Download Differences Among Primary Care Physicians

Copyright 1999 by
The Cerontological Society of America
The Cerontologist
Vol. 39, No. 4, 417-425
This study examined differences by specialty of primary care physicians in managing
suicidal and depressed geriatric patients. A probability sample of 300 Illinois physicians
drawn from the American Medical Association Physician Masterfile was surveyed.
Significant differences were found between specialties in estimates of the prevalence of
psychiatric disorders; use of assessment procedures, treatment approaches, and referrals;
perceptions of obstacles to providing mental health care; and confidence in diagnosing
and treating depression and suicidality. Meeting the mental health needs of the rapidly
growing older population will require a greater emphasis on geriatric mental health and
consistency across primary care specialties.
Key Words: Elderly suicide, Mental health, Physician specialty
Managing Depressed and Suicidal
Geriatric Patients: Differences
Among Primary Care Physicians
Mark S. Kaplan, DrPH,1 Margaret E. Adarriek, PhD,2 and Alvin Calderon, MA3
even more responsibility for diagnosing, treating, and
managing all areas of health care, including care for
ambulatory patients with mental disorders (Lanier &
Clancy, 1996; Lebowitz, 1996; Wells, Sturm, Sherbourne,
& Meredith, 1996). However, the quality of care provided by primary care physicians to such geriatric
patients and the effects of such care are of growing
concern, especially in light of the restrictions imposed
by managed care plans (Unutzer et al., 1997).
A variety of constraints inhibit primary care physicians from assessing and treating mental health symptoms. Productivity standards in managed care, with
tighter scheduling of shorter appointments (the average patient-physician contact is 6 to 7 minutes; Cohen
& Cairl, 1996), appear to discourage them from exploring emotional problems with their elderly patients
(Waitzkin, Britt, & Williams, 1994). In addition, cost
containment mechanisms restrain physicians from addressing the mental disorders of physically ill geriatric
patients (Lebowitz, 1996).
Moreover, many physicians do not recognize or screen
for the prevailing patterns of mental disorders among
ambulatory geriatric patients (Henderson, 1990; Lazarus,
1995; Lichtenberg, Gibbons, Nanna, & Blumenthal,
1993; Rapp & Davis, 1989; Waxman & Carner, 1984).
According to Spitzer and his colleagues (1994), generalists are not sufficiently knowledgeable about diagnostic criteria and are not certain about the best questions to ask to evaluate whether those criteria are met.
Therefore, they not only underrecognize mental disorders (Spitzer et al., 1994), but also underrefer patients for psychiatric consultations (Cassata & KirkmanLiff, 1981; Rogers, Wells, Meredith, Sturm, & Burnam,
1993), inappropriately prescribe psychotropic medications (Eisenberg, 1992; Kamerow, Pincus, & MacDonald,
1986; Katon, Von Korff, Lin, Bush, & Ormel, 1992;
As many as 45% of the elderly patients who are
seen by primary care physicians experience some level
of depression (Rapp & Davis, 1989; Wells, Hays, Burnam,
Greenfield, & Ware, 1989). The high rate of suicide
among elderly persons (Mclntosh, Santos, Hubbard,
& Overholser, 1994) and their greater likelihood of
seeking mental health care from primary care physicians than from mental health specialists (Cohen &
Cairl, 1996) underscore the importance of investigating how physicians manage depressed or suicidal elderly patients. Although numerous studies have examined how primary care physicians assess and treat psychiatric disorders (e.g., Badger et al., 1994; CooperPatrick, Crum, & Ford, 1994), few have focused on
clinical practices with depressed geriatric patients
(Callahan et al., 1994; German et al., 1987; Rapp &
Davis, 1989).
Primary care physicians have been increasingly involved in providing psychological interventions (Olfson
etal., 1995; Orleans, George, Houpt, & Brodie, 1985;
Spitzer et al., 1994). Furthermore, with the continuous expansion of managed care plans and hence the
greater importance of primary care, they may assume
This research was supported by grants from the Retirement Research
Foundation (94-148) and the University of Illinois at Urbana-Champaign
Campus Research Board. An earlier version of this article was presented
at the 43rd annual meeting of the American Society on Aging, March
1997, Nashville, TN. The authors thank the Illinois State Medical Society,
Dean C. C. C. O'Morchoe, Dr. Nestor Ramirez, Jon Rhoades, Wendy Almelen,
Dr. Bentson McFarland, Dr. Larry Goldman, and Olga Celing for their
assistance with this project.
