriatr Psychiag 1996;4(suppl 1):S3-
6.

Callahan CM, Hendrie HC, Dittus RS, Brater Dc, Hui S1. Depression in late
life: The use of clinical characteristics to focus screening efforts. Journal of
Gerontology 1994;49(1):M9-14.

Fischer LF, Wei F, Rolnick SJ, Jackson JM, Rush WA, Garrad JM, Ni NM,
Luepke LJ. Geriatric depression, antidepressant treatment and healthcare
utilization in a health maintenance organization. J Am Geriatric Socieg
2002;50(2):307-12.

58

 

10.

ll.

12.

13.

14.

15.

16.

17.

18.

Broadhead EW, Blazer DG, George LK. Depression, disability days and days
lost from work in a prospective epidemiological survey. JAMA
1990;264:2524-8.

Ashby D, Ames D, West CR. Psychiatric morbidity as a predictor of
mortality for residents of local authority homes for the elderly. Int J Geriatr
Psychiam 1991 ;6:567-73.

Parmalee PA, Katz IR, Lawton Mp. Depression and mortality among
institutionalized aged. J Gerontol 1992;47:P3-5.

Callahan Cm, Wolinsky FD, Stump TE, Nienaber NA, Nui SL, Tierney WM.
Mortality, symptoms and functional impairment in late-life depression. 1
General Internal Medicine 1998;13(11):746-52.

Schoevers RA, Geerlings MI, Beekrnan AT, Penninx BW, Deeg DJ, J onker C,
Van Tilburg W. Association of depression and gender with mortality in old
age. Results from the Amsterdam Study of the Elderly (AMSTEL). British
Journal of Psychiatry 2000; 177:336-42.

Schleifer SJ, Keller SE, Bond RN. Major depressive disorder and immunity.
Arch Gen Psychiag 1989;46:81-87.

Thomas DR, Zdrowski CD, Wilson MM, Conright KC, Lewis C, Tariq S,
Morley JE. Malnutrition in subacute care. Am J Clin Nutr 2002;75(2):308-
13.

Vaccarino V, Kasi SV, Abramson J, Krumholz HM. Depressive symptoms
and risk of functional decline and death in patients with heart failure. J Am
College of Caldiolpgy 2001 ;38(1): 199-205.

 

Blazer DG, Bachar JR, Manton KG. The epidemiology of depression in an
elderly community population. Gerontologist 1987;27:281-7.

59

 

19.

20.

21.

22.

23.

24.

25.

26.

27.

Blazer DG, Bahar JR,Hughes DC: Major Depression with melancholia: A
comparison of middle aged and elderly adults. Journal of the American
Geriatric Sociegy 35:927-932, 1987

Kessler RC, McGonagle KA, Swartz M, Blazer DG, Nelson CB. Sex and
depression in the National Comorbidity Survey. 1: Lifetime prevalence,
chronicity and recurrence. J Affect Disord 1993;29(2-3):85-96.

Narrow WE, Reiger DA, Norquist G, Rae DS, Kennedy C, Arons B. Mental
health service use by Americans with severe mental illness. Soc Psychiap'y

Psychiag Epidemiology 2000;35(4): 147-55.

Weissman MM, Wickrarnaratne P, Merikangas KR, Leckman JF, Prusoff BA,
Caruso KA, Kidd KK, Gammon GD. Onset of major depression in early
adulthood. Increased familial loading and specificity. Arch Gen Psychiagy
1984;41(12):1136-43.

Kukull WA, Koepsell T K, Inui TS, Borson S, Okimoto J, Rasking JL.
Depression and physical illness among elderly general medical clinic patients.
J Affect Disorder 1986;10:153-162.

Turrina C, Caruso R, Este R, Lucchi F, Faxxari G, Dewey ME, Errnentini A.
Affective disorders among elderly general practice patients. Br J Psychiag
1994;165:533-537.

Oxman TE, Barrett J E, Barrett J, Gerber P. Symptomatology of late-life
Minor Depression among primary care patients. Psychosomatics
1990;31(2):174-180.

