Download Outpatient Mental Health Service Use by Older Adults After Acute

Brief Reports
Outpatient Mental Health Service
Use by Older Adults After Acute
Psychiatric Hospitalization
Hong Li, PhD
Enola Proctor, PhD
Nancy Morrow-Howell, PhD
Abstract
This study described outpatient mental health service used hy elderly patients discharged from
acute inpatient psychiatric treatment for depression, (Ltsessed sen'ices barriers, and identifiedfactors
related to the use of outpatient mental heallh sen'ices. The sample consisted of 199 elderly patients
discharged home from a geropsychiatric unit of an urban midwestern hospital. Multivariate logistic
regression was used to identify factors associated with use of various mental health sen'ices. Ahnost
three quarters ofthe elderly patients saw a p.sychiatri.st within 6 weeks postdischarge, but few used
other outpatient mental heallh services. The mostfrequently reported barriers to use included (I) cost
ofsen>ices, (2) personal belief that depression would improve on its own, and (3) lack of awareness of
available services. The use of various outpatient sen'ices was differentially related to predi.Kposing,
need, and enabling factors. Female patients, those residing in rural areas, and those who wanted to
solve their problems on their own were less likely to use outpatient mental health sen'ices. Patients
who teported greater levels of functitmal impairment, resided in rural areas, and petveived that
getting services required too much time were less likely to see a psychiatri.st in the postacute period.
African American patients were more likely than whites to use day treatment programs. This may
be related to lhe fact that most day treatment centers were located in areas where the majority of
residents were African Americans,
Introduction
Major depression affects up to 5% of older adults in this country.' Il is also the most common reason thai older adults are admitted to acute inpatient psychiatric treatment." Data from the
2000 National Hospital Discharge Survey showed that nearly 1.5 million (4.7% of all hospital
discharges) patients were discharged from a non-Federa! short-stay hospital for severe mental disorders and 9.2% of these patients were older adults.^ There is evidence showing that most elderly
Address correspondence to Hong Li, PhD. Schm^I of Social Work, University of Illinois at Urbana - Champaign. 1207 W
Oregon, Urbana. IL 61801. E-mail: hongli@uiin:,edu.
Enola Proctor, PhD, is from George Warren Brown School of Social Work. Washington University in Sl LjJuis,
Nancy Morrow-Howell, PhD. is from George Warren Brown School of Social Work, Washington University in St Louis.
Journal of Bebavioral Health Services & Research, 2005, 32(1). 74-84. © 2005 Nalional Council for Community
Behavioral Healthcare.
74
The Journal of Behavioral Health Services & Research
32:1 January/March 2005
palients respond favorably to inpatient psychiatric treatment.'' However, due to shortened inpatient stay, depressed elders rarely complete full treatment while hospitalized. Rather, they are expected to complete treatment and recovery in the community or in extended care facilities after
Posiacute mental health services at outpatient settings are particularly important because most
depressed elders are discharged to home rather than to institutional settings/ and because the recovery
process is longer for elderly patients."'^ Indeed inadequate postacute services may undermine the
effectiveness of inpatient care.'" It remains unknown whether patients' improvements achieved
during hospitalization can he maintained or enhanced by outpatient mental health services.'' A logical
step in addressing this question is to first understand outpatient mental health services provided to
elderly patients, the service barriers encountered, and factors associated with patterns of service use.
In the many studies of health service use guided by Andersen's behavioral model,'" need has
emerged as the strongest predictor of mental health service use by elderly persons.'•*"'' Predisposing
and enabling factors inlluence mental health service use more strongly than they do use of physical
health services.'" African Americans and Hispanics are less likely than whites to use mental health
services.'-^ Female patients are more likely to visit psychiatrists than their male counterparts."' Individuals with high socioeconomic status are more likely to visit a mental health specialist.'^ Patients
living in rural areas are less likely to use certain mental health services than their urban counteiparts.'^
In addition, elderly patients encounter service barriers that hamper the use of outpatient mental health
services.'^' These barriers include out-of-pockel cost for medications (patient barriers), lack of information and stigma (patient and provider barriers), and limited time and ability to track chronic
conditions such as depression In primary care (provider harriers).