Address corresponaence to Mark S. Kaplan, DrPH, School of Community Health, Portland State University, P.O. Box 751, Portland, OR
97207-0751. E-mail: [email protected]
2
Office of Gerontology, School of Social Work, Indiana University, Indianapolis, IN.
3
Medical Scholars Program, University of Illinois at Urbana-Champaign,
Urbana, IL.
Vol. 39, No. 4, 1999
417
Lazarus, 1995), and often attribute psychiatric symptoms either to changes expected with age or to concomitant physical disorders (Solomon, 1996).
Depression, a common risk factor for suicide, is one
of the mental health problems that primary care physicians often do not recognize (Bell, 1997; Greene &
Adelman, 1996). Studies have indicated that primary
care physicians do not diagnose depression in at least
50% to 75% of their depressed elderly patients (Stults,
1984; Wells et al., 1996). Furthermore, in a randomized clinical trial, Callahan, Hendrie, Nienaber, and
Tierney (1996) found that nearly 5% of clinically depressed older patients seen in a primary care setting
were potentially suicidal.
Numerous studies have found that most elderly persons who committed suicide—as many as 75%—had
seen their physicians shortly before their deaths (Conwell, Olsen, Caine, & Flannery, 1991; Cooper-Patrick
et al., 1994; Courage, Godbey, Ingram, Schramm, &
Hale, 1993; Frierson, 1991; Lin, Von Korff, & Wagner,
1989; Tobias, Pary, & Lippmann, 1992). Despite these
findings, little is known about the interactions between
suicidal elderly persons and their primary care providers, such as the extent to which these physicians
inquire about methods of suicide available to the patients. Less is known about whether these interactions
vary by specialty of the primary care physician. One
recent study has suggested that family physicians had
a more positive attitude or orientation toward geriatric patients with depression compared to general internists (Glasser & Gravdal, 1997).
Because family practice, internal medicine, and geriatric internal medicine specialists are the primary care
providers that elderly persons see most frequently
(Selma, Schwartz, Koch, & Nelson, 1993), they are in
a good position to detect and manage geriatric depression and suicidality. In this article, we report on a
study of these three specialties' use of assessment procedures, treatment approaches, referrals, perceptions
of obstacles to providing mental health care, ana confidence in diagnosing and treating depression and
suicidality. We also discuss innovative approaches to
mental health treatment in primary care practice as
well as the implications of the findings for medical
education and training.
Instrument
The instrument, consisting of 20 single- and multiple-response items, was developed on the basis of
consultation with leading scholars, recommendations
from the literature, and extensive pilot testing. It included questions on the physicians' primary and secondary specialties, practice patterns and training, years
in practice, and demographic information (e.g., age,
gender, and ethnicity) as well as on their practice settings (outpatient, inpatient, nursing home, communitybased clinics, or other) and the location of their practices (rural, urban, or suburban). In addition, several
questions were designed to measure the physicians'
confidence in their ability to perform specific diagnostic and treatment tasks related to their geriatric
patients' mental health.
Procedures
In May 1996, packets that included a cover letter
(signed by the investigators and the dean of the University of Illinois College of Medicine), a letter of
endorsement from the Illinois State Medical Society
(ISMS), a postage-paid return envelope, and the instrument were sent to these 300 physicians. Every
effort was made to reach the nonrespondents, including follow-up mailings and contacts by telephone and
fax. Several sources were consulted (including the
ISMS, the Directory of Medical Specialists [1996], and
telephone directories) to verify the nonrespondents'
addresses. Twenty-seven physicians' names were removed from the mailing list because they were ineligible for the study (i.e., were not practicing in Illinois,
were not MDs or DOs, were retired, had no geriatric
patients, or were deceased). An additional nine surveys were undeliverable. Of the remaining 264 valid
addresses, 167 usable surveys were returned, for a 63%
response rate. This rate compares favorably with other
surveys of physicians (Frank, Rothenberg, Brown, &
Maibach, 1997; Mort, Edwards, Emmons, Covery, &
Blumenthal, 1996). Furthermore, the respondents did
not differ from the study population on such characteristics as gender, age, and ethnicity.