Olfson M, Shea S, Feder A, Fuentes M, Nomura Y, Gameroff M, Weissman
MM. Prevalence of anxiety, depression, and substance use disorders in an
urban general medical practice. Arch Fam Med 2000;9(9):876-83.

Bruce ML, McNamara R. Psychiatric status among the homebound elderly:
an epidemiologic perspective. J Am Geriatr Soc 1992;40(6):561-6.

60

 

28.

29.

30.

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Bruce ML, McAvay GJ, Raue P], Brown EL, Meyers BS, Keohane DJ,
J agoda DR, Weber C. Major depression in elderly home health care patients.
Am J Psychiatly 2002;159(8):1367-74.

www.agingstats.gov
Hawes C, Rose M, Phillips OS. A national study of assisted living for the

frail elderly. Executive summary: Results of a National survey of facilities,
hhtp://§pe.os.dhhs.gov/deltcp/reports/facreses.htrn.

Sullivan JG. Redefining long-term care. Contemmpag Long Term Care,
l998;August:60-64.

www.aspe.hhs.gov

www.hhs.gov

www.carescout.com/resources

www.cdc. gov

Jones A. The National Nursing Home Survey: 1999 summary. Vital Health
M 2002;Jun (152):1-116

Kinney V. OBRA nursing home rules target aides, mentally ill. Health Prog
1989;70(4):19-61.

McGannon A. F arnilies in distress: Making a long-term placement for
someone with Alzheimer’s disease. American Journal of Alzheimer’s Care
and Related Disorders & Research 1993; Jan/Feb:2-5.

Chapin R, Dobbs-Keeper D. Aging in place in assisted living: philosophy
versus policy. Gerontologi st 2001 ;41(1):43-50.

www.census. gov

61

 

 

41.

42.

43.

45.

46.

47.

48.

49.

50.

www.michigan. gov

www.nchs.gov

McCuskor J, Cole M, Keller E. Effectiveness of treatments of depression in
older ambulatory patients. Arch Intern Medicine 1998;158:705-712.

NIH Consensus Development Panel on Depression in Later Life. Diagnosis
and treatment of depression in late life. JAMA 1997;278:1186-1190.

Alexopoulos GS. Katz IR, Reynolds CF, Carpenter D, Docherty JP.
Pharmacotherapy of Depressive Disorders in Older Patients. Post uate
Medicine 2001;Special Reportzl-86

Colenda CC, Wagenaar DB, Mickus M, Marcus SC, Tanielian T, Pincus HA:
Comparing clinical practice to guideline recommendations for the treatment of
depressed geriatric patients: Findings from the American Psychiatric Practice

Research Network. American Journal of Geriatric Psychiapy, in press.

Spitzer RL, Kroenke K, Williams JBW. Validation and utility of a self-report
version of the PRIME-MD: The PHQ primary care study. JAMA
1999;282(18):1737-1744.

Spitzer RL, Williams JB, Gibbon M, First MB. The Structured Clinical
Interview for DSM IV: History, rationale and description. Arch Gen
Psychiat_ry 1999;49(8):624-9.

Hamilton M. Development of a rating scale for primary depressive illness.
Br J Soc Psychol 1967;6(4):278-96.

Lawton MP, Moss M, Fulcomer M, Kleban MH. A research and service
oriented Multilevel Assessment Instrument. Journal of Gerontology
1982;37(1):91-99.

62

51.

52.

53.

54.

55.

56.

57.

58.

59.

Folstein MF, Robins LN, Helzer JE. The Mini-Mental State Examination.
Arch Gen PsvchiMl983; 40(7): 812.

Shrout PE. Measurement reliability and agreement in psychiatry. M

Methods Med Res 1998;7(3):301-17.

Herbert LE, Beckett LA, Scherr PA, Evans DA. Annual incidence of
Alzheimer disease in the United States projected to the years 2000 through
2050. Alzheimer Dis Assoc Disord 2001;15(4): 169-73.

 

Mulsant Bh, Ganguli M. Epidemiology and diagnosis of depression in late
life. J Clin Psychiat_ry 1999;60(suppl 20):9-15.

Ventura J, Liberman RP, Green MF, Shaner A, Mintz J. Training and quality
assurance with the Structured Clinical Interview for DSM-IV TR (SCID-I/P-
I/P). Psvchietry Res 1998;79(2):163-73.