Concemed about continuity of inpatient and outpatient mental health treatment, a few researchers
have focused on the use of postacute mental health services.'"''*'' Kruse and colleagues'** studied
patients" attendance at the first appointment after psychiatric hospital discharge. They found that
minority patients and rural patients were less likely to attend these appointments than their counterparts. Olfson and colleagues'** studied the linkage hetween inpatient care and outpatient care
for patients with schizophrenia. They reported that the contact hetween palients and outpatient
clinicians before discharge significantly improved patients' compliance with psychiatric outpatient
visits.
The previous studies have contributed to growing body of knowledge of using postacute mental
health services, and they also revealed some limitations. Kruse and colleagues"* focused on the first
appointment after discharge. It is not known whether the first appointment was for patients' physical
or mental health problems. To expand the understanding of postacute mental health service use, the
present study examined elderly patients' use of a broad range of service options, including visits
to psychiatrists, primary care physicians, other mental health specialists (e.g., psychologists, social
workers, geriatric nurses) day treatment centers, and clergy. In addition, this study assessed service
barriers that elderly patients might encounter while using these services. The specific questions of
the study were as follows: (I) What outpatient mental health services are used by older adults with
major depression after hospital discharge? (2) What services barriers do older adults experience in
accessing outpatient mental health services in the postacute period? (3) What factors are associated
with depressed older adults' use of outpatient mental health services?
Methods
Sample
Elderly patients in the .study were recruited from the geropsychiatric unit of an urban midwestem
teaching hospital. They (1) were 65 years of age or older, (2) were hospitalized for treatment of
depression, (3) had a DSM-IV Axis I diagnosis of depression, and (4) were discharged to a community
Outpatient Mental Heulth Service Use
U et al.
73
setting rather than a nursing home, hospital, or other institutional settings. Patients with cognitive
impairment were retained if they wereadmitted to the unit for treatment of depression. Two patients
were excluded because they were unable to provide reliable detiiographic and health infomiation
(according to the assessment of hospital staff) and lacked a collateral source of information. Diagnosis
was confirmed through chart review by the unit's medical director, a geropsychiatrist,
Paiients who met the above inclusion criteria were asked to participate in the study and were
provided written informed con.sent for study participation during hospitalization. If the palicnt was
cognitively impaired, according to a standardized test administered by the unit nurse as well as nurse
clinical judgment, consent was obtained from botb the patient and a family member. The consent
rate was 73.5%, and sample size was 199. At 6 weeks after discharge, 3 palienis died, 9 refused to
continue participating. I moved to an undisclosed location, and 15 moved to nursing homes. As a
result, this siudy included 171 patients or proxies, A totai of 56 proxies (32.8%) were used when
patients were loo frail to participate in the interview. Paiients included in the 6-week follow-up
interview did not differ significantly from those who were excluded on age, race, or marital status.
Subject recruitment occurred between March 1997 and May 2000. During the study period, elderly
patients were interviewed before discharge and at 6 weeks after discharge, via a structured telephone
interview. Data such as patients' demographic and mental health information were extracted from
medical records.
Measures
Outpatient mental health service «.sc
Similar to the Epidemiologic Catchment Area (ECA) studies,^" the mental health services assessed
in thi.s study were provided by psychiatrists, primary care physicians, day treatment programs, inhonie psychiatric counselors, office-based psychiatric counselors, and clergy. To measure outpatient
mental health service use. elderly patients or their proxies (e.g., family metnbers) were asked whether
or not they used the above-listed services after discharge and the number of visits to these services.
Although self-reported use of psychiatric services has disadvantages, it has been widely used in
mental health service research.''
Psychiatrists included bolh admitting psychiatrists seen again after discharge and other psychiatrists. To identify use of primary care physicians for mental health services, patients were asked
whether such visits to primary care doctors were for mental health problems or for physical health
problems. Day treatment centers were outpatient programs that provided psychiatric assessment
and psychopharmacological and psychotherapeutic treatinent to patients with psychiatric disorders.