Analysis
Method
To determine whether there were intergroup differences, we compared the three specialties on the
various physician and practice characteristics using
analysis of variance for continuous variables and chisquare for categorical variables. To correct for the possibility of Type I error due to multiple comparisons, a
more stringent significance level (p < .01) was used
than the conventional p < .05. We then performed
logistic regression and analysis of covariance using
physicians' specialties and any potential confounders
(demographic and practice characteristics that had a
univariate association with physicians' specialties) as
independent variables. Except for the continuous-level
confidence items, all other dependent variables (assessment procedures, treatment approaches, referrals,
and obstacles to mental health care) were categorical.
Sample
We obtained a probability sample of primary care
physicians in Illinois from the American Medical Association (AMA) Physician Masterfile, a continuously updated and primary source of data on all physicians in
the United States (Grumbach, Becker, Osborn, & Bindman, 1995). From the population of 4,980 family practice and internal medicine physicians listed in Illinois
(excluding students and retirees), we drew a sample
of 300 physicians: a stratified random sample of
247 family practice and internal medicine physicians
chosen according to their proportion in the state and
the entire population of 53 internal medicine geriatric
physicians.
418
The Gerontologist
The probability levels of the odds ratios (OR) were
ascertained using Wald's test (Morrow-Howell & Proctor, 1992).
spondents (85%) practiced in urban or suburban communities. Three quarters estimated that at least 25%
of their patients were aged 65 or older, indicating that
the respondents had substantial experience with
elderly patients. Of the seven characteristics shown
in Table 1, three (practice setting, practice location,
and percentage of patients 65+) had categories that
varied significantly among the three specialties and
were therefore used as control variables in subsequent
analyses.
Results
Physician Characteristics
Of the 167 respondents, 77% were male, and 74%
were Caucasian, 17% were Asian American, 4% were
African American, and 3% each were Latino and "other"
(see Table 1). The respondents ranged in age from 30
to 80, with a mean age of 46 years {SD = 10.6); 40%
were aged 3 5 ^ 4 . With regard to specialties, 43% were
in general internal medicine (GIM), 27% were in general family practice (GFP), and 30% had a primary or
secondary specialty in geriatrics (GER). Of the GERs,
92% of those with a primary specialty and 86% with
a secondary specialty reported being certified.
The respondents had been in practice from 1 to
53 years, with a mean of 14 years (SD - 9.8); the
largest group (43%) had been in practice 10 or fewer
years. The majority reported seeing patients in outpatient (87%) or inpatient (78%) settings, and 55% saw
them in nursing homes. In addition, most of the re-
identification of Mental Health Problems
In response to the question on which of five mental health problems (anxiety disorders, organic brain
syndrome, affective disorders, phobias, substance abuse)
was the most common among their elderly patients,
37% of physicians mentioned affective disorders (including depression), 27% mentioned organic brain syndromes, and 26% mentioned anxiety disorders; none
mentioned phobias or substance abuse (see Table
2). Of the five types of disorders, only one—anxiety
disorders—showed a significant difference (p < .01)
among the specialties. When asked to estimate the
Table 1 . Comparisons of Demographic and Practice Characteristics
Total
(n = 167)
GERa
(n = 50)
GFP
(n = 45)
GIM
(n = 72)
Sex (%)
Male
Female
77
23
78
22
71
29
81
19
Ethnicity (%)
Caucasian
Asian American
African American
Latino
Other
74
17
4
3
3
81
15
2
0
2
81
14
0
5
0
65
19
8
3
5
Age (M)
46.3
(SD = 10.6)
47.3
(SD = 9.4)
43.8
(SD = 11.1)
47.1
(SD = 10.9)
Years in practice (M)
14.3
(SD = 9.8)
16.1
(SD = 8.9)
12.1
(SD = 10.9)
14.4
(SD = 9.6)
Practice setting (%)b
Outpatient
Inpatient
Nursing home
Community-based clinic
Other
87
78
55
12
15
82
80
72*
12
22
91
76
58
13
16
89
79
41*
11
9
Practice location (%)
Suburban
Urban
Rural
46
39
15
54
39
6
32
28
39*
49
45
6
Patients aged 6 5 + (%)
1-25%
26-50%
51-75%
76-100%
25
30
24
21
4
20
22
54*
50*
33
13
2
25
33
32
10
Note: GER (geriatrics), GFP (general family practice), GIM (general internal medicine).