Rosen J, Rogers JC, Marien RS, Mulsant BH, Shahar A, Reynolds CF.
Control-relevant intervention in the treatment of minor and Major Depression
in a long-term care facility. American J Geriatric Psychiafl 1997;5(3):247-
257.

SAS Institute, Cary, NC.

Rosner B. Fundamentals of biostatistics. Wadsworth Publishing Co, Belmont,
CA, 1995.

Llorente MD, Olsen EJ, Leyva O, Silverman MA, Lewis J E, Rivero J. Use of
antipsychotic drugs in nursing homes: Current compliance with OBRA
regulations. J AGS 1998;46(2):198-201 .

 

63

 

APPENDICES

64

APPENDIX A

RESEARCH CONSENT FORM

DEPRESSION IN ASSISTED LIVING

65

 

APPENDIX A

RESEARCH CONSENT FORM

DEPRESSION IN ASSISTED LIVING

 

Introduction

Depression is a common problem for many assisted living residents. The goal of this

 

research is to understand how common depression is in the assisted living facility and
understand what types of mental health care are currently being provided for the assisted

living resident.

Study Overview

The purpose of this study is to understand how often depression occurs in the assisted
living facility and to help us understand how it may be treated. Faculty members from
Michigan State University’s Department of Psychiatry will be conducting the study.
Members of this research team include MSU faculty and master's trained mental health

clinical specialists experienced in working with older adults and their families.

If you agree to participate in this depression study, a member of the research team will

interview you three times during the next 3 months. During the first interview, the

66

master's trained mental health clinical specialists will ask you questions from a number of
standard screening instruments (the PRIME MD PHQ, the Hamilton Depression Rating
Scale, the Mini Mental State Examination and the Multilevel Assessment Inventory).
This interview will take about one hour to complete. You will be notified of the results
of testing at the conclusion of the interview. If you have evidence of poor memory, we

will refer you to your primary care physician for further testing.

If you quality for the study, you will then participate in a SCID interview designed to
help the research team understand how your mood is doing. This interview may take up
to 2 hours to complete. We will complete the interview at your own pace, taking breaks
as needed throughout testing. Finally, a third brief interview will follow in about one
month to learn more about any types of depression treatment you may be receiving
(counseling, medications, activity therapy, etc). This interview will last about 30
minutes. If you are unable to remember the names of your medications or medical
conditions, we will ask the assisted living staff to provide us with a list of these items.
By signing this consent you agree that the assisted living staff can provide this medical

information for our study.

If your testing indicates the presence of depression, you may be offered a number of
treatments including medication, therapy or therapeutic activities. Specific treatment
options and decisions should be discussed with you and your physician. The research
team will offer one free evaluation session at the MSU Psychiatry Clinic with a trained

mental health professional who may help triage your case to appropriate mental health

67

 

professionals in the community. Treatment beyond this session will be the responsibility
of the study participant and may vary depending upon community and financial

resources.

Risks

This project should present minimal risk to you. Possible risks include increased anxiety
or depression that may occur as you are interviewed by the research team. Members of
the research team recognize the sensitive nature of depression discussions. They will
insure that the interview setting is private and confidential. If you were to become
anxious or upset during the interview, the research team member will immediately stop
the interview, contact the assisted living social worker and additional support will be
provided for you. If you feel suicidal or homicidal during the interview, the assisted
living social worker will work with you to help you with these feelings, assure your

safety and help you obtain additional mental health help.

The risk of confidentiality will be minimized by never identifying you by name. All

study data will be kept confidential, kept in a locked secured area within the research

office. Study results will never identify a patient, family, guardian or facility by name.

68

 

 

Benefits

You may directly benefit from this study in that if depression is identified, you and your
doctor along with the assisted living staff will be notified. You may be offered a variety

of treatments to help with this condition.

 

T
l}
'4

Costs

There will be no cost to you or your family.

Questions

If you have further questions, you may contact Dr Deborah Wagenaar (517-353-4362) or
her master’s trained mental health clinical specialists. For questions relating to your
rights as a study participant, please call Dr Ashir Kumar, MD, Chairperson of the

University Committee on Research Involving Human Subjects at 517-355-2180.