Some of the day treatment programs were administered by outpatient mental health centers and some
were administered by hospitals. In-home psychiatric counseling included visits from geriatric social
workers and nurse practitioners. Office-based psychiatric counseling included visits to psychologists,
social workers, and other types of trained counselors. Patients were aiso asked whether they received
counseling from clergy for mental health problems.
Predisposing factors
Consistent with Andersen's behavioral model, the 4 predisposing factors were patient's age,
gender, race, and educational attainment. Age, race, and gender were abstracted fTom patient's bospital admission records, and educational attainment was gathered through self-report and collapsed
inlo 2 categories: less than high school versus higb school or more education.
Enabling factors
Enabling factors were patient marital status (married or not), income, living arrangement (living
alone or not), residence location (urban or rural), insurance status, and perceived service barriers.
76
The Joumal of Behavioral Health Services *fi Research
32:1 January/March 2005
The annual household income had 2 categories: equal to or less than $ 10.{)()0 and more than $ 10,CX)O.
Residence location was extracted from their hospital admission records, and coded as rural (in an
area with a population less than 2500) or urban. Medicaid enrollment and private healthcare plan
coverage were dichotomous variables and extracted from hospital records.
The measure of service barriers was adopted from a study focusing on barriers to seeking treatment
for major depression.^' It included issues such as cost of services; distance/lack of transportation;
patients' perception that ihe service will not do any g(X)d; patients' perception that they can solve the
problem on their own; and patients' concem that other people may find out about the mental health
problem (the complete list of barriers is shown in Table 3). At the 6-wcek telephone interview, all
patients were asked to report whether they experienced any of the listed barriers lo use of mental
health services since hospital discharge.
Needfactors
Need factors included whether the depression had psychotic features, whether this was the first
episode of affective disorder, general psychiatric severity, pyschosocial functioning, and functional
status. At the lime of hospitalization, the geropsychiatrist. the director ofthe unit, recorded presence
or absence of psychotic features among the depressive symptoms and if this was the first episode or
not.
General psychiatric severity was measured by the Brief Psychiatric Rating Scale (BPRS). The
BPRS has 18 items, each with a 7-point response option, with the total score ranging from 0 to
108. The higher scores indicated more severe .symptomatology. The BPRS has been successfully
used to evaluate the severity of psyehopathology among inpatients.-' In addition, the psychosocial
functioning was also assessed by unit slaiT, using the Global Assessment of Functioning (GAF). The
GAF scores range from 0 to IOO, with higher scores indicating better functioning. The reliability and
validity of GAF have been demonstrated in studies of subjects from different populations, including
psychiatric inpatients and aftercare patients.-** The BPRS and GAF were routinely administered by
unit staff. The BPRS and GAF scores used in this sludy were the last ones recorded in patients'
medical records during hospitalization.
On the basis ofthe patient's medical records, a nurse practitioner completed tbe Cumulative Illness
Rating Scale, which quantified chronic medical problems.^^ The Cumulative Illness Rating Scale
was moditied for use witb a geriatric population.-^ The severity of illness in 13 organ systems was
rated on a 5-point scale (0-4), with 0 being no problem and 4 being extremely severe. The score of
each organ system was summed. The total score ranged from 0 to 42. with higher scores indicating
greater illness. Tbis scale has been used wiib medically and psycbiatrically impaired elderly and has
established interrater reliability (0.78-0.88) and face validity.^*
Questions in tbe predischarge interview assessed patients' ability to perform activities of daily
living al the time of hospital discbarge including walking, eating, using toilet, grooming, dressing,
and bathing, and instrumental activities of daily living including taking medicine, preparing meals,
sbopping, managing money, transportation, and housekeeping, using the Multidimensional Functional Assessment of Older Adults.^^ Ability on each of 13 items was assessed using a scale of 0 to 2.
The summing yielded a score ranging from 0 to 26 with bigher scores indicating better functioning.
Finally, type of respondent consisted of 2 categories: self or proxy response. This variable was
included in the analysis to control for the effects of using proxy respondents on outpatient mental
health service use.