*Significantly different from other specialties (p < .01).
includes GFP and GIM physicians with a primary or secondary specialty in geriatrics.
b
Multiple response item.
Vol. 39, No. 4, 1999
419
percentage of their elderly patients who had diagnoses
of depression in the past 12 months, 19% indicated
that 26-50% of their elderly patients were depressed.
the mental health status of their elderly patients. Virtually all respondents (99%) reported interviewing the
patients directly, 83% said that they also interviewed
the patients' family members, and 66% indicated that
they included a medical workup in their assessments.
Only 28% noted that they used formal questionnaires
or scales in their assessments or took into account information obtained by other members of the medical
staff from patients (25%) and their families (21%).
Differences among the three specialties in the use
of assessment procedures were tested for statistical
significance after adjusting for potential confounders
(percentage of patients 654-, practice location, and nursing home practice). We found that the GERs were
significantly more likely than the GFPs (OR = 3.71,
p = .0001) to use questionnaires or scales and to have
staff interview family members to assess their elderly
patients' mental health status. The GIMs were less likely
than the GFPs to use medical workups (OR = .58,
p = .03) and questionnaires or scales (OR = .19,
= .0000) and to have staff interview family memers (OR = .47, p = .02).
Assessing Mental Health Problems
Treating and Referring Patients
With regard to their use of the six assessment techniques examined (see Table 3), the respondents reported using an average of three approaches to assess
The most common approach to treating depression
in elderly patients was to prescribe medications (92%
of the respondents), followed by the use of brief sup-
Table 2. Physicians' Estimation of the Most Prevalent
Mental Health Problems (Percentage)
Affective disorders
Organic brain syndromes
Anxiety disorders*
Multiple conditions3
Other
CER
(n - 49)
CFP
(n = 44)
GIM
(n - 72)
37
39
14
6
4
27
23
43
7
0
42
21
24
13
1
Note: GER (geriatrics), CFP (general family practice), CIM (general
internal medicine). Physicians were asked: "Among the patients
65 and over you see in your practice, what is the most common
mental health problem?"
* Difference between all pairs of specialties are significant at
p < .01.
a
Some physicians provided multiple responses to a singleresponse item.
Table 3. Physicians' Assessment, Intervention, and Referral Patterns With Depressed and Suicidal Patients Aged 65+, by Specialty
(Percentage)
Assessment of mental health status
Interview patient
Interview family members
Medical workup
Questionnaires or scales
Staff interviews patient
Staff interviews family members
Intervention used with depressed patients
Prescribed medication
Brief supportive counseling
Observed and managed without medication
Long-term psychotherapy
Hospitalization
Electroconvulsive therapy
Referrals for treating depression
Prescribed medication
Let referred clinician decide treatment
Long-term psychotherapy
Brief supportive counseling
Hospitalization
Electroconvulsive therapy
Referrals for suicidal patients
Psychiatrist
Hospital emergency room
Crisis assessment team
GER
(n = 50)
CFP
(n = 45)
GIM
(n = 72)
98
88
76
65
35
39
100
80
73
22
27
20
100
82
54
6
18
10
98
88
41
25
33
12
89
87
42
22
7
2
90
75
39
19
10
1
70
31
47
39
51
43
39
39
44
44
30
12
44
58
42
33
22
4
92
82
36
32
21
14
9
85
45
21
13
14
11
47
33
Mental health clinic
14
Psychologist
Social worker
14
10
Note: GER (geriatrics), CFP (general family practice), GIM (general internal medicine). All items allowed multiple responses.