69

Consent

I have read this consent form and my questions are answered thus far. I voluntarily
consent to participate in this study with the understanding that I may withdraw my

consent at any time.

 

Resident Signature Date

 

Witness Signature Date

70

 

 

APPENDIX B

MINI MENTAL STATE EXAMINATION

71

 

9:9"

APPENDIX B

MINI MENTAL STATE EXAMINATION

ORIENTATION

1. What is the: season date year month day

 

2. What is the name of this:

city state county place floor of this building

REGISTRATION

3. Name three objects, "apple-table-penny". Score one point for each correct on FIRST

TRIAL.

ATTENTION AND CALCULATION

4a. Ask the patient to subtract 7 from 100 and then continue subtracting 7 until told to

stop. Stop after 5 subtractions. Score one point for each correct subtraction.

72

4b. Spell "world" backwards. Record number of correct letters. DO NOT include this in

total score.

RECALL

5. Recall 3 objects repeated earlier. Score one point for each.

LANGUAGE

6. Point to objects one at a time. Score one point for each object correctly named.

7. Repeat "no ifs and or buts". Allow only one trial. Score one point for EXACT

repetition.

8. Three state command: "Take this piece of paper in your right hand, fold it in half,

and give it back to me". Score one point for each stage.

9. Read and obey "Close your eyes". Score one point if he closes his eyes.

10. Ask the patient to write a sentence. Score one if the sentence is sensible and has a

verb and a subject.

1 1. Copy intersecting pentagons.

73

APPENDIX C

THE PRIME MD PHQ

74

 

Appendix C

PRIME MD PHQ

For the last 2 weeks have you had any of the following problems nearly every day?

1. Trouble falling or staying asleep, or sleeping too much yes no
2. Feeling tired or having little energy? yes no
3. Poor appetite or overeating? yes no
4. Little interest or pleasure in doing things? yes no
5. Feeling down, depressed or hopeless? yes no

6. Feeling bad about yourself---or that you are

a failure or have let yourself or family down? yes no

7. Trouble concentrating on things, such as reading

the newspaper or watching television? yes no

75

 

10.

Being so fidgety or restless that you were moving around

a lot more than usual? yes no

In the last 2 weeks, have you had thoughts that you

would be better off dead or of hurting yourself in some way? yes no

Are answers to five or more of #1 to #9 yes?

76

 

Appendix D

MULTILEVEL ASSESSMENT INSTRUMENT

 

 

77

Appendix D

MULTILEVEL ASSESSMENT INSTRUMENT

A. Type of housing

B. Do you live alone?

C. Who lives with you?

1.

2.

M1.

About how often do you go out of this house in good weather?

About how often do you leave the neighborhood?

Do you own and drive a car now?

Do you use the telephone?

Why is it that you don't use the telephone?

Do you get to places out of walking distances?

Why is it that you have help/don't go at all?

Can you get to places out of walking distance?

How long have you been having these difficulties?

Have you changed how frequently you get to places outside of walking
distance?

Have you changed the method you use to get places outside of walking
distance?

How did you change this?

Why did you change this?

How many city blocks do you usually walk on a typical day?

78

 

Mla
Mlb
Mlc
Mld
Mle
le
Mlh
M2

M2b
M20
M2d
M20
M2f

M2g

Do you walk 3 blocks/1/4 mile?

Why is that you don't walk?

Can you walk without help from others?

About how long have you been having these difficulties?
Have you changed how frequently you walk?

How did you change this?

Why have you changed this?

How many times a day do you climb a flight of stairs?
Why is it that you don't climb stairs?

Can you walk up and down a flight of stairs?

About how long have you been having these difficulties?
Have you changed how frequently you climb stairs?
Have you changed the method that you use to climb stairs?

How did you change this?

7. Do you shop for groceries?

8. Do you prepare your own meals?

9. Do you do your own housework?

10. Do you do your own handyman work?

11. Do you do your own laundry?

12. Do you take any medicines or use any medications?

13. Do you manage your own money?

14. Do you ever have trouble getting to the bathroom on time?

79

 

   

llillilllllilililiillil