Data analysis
Central tendencies and distributions of all study variables were examined. Since perceived barriers
and use of different types of mental healtb services were measured dichotomously, chi-square tests
Oittpatieni Mental Health Sen'ice Use
U el aL
11
were used to examine associations. Barriers found to be significant in hivariate comparisons at the
level of 0.10 or less were used along with predisposing, enabling, and need factors to predict service
use, using logistic regression analyses. Logistic regression yielded parameter estimates that can be
converted to odds ratios, which indicate the direction and magnitude of a factor's association with
service u.se.
Impution was used to avoid losing observations due to missing data. Two variables had missing values. General psychiatric severity (BPRS) at time of discharge had 30 missing observations
(17.5%), and psychosocial functioning (GAF) at time of discharge had 10 (5.8%). These data
were missing from the hospital records. A "hot decking" procedure was used to impute these 2
variables.'" A correlation matrix is used to determine the most highly correlated variahles with
the variahie that had missing values. The data were sorted by that correlate and a random variable
(tie breaker). Using the sorted data set, the missing observations of the variable were replaced by
the observation that appears for the preceding participant. As a result, missing observations were
replaced with a value from a case that is similar on a highly correlated variahie. To compare the
effect of this imputation method, missing ohservations were also imputed by mean substitution,
where the mean BPRS and GAF of the total sample were inserted for missing values. Comparison
of these 2 imputation methods showed no changes in signs and statistical significance of model
coefficients.
Multicollinearity between independent variables was also assessed. The tolerance values were
in the acceptable range (TOL > 0.25) and indicated that none of the variahles had the problem
of multicollinearity.^'^ The significance level for this analysis was set at less than or equal to 0.05
(2-tailed).
Other studies have found that eldedy patients with mental disorders frequently sought help from
primary care physicians."* However, in this study, only 4 elderly patients consulted their primary care
physicians for purposes of mental health treatment. This minimum use of primary care physician
reflected the fact that these elderly patients were discharged from inpatient psychiatric treaiment.
Because of the small sample size, use of primary care physician was not analyzed hy multivariate
logistic regression.
Results
Characteristics of sample
Table I summarizes the characteristics of elderly patients in the sample. The mean age of respondents was 76.2 years ranging from 65 to 98 years. A majority ofthe respondents (70.8%) were
female, and more than one half were unmarried. Of the total sample, 14.6% were African Americans.
Almo.st one third had an annual household income equal to or less than $IO.O(M). and two thirds had
less than a high school education. In addition, close to one fifth of patients lived in rural communities.
All had Medicare. Less than one tenth were enrolled in Medicaid, and nearly one fifth had private
healthcare plans.
Pattems of outpatient mental health service use
.
J^
Thefindingsshown in Table 2 indicate that more than three quarters of elderly patients (81.9%) used
mental health services at 6-week postdischarge. A vast majority ofthe visits were to a psychiatrist.
Of these psychiatrists, nearly two thirds were patients' admitting psychiatrists. In addition, 14.6%
of patients used day treatment programs. Close to one tenth of patients used in-home mental health
services provided by geriatric nurses and social workers. Only 4.7% of patients visited mental health
specialists such as psychologists, social workers, and counselors outside of their homes. Finally,
15.2% of patients saw clergy for mental health counseling.
»
T
78
The Joumal of Behavioral Health Services & Research
32:1 January/March 2005
Table 1
Patient eharacleristies (A' = 171)
Variables
Percentage
Age
Gender
Male
Female
Race
White
Black
Education
Less than high sehool
High school and more
Marital status
Married
Not married
Living arrangement
With someone
Alone
Income
M (SD)
76.2(7.1)
29.2
70.8
85.4
14.6
66.3
33.7
42.7
57.3
69.6
30.4
< 10.000
32.4
> 10.000
Medicaid
Yes
No
Other insurance
Yes
No
Residence
Urban
Rural
67.6
7.0
93.0
82.5
17.5
80.7
19.3
Table 2
Outpatient mental health service use at 6 weeks {n = 171)
Outpatient mental health services
Use, % (/t)
No. of visits, M (SD, range)
Any community MH services
Psychiatrists
Admitting psychiatrists
Other psyehiatrists
MD for MH problems
Day treatment programs
Counseling services
Omce-based MH Specialists
In-home MH specialists
Clergy for MH problems
81.9(140)
74.9 (128)
63.7(109)
11.8 (20)
2.4 (4)
14.6 (25)
15.1 (28)
4.7 (8)
9.9 (17)
15.2 (26)
10.1 (20.2,1.0-172.5)
2.5 (4.1. LO-30.0)
2.6(4.3, 1.0-30.0)
2.1 (2.4, 1.0-12.0)
2.3 (1.3,1.0-4.0)
17.2 (8.9, 2.0-30.0)
19.4(34.3. I.O-I69.5)
2.1 (1.0, 1.0-4.0)
18.7 (14.8, 1.0-54.0)
2.9 (2.6. 1.0-11.0)
Outpatient Mental Heallh Service Use
LI et al.