420
The Gerontologist
portive counseling (82% of the respondents). It is important to note that these data do not mean that 92%
and 82% of depressed elderly patients are treated with
medications and brief supportive counseling. Hence,
although 9 of 10 physicians reported using medications to treat depression, 4 of 10 reported observing
and managing depression without medications. This
apparent inconsistency in the data is most likely the
result of physicians describing their approaches with
different patients. In actual practice, most respondents
used a combination of approaches to treat patients.
For example, 64% prescribed medication along with
brief supportive counseling, 2 1 % prescribed medication in conjunction with brief counseling and longterm psychotherapy, and 13% prescribed medications
only. In comparing the three specialties, we found a
similar pattern in their approaches to intervention. However, the GERs were significantly more likely than the
GFPs (OR = 3.41, p = .004) to report using hospital ization.
The respondents were much more likely to refer
depressed elderly patients for long-term psychotherapy,
hospitalization, and electroconvulsive therapy (41%,
33%, and 20%, respectively) than to initiate these treatments themselves (22%, 16%, and 5%, respectively).
Prescribing medications was the most common treatment for which referrals were made—reported by 51%
of the respondents. The GERs were significantly more
likely than the GFPs (OR = 2.02, p = .02), and the
GlMs were significantly less likely than the GFPs
(OR = .58, p = .04), to refer for hospitalization. Likewise, the GERs were significantly more likely than the
GFPs (OR = 3.28, p = .003), and the GlMs were
significantly less likely than the GFPs (OR = .25, p =
.002), to refer for electroconvulsive therapy. Although
the differences between specialties were statistically
significant, it may be useful to point out that only
1 of 45 GFPs and 1 of 72 GlMs actually recommendea
this procedure. The GlMs were nearly twice as likely
as the GERs (OR = 1.80, p = .01) and the GFPs
(OR = 1.80, p = .01) to let the clinicians referred
to decide the course of treatment.
Physicians in the three specialties used similar referral practices with suicidal elderly patients. The majority (86%) would refer these patients to psychiatrists,
followed by a hospital emergency room (43%), a crisis assessment team (27%), a mental health clinic (15%),
psychologists (14%), and social workers (10%). There
were no significant differences by specialty in the types
of referral resources used for suicidal geriatric patients.
With regard to assessing access to suicide methods
(Table 4), we found significant differences by specialty,
with the GERs more likely than the GFPs to ask patients (OR = 2.26, p = .01) about their access to
firearms. The GlMs were half as likely as the GERs
to ask patients (OR = .49, p = .004) or their family
members (OR = .49, p = .005) about access to firearms. Overall, 58% of the physicians reported asking
patients or their families about firearm access. Interestingly, the respondents were more likely to ask their
depressed and suicidal patients about the misuse of
medications {77%) than about the use of firearms as a
suicide method (56%).
Vol.39, No. 4, 1999
421
Table 4. Physicians' Methods of Assessing Suicidality
(Percentage)
Asked patient about
firearms access
Asked family of patient
about firearms access
Assessed patient for
intentional misuse of
medication
GER
(n - 50)
GFP
(n = 45)
GIM
(n = 72)
75
61
39
67
50
30
88
76
71
Note: GER (geriatrics), GFP (general family practice), GIM (general
internal medicine).
Barriers to Providing Mental Health Care
The respondents' perceptions of barriers to providing mental health care were similar across the three
specialties (see Table 5). The two most common obstacles reported by all three specialties were patients'
"unwillingness to seek help" (76%) and "not complying with treatment" (63%), followed by the lack of time
(47%). The GlMs were significantly more likely than
the GERs (OR = 2.04, p = .002) to note the lack of
time.
Other obstacles reported were restrictions on mental health coverage (29%), lack of referral resources
(20%), not the physician's area of expertise (20%), insufficient training in geriatric mental health (18%), lack
of financial reimbursement (15%), and concern about
stigmatizing the patient (8%). Overall, the respondents
more frequently identified health system and patient
issues, rather than physicians' knowledge or attitudes
(e.g., concern about stigmatizing patients), as barriers
to providing mental health care.
The respondents were asked to rate their level of
confidence, on a scale ranging from 1 (not at all) to 5
Table 5. Obstacles to Providing Mental Health Care
to Patients Aged 6 5 + , Cited by Physicians (Percentage)
Patients unwilling to
seek help
Patients not complying
with treatment
Lack of time
Restrictions on mental
health coverage
Lack of referral resources
Not my area of expertise
Insufficient training in
geriatric mental health
Lack of financial
reimbursement for MD
Concern about stigmatizing
patient
GER
(n = 50)
GFP
(n - 45)
GIM
(n = 72)
80
80
72
69
31
66
43
57
60
37
16
8
32
32
18
22
15
28
8
25
19
14
23
10
8
11
7
Note: GER (geriatrics), GFP (general family practice), GIM (general
internal medicine); multiple response item.
(very confident), in four areas of their practice with
elderly patients: diagnosing depression, treating depression, assessing suicidality, and treating suicidality (see
Figure 1). As a group, they were most confident about
diagnosing depression {M = 4.0, SD = 1.1), followed
by treating depression {M = 3.8, SD = 1.0), assessing
suicidality {M = 3.6, SD = .8), and treating suicidality
(M = 2.4, SD = .8).
In examining differences by specialty after adjusting for potential confounders, we found that GERs
were significantly more confident than the other specialties in treating depression (F = 2.93, df = 2, p <
.001). They were also more confident in diagnosing depression and assessing and treating suicidality,
but these differences were not statistically significant.
The GIMs rated themselves as less confident than
the other specialties in assessing suicidality (F = 3.78,
df = 2, p = .026) and in treating suicidality (F =
3.62, df= 2, p = .030).
Discussion
A few points must be acknowledged in interpreting
the results. First, although the respondents were representative of the population of Illinois physicians on
key demographic variables, we have little information
on the nonrespondents, who may have been less inclined to identify or treat the mental health problems
of their geriatric patients or less well trained and less
confident to do so than the respondents. Second, the
respondents may have sought to describe their activi-
ties and level of competence in the most favorable
light. Third, the information on the sample obtained
from the AMA Physician Masterfile (Grumbach et al.,
1995) may not have been precise, given that 15% of
the pilot sample and 12% of the final sample did not
meet the study criteria or could not be located. Fourth,
although the data collected were often categorical, rather
than continuous (to promote ease of administration),
the level of measurement of some variables (e.g., "%
of patients 65 + ") precluded more precise statistical
analyses.
Despite the expectation that the physicians would
differ by specialty in their care of elderly persons with
mental health problems, overall we found more similarities than differences. For example, the GFPs, GIMs,
and GERs were similar in the extent to which they
prescribed medications for depression and used brief
supportive counseling, as well as in their referral patterns for suicidal patients and perceptions of obstacles
to providing mental health care. Other results followed
the expected pattern: the GERs were significantly different in their management of geriatric depression and
suicidality. Significant differences were found among
the specialties in their estimates of the prevalence of
psychiatric disorders and their use of assessment procedures, treatment approaches, and referrals. For example, among the three groups, the GERs were more
confident in diagnosing and treating depression and
suicidality; the GFPs generally ranked second on most
assessment, treatment, and referral measures, and the
GIMs ranked third. In addition, the results regarding
Physicians' confidence in . . .
Very Confident
• GER(n = 50)
DGFP(n = 45)
• GIM(n = 72)
2 -
Not at all
1
diagnosing
depression
treating
depression
assessing
suicidality
treating
suicidality
Figure 1. Physicians' confidence in assessing and treating depression and suicidality, by specialty. Note: GER (geriatrics), GFP (general family practice), GIM (general internal medicine). *Significantly different from other groups (p < .05) after adjusting for % of
patients aged 65 + , practice location, and nursing home practice.
422
The Gerontologist
the most common mental disorders are consistent with
other studies of elderly persons in primary care settings (Pearson, Conwell, & Lyness, 1997).