79
Table 3
Service barriers and relationships tooutpatieni mental heallh service use (ri = 171)
Service barriers
% (n)
Any barrier
Barrier I: Costs of services
Barrier 2: Unaware of where get service
Barrier 3: Transportation/distance
Barrier 4: Take too much time
Barrier 5: Service will not do any good
Barrier 6: Get better by itself
Barrier 7: Solve the problem on their own
Barrier 8: Other might find out about their problem
Barrier 9: Concem about what others might think of if they went for help
•M = 2.().SD= 1.7.
77.2(132)*
40.6 (69)
24.1 (41)
20.0 (34)
10.0 (17)
20.0 (34)
21.2 (36)
41.8 (71)
11,8 (20)
10.6 (18)
J,
Ofthe patients who used services, the number of visits varied considerably. Patients had an average
of 2.5 (SD = 4.1) visits to their psychiatrists in the 6-week observation period. Among those who
used in-home mental health services, patients received an average of 18.7 (SD = 14.8) visits from
these services ranging from 1 to 54.
1'
Perceived barriers to the use of outpatient mental health services
As can be seen in Table 3, more than three quarters of elderly patients reported one or more service
barriers, with tbe sample identifying a mean of 2.0 (SD = 1,7) barriers. The most frequently reported
barriers were related to patients' perception that they could solve the problem on their own (41.8%),
patients" concem about costs of services (40.6%), and patients' unav^'areness of where to get services
(24.1%).
Factors associated with use of outpatient mental health services
To identify significant barriers to the use of ditferent types of mental health services, chi-square
tests were performed. Three service barriers were significant at less than ,10, and were thus included
in mullivariate analyses: patients' perception that depression would improve on its own, that services
would not do any good, and that they couid solve the problem on their own.
In addition to these barriers to mental health services, other factors associated with elderly patients'
use of different types of menial health services were examined in 5 separate multivariate logistic
regression analyses. Three out of the 5 overaJI regression models were statistically significant: use
of "any" outpatient mental health .services, use of services provided by psychiatrists, and use of day
treatment. The models predicting use of counseling services and clergies for mental health problems
were not significant.
In relation to use of any outpatient mental health services, patients' chronic health conditions and
residence locations were statistically significant (see Table 4). Elderly patients with more severe
health conditions were less likeiy to u.se any outpatient mental health services than elderly patients
with less severe health conditions (odds ratio [OR] = 0.85. 95% CI: 0.73-0.99). Compared with
tbeir rural counterparts, urban elderly patients were almost 3 times as likely, in terms of odds, to use
any outpatient mental health services (OR = 2.97,95% CI: 1.01-8.76).
80
The Journal of Behavioral Heallh Services & Research
32:1 January/March 2005
In relation to the use of mental health services provided by psychiatrists, only urban/rural status
was statistically significant. Urban elderly patients were 4.5 times as likely, in terms of odds, to use
psychiatrists as were rural elderly patients (OR = 4.52, 95% CI: 1.70-11.97).
A similar model was used to identify factors associated with use of day treatment programs. Since
day treatment centers were not available to the rural elderly patients in the study, patients* residence
location was dropped from the analysis. Ethnicity was the only factor that w as significantly related to
use of day treatment programs. Compared with their African American counterparts, white patients
were 7% as likely, in terms of odds, to use day treatment (OR = 0.07. 95% CI: 0.02-0.32). It is
probable that this finding is confounded with the fact that most day treatment centers attended by
study participants were located in areas where the majority of residents were African Americans
(Tahie 4).
Discussion
This study revealed several important findings with implications for outpatient mental health
providers. A vast majority of elderly patients made follow-up visits to mental health serviee providers
at 6 weeks postdischarge; ihis suggests continuity of inpatient and outpatient treatmeni. However,
because these older adults had depression severe enough to warrant hospitalization, it is troubling
that nearly one fifth of elderly patients did not use any mental health services within 6 weeks after
their discharge.
The findings of this study further showed that psychiatrists, including both admitting and other
psychiatrists, played a leading role in the early stage of postacute care. In contrast, only a few elderly
patients sought help from primary care physicians or other outpatient mental health specialists,
including psychologists, social workers, and geriatric nurses. This observed pattern may he unique
at the earlier stage of postacute periods because follow-up care provided by psychiatrists may reduce
or delay the use of services offered by other mental health specialists.
Yet a relatively high percentage of elderly patients diseussed their mental health conditions with
their clergy. A growing amount of literature has documented that religious leaders are actively
involved in counseling patients with mental illness.'"^ Because of the trust built over years, no cost,
and easy access, elderly patients may feel less hesitant to approach clergy wilh their mental health
problems than to approach mental health specialists.
In seeking follow-up mental health services, the elderly patients in this study experienced similar
harriers as those reported by other researchers.'*-^^ The finding that a relatively large proportion of
elderly paiients still believed that they could solve the problems on their own and were still unaware of
community services may suggest areas of improvement in patient education and discharge planning.
Tn predicting the use of follow-up services, patients' chronic health conditions were related to use
of mental health services. Elderly patients with less severe chronic conditions were more likely to
use any psychiatric services. This finding suggests that medical conditions comorbid with depression
can pose barriers to mental health service use. Patients' chronic conditions may present "competing
demands" to the mental health condiUons: that is, these conditions compete for the resources and
energy of the depressed older adults, who may prioritize attention to these conditions over attention
to depression.
Finally, consistent with other studies, rural patients were at a disadvantage in accessing outpatient
mental health services.'^ There is evidence suggesting that shortage of psychiatrists and mental
health specialists and fragmentation of community mental health service systems in rural areas are
some ofthe main obstacles faced by rural patients.'^
This study has several limitafion.s. The study sample was limited to elderly patients who all had
a diagnosis of depression and were discharged from a geropsychiatric unit of one general hospital.
While this design enhances control of many patient factors and in-hospital treatment factors, it may
be vulnerable to the influence of the location of the study hospital and variations in the availability
Outpatient Mental Health Service Use
LJ et ai
81
(N
o
o
i
a,
r-'
£
o
rf
O
O f^i
oq ov
oooo
oooooo
oo
^^ VO ^D C^
O C* C^ O
II
^^ ^5 ^5 O^ Oi ^^
' ^ *^
O ^5 ^5 O ^O O
O ^5
II
111
°^ "+
O 1^
O O
-^"
C^ ^
~^ ~^
^ **D
*O ^"
II
p
ci
^2
SI
o o
o
r- o In 'Q
o o o o
so
oo —•
o o o o
o o
E ro- ino
—
06
43
51
o
B
o o
46
04
r- ro
o o ra o
s .a
02
04
12
02
II
70
83
07
fl
Ov 00
^
p
o o ^
d d
s;
X
op
R
e~
II
rj
3
o
l
m o 00 o 00 o
o o o o
o
o o o
o —o o
o o o o o o
I
I
00
in
00
o o
r- "n
o
— o tn
o
40
62
09
m
o o
05
06
16
02
02
34
3
o o
ri
02
40
§2
03
02
87
70
i'E
oq
d
in
in
d^ ^
d d
d d X
o —
I
a.
•B
$
6-
c
'>
cut
S
•a
g.
-e
u
« o
a
tl.
c
I
82
<
a>
O
u u
T3
OJ
fl
w
O
E—
a
^ '^ V
u2 a . O
oo
TD
[«
^
tu- ^
a
—
>i
•
—
.