Despite the recommended protocol of conducting
complete medical workups (Depression Guideline Panel,
1993; Tobias et al., 1992) and using assessment instruments (Lichtenberg et al., 1993; Mclntosh et al.,
1994) with elderly patients who may be depressed or
suicidal, we found wide variations in the use of these
approaches by specialty. For example, only 6% of the
GIMs versus 65% of the GERs reported using questionnaires or scales to assess mental health status. The
most common approach both for treating depression
(92% of the respondents) and for making referrals for
depression (51% of the respondents) was the prescription
of medications. This finding raises questions about the
heavy reliance on psychotropic medications (Semla et
al., 1993; Willcox, Himmelstein, & Woolhandler, 1994)
and the underuse of psychosocial interventions as
approaches to treating geriatric depression. Although
studies have demonstrated the efficacy of individual
and group treatment of geriatric depression (Callahan,
Hendrie, & Tierney, 1996; Mossey, Knott, Higgins, &
Talerico, 1996; Rosen et al., 1997; Scogin & McElreath,
1994), little is known about the reasons why psychosocial interventions are not more widely used in primary care settings.
Although the two most common obstacles to mental health treatment reported by the majority of physicians were "patient unwilling to seek help" and "patient not complying with treatment," the factors that
account for the physicians' perceptions are not clear.
It is likely that several factors play a role. For example,
physicians may initiate discussions about depression,
but patients are resistant, or physicians may hold their
patients responsible for not communicating their symptoms. Another factor could be that if physicians are
rushed, they may not create a climate in which patients feel Tree to discuss their symptoms. Also, patients may think it is natural for them, as elderly people,
to feel depressed. It is also possible that patients and
some physicians are not familiar with the signs of latelife depression or know it is treatable. The fact that
the one barrier that the GIMs were the most likely to
report was the lack of time may partly explain why
the GIMs were less likely to use time-consuming assessment procedures, such as medical workups and
questionnaires or scales. In addition, the finding that
the GIMs had the least confidence of the three groups
in diagnosing and treating depression is supported by
other studies (e.g., Banazak, 1996). However, it is uncertain how the findings on physician confidence relate
to clinical outcomes.
The data also raise questions about the adequacy
of lethality assessment of depressed and suicidal elderly patients by primary care physicians. For example,
despite the fact that firearms are the most common
suicide method among both elderly men and women
(Adamek & Kaplan, 1996a, 1996b; Kaplan, Adamek,
& Johnson, 1994), a sizeable proportion (44%) of respondents reported that they did not ask their
depressed and suicidal elderly patients about their
access to firearms. It is unclear why physicians were
Vol. 39, No. 4, 1999
423
more likely to inquire about the misuse of medications than about firearms as suicide methods. Could it
be a reflection of physicians' reluctance to probe about
guns or a lack of information about the use of firearms among suicidal older adults?
There have been longstanding questions about how
well the mental health needs of geriatric patients are
met (Butler, Lewis, & Sunderland, 1991). According
to Mechanic (1989, p. 55), this uncertainty is due, in
part, to the "lack of knowledge among physicians in
recognizing mental health problems and making appropriate referrals, or attitudes among physicians that
inhibit appropriate care and referral." In light of the
increasing enrollment of elderly people in managed
care plans, there may be an even greater cause for
concern. Primary care physicians have become an important point of entry for both the general health care
and behavioral health care systems and are likely to
play an even greater part as their gatekeeping functions expand, particularly for elderly patients (Lebowitz,
1996). Primary care providers are increasingly expected
to provide mental health care themselves (Lanier &
Clancy, 1996). In this regard, our finding that some
primary care physicians lack confidence in managing
geriatric mental health care prompts two important
questions: Are primary care physicians in managed
health care plans prepared to assume an expanding
role that includes mental health care? And, are elderly patients in increasing jeopardy of not having their
mental health needs met?