>.
.
-
^
>.
.
—
•
JJ in o r2
u
Cfl « CU
5 *- == -q S
^.b ii J ;S
U t l , CL, a. g S _J
^
^
^
g i2 £2 E2 "o
^
J J _<L) _(U in
iii ra fl S >,
OS CO CD ca H
tu
The Journal of Behavioral Health Sen'iees & Research
32:1 January/March 2005
of outpatient mental health services, thus limiting generalizability of this study. Also, most of the
data used in this study were self-reported, and self-report data have disadvantages. For example,
patients may not be able to distinguish the reasons for a doctor visit because of the comorbidity of
their conditions. However, seif-reported information has been widely used in mental health service
research'' and was the most feasible approach in this study. Finally, the relationships between use
ot serviee and significant independent variables should not be interpreted as unidirectional or causal
because the data used for this analysis were cross-sectional.
Implications for Behavioral Health
Postacute mental health services play a critical role in elderly patients' recovery from severe
depression. To enhance treatment continuity and overcome the service barriers, psychiatrists with
the support of discharge planners need to make persistent efforts to educate patients about their
conditions and their need for continuing care as per practice guidelines and to connect tbem to local
outpatient services. Given patients' contacts witb clergy, training clergy members on assessing and
treating mental illness may improve their effectiveness in recognizing mental illness and linking
patients to needed services.
Rural patients remain vulnerable to the lack of postacute follow-up services. Strategies sueb as integration of primary care and psychiatric care and linkage between primary care and outpatient mental
bealtb centers can be used to improve rural patients' access to and use of outpatient services.'*""
Mental health serviee providers need to understand that depressed older adults with pressing medical
problems are less likely to use mental health services, even after psyehiatric hospitalization. Thus
lhe primary healthcare system needs to improve its capacity to deal with mental health problems.
At the same time, bringing mental health services into the bome tbrough psychiatric social workers
and nurses who provide home care .services may be another strategy that engages depressed elderly
patients with poor physical health.
References
!. Gurland B. Cross P. Katz S. Epidemiologica] perepecrives tin opportunities for treatment of depression. Aineri^an Journal of Geriatric
Psychiatry. t996;4:S7^SI3.
2. Ertner S. Hermann R. Inpatienl pKychiairic ircatmcnt of elderly Medicare beneficiaries, \9<i{i~\'^\. Psychiotric Services. 1998;49:I1731179.
3. HiillMJ.OwingsMF. 2(XK)NationaiHospila] Discharge Survey. AJiwirc Du/u. 2002:329:1-19.
4. HinrichscnG. Recovery and relapse fmm major depressive ttisiirder in lhe elderly.-4merii-fl/jiffHnnj/()/7'.Tyc"''iar/>'. 1992; 149; 1575-1."179.
5. Davis t.. Wells K. Rogers W. Benjamin B. ct al. Effecis of Medicare')( prospective payment syslem on service use by depressed elderly
inpalicnts. Psyvhiairic Sen'ices. I99.*i;46:l 178-1184.
6. Boyer C, McAlpine D. Pottick K. el al. Iclentifytng risk faclois and key strategies in linkage lo outpatient psychiatrie care. American Journal
ofPayt-hiiitry. 2000; 157:1592-1598.
7. Welh K. Norquist G. Benjamin B. cl al. Quality of antidepresKani medicaiions prescribed at dischiir^e [o depressed elderly paiienU in
general medical hospital.^ before and urier prospective payment system. General Hospital Psycliiairy. 1994; 16:4-5.
8. Caliahitn C. Hui S, Nienaber N, ct al. Longitudinal study of depression and health services use among elderly pdmary care patients. Jourmil
of lhe American Geriiiirics Society. 1994;42;X3.V838.
9. ColeM, BellavanceF. The pmjinosis of depression in o\d a^e. Amerimn Journal of Geriatrii Psychiatry. 1997;5;4-I4.
10. George L. Conununiiy and home care for mentally ill older adults. In: Birren J. Sloane R, Coben G, el al. ed. Handbi>ok of Mentul Heatih
onilAginfi. San Diego. Calif: Academic Press: 1992.
11. Hinrichsen G, Hernandez N. Faet<irs associated with recovery from and relapse into major depressive disorder in the elderly. American
Journal of Psychiatry. 1993;150:1820-1823.
12. Andersen R. Revisiting the behavioral model and acce.ss 10 medieal eare: does it matter? Joiima/o/Wefl/rttCSor/fl/BfAaWor. 1995:36:1-10.
I.T. Hu T. Snowden L. Jcrrel J, et al. Ethnic populations in public mental health: services choice and level of use Ami-rican Journal qfPuhlk
Health I992;8I:I429-1434.
14. Kessler R. Zhao S. Kat/ S. et al. Post-year use of outpatient services for psyehiatric problems in the National Comorbidity Survey. American
Joumal of Psychiatry. 1999:156:115-123.
15. PadgL-ti D. Patrick C. Bums B. et al. Hthniciiy and the use of ou^>atient mental health services in a national insured population. <
Joumal of Public Health. 1994:84: 222-226.
Outpatient Mental Health Service Use
U et al.
83
16. Unut/cr J. Kuion W, Suiliv.tn M. ci al. Treating depre!»cd older tidutts in primary care: nairowing the gap belween efficacy and efTcctivciieBs.
77(c MiHxMiA Quarterly. 1999;77;225-230
17. Sullivan G, Jackson C. Spdizcr K. Characlcri.ilics and service use of seriou.sly mentally ill persons living in rural area.s. Psychiatric Services.
19%;47:57-h!.
1 a. Kruse GR. R»hlund B. Facinrs associated wilh iittendance ai a first appointment afler dischstse from a psychiatric hosfrital. P.tychiatric
Ser,'ices. 2()02;4:47 3-^76.
19. Olfson M. Guiinlinii M. Struening E, el al. Barriers lo lhe treatment of sociai anxiety. American Journal ofPsycttiatry. 2000; 157:521-527.
20. Shapiro S. Skinner EA, Knuner M, et al. lliiliiriition nf health and mental health services: three Epidemiologic Catchment Aim sites.
An-hives of General Fivchiairy. 1984:42:89-94.
21. Kes-;lcr R, Berglund P. (X-mler O. el al. The epidemiology of major depressive disorder results from the Katiunal Comorbidity Survey
RepHcuiion (NCS-Rl, JAMA 2(X)3:289:3(W5-3IO522. Btunienihat R, Endknit J, Barriers lo seeking treatmenl ftir major depression. Deprrssioniind Atixitriy. l99()/1997;4:273-278.
23. Overall J, Gorham D. The Brief Psychiatric Rating Scales (BPRS): retent developnienis in asccrtainmeni and scaling. Psyrhopharmacology
Bullciin. l988;24:97-99.
24. Endicott J. Spit/er R. Fleiss J. ct sl. The global assessment scale: a procedure for measuring over^l severity of psychiatric disturbance.
Archives of General Psychiatry. I976;33:766-771.
25. Linn B. Linn M.Gurel L. Cumulative \HnciisR9tin$Scale.Joumat of the American Geriatrics Society. 196S; 16:622-625.
26. Miller M. Paradi.s C. Houck P, el al. Rating chronic medical illness burden in geropxychiatric practice and research: application ofthe
Cumulalivc Illness Rating Scale. P.sychiairy Re.smrch. 1993;4|;237-248.
27. ViWcnh&amG. Multidimensionalt'unctional Assessment of Older Adult: The Puke Older Americans Resources and Services Pmcedurts.
Hillsdale, NJ: Erihuum; 1988.
28. Allison P. Missinf! tiaia. Thousand Oaks. Calif: Sage; 2002.
29. Fox J. Regression Diagnostics. Ncwbury Parks. Calif: Sagc:199L
30. Taylor R. Ellison CL. Chatters L. el al. Mental health services in faith communities: the role of clergy in back churches. Sociat VhA.
2000;45:73-82.
http:// jbhsr.fmhi.usf.edu
84
The Joumai of Behaviorai Health Services & Research
32:1 January/March 2005