Recommendations
Despite calls for increased referrals to a broad range
of mental health specialists with expertise in psychosocial issues (e.g., Badger, Ackerson, Buttell, & Rand,
1997; Waitzkin et al., 1994), 9 out of 10 physicians
in this study referred their suicidal elderly patients to
psychiatrists, and relatively few referred them to psychologists or social workers. Collaboration among members of an interdisciplinary team must be made an
integral part of medical education. Issues that need to
be covered include the roles of ancillary professionals, such as social workers, psychologists, and nurse
practitioners, and how to gain access to them. There
is evidence that the use of diverse personnel could
enhance the role of primary care physicians who treat
elderly persons (Netting & Williams, 1996). Furthermore, the investment in fostering collaborative approaches may lead to more cost-effective interventions
(Katon et al., 1997). The use of other mental health
professionals may also address the concern that psychiatrists often refuse to see Medicare patients due
to low reimbursement (B. McFarland, personal communication, April 23, 1998).
Innovative approaches are needed to improve geriatric mental health care in primary care settings. One
idea is to develop an office-based system to screen
for affective symptoms that could reduce the burden
on physicians (S. Counsell, personal communication,
November 7, 1997). In such a system, a social worker
or nurse practitioner may act as a consultant to the
primary care physician and routinely screen for and
treat mental health problems among the physician's
geriatric patients. Another alternative may be a computerized telephone screening system for mental health
symptoms. Baer and colleagues (1995) developed a
low-cost, confidential method of screening for depression
using computerized digital voice recordings and touchtone responses, and Gonzalez, Costello, La Tourette,
Joyce, and Valenzuela (1997) found support for an
automated voice-interactive program for screening for
depression in English and Spanish. An innovative way
of coordinating the efforts of a primary care physician
and community providers (e.g., adult day care, mental health centers, and other outreach services) serving at-risk elderly persons would be for the physician
to have a social worker or other professional on the
staff who is familiar with community resources.
As a way to improve physicians' care and communication with emotionally distressed elderly patients,
we recommend that medical training in geriatric mental
health be expanded to include a greater emphasis on
psychosocial interventions in conjunction with psychotropic medications. As Leggatt (1996, p. 250) noted,
physicians who are "trained in techniques of psychological assessment, problem-solving and development
of strategies to help better cope with future difficulties, had better outcomes with their patients." Therefore, general medical education and continuing education courses for primary care physicians should include training in interviewing and counseling and
information on educational and self-help programs to
improve physicians' ability to identify and treat mental illness in elderly patients. Furthermore, primary care
physicians could more frequently use available assessment instruments for detecting depression with their
geriatric patients (e.g., the short Geriatric Depression
Scale, the Beck Depression Inventory, or the Hopelessness Scale; Valente, 1997). In addition, increased
public education about geriatric depression may enhance patient-physician cooperation.
We also recommend that primary care physicians
of all specialties conduct firearm-availability histories
with their depressed and suicidal elderly patients (Adamek & Kaplan, 1996a). Although few residency programs offer training in firearm safety counseling (Price,
Bedell, Everett, & Oden, 1997), there is evidence that
patients are amenable to physicians counseling them
about firearms (Price, Clause, & Everett, 1995). Reference guides that are specifically designed to assist physicians in detecting, diagnosing, and treating geriatric
depression and suicidality need to be developed and
widely disseminated ana should include information
on firearms.
Conclusions
The education and training of physicians must
change to meet the mental health needs of the rapidly growing older population. More geriatricians and
physicians who are trained in geriatric mental health
will be needed, particularly if future cohorts of elderly
persons are more likely to seek such care. Furthermore, there must be adequate and more consistent
geriatric mental health training in all primary care spe424
cialties. Research directed at improving primary care
physicians' ability to detect, diagnose, and treat mental illness is urgently needed, especially in light of the
anticipated increase in primary care practice devoted
to elderly people (Stults, 1984). Research is also needed
to assess the extent to which the health care system
affects physicians' ability to provide high-quality mental health care to geriatric patients. For example, little
is known about the organizational and individual factors associated with physicians' use of referral resources
(Lanier & Clancy, 1996). Future studies might investigate whether some of the variability in treatment approaches is due to patient factors such as severity
of illness. It is also important to know when and
why elderly patients do or do not raise mental health
issues in their primary care visits.
Primary care physicians can play a critical role in
suicide prevention (Murphy, 1986). One of the targets of trie Healthy People 2000 initiative is to reduce
the suicide rate among older White men by 8% to
39.2 per 100,000 (National Center for Health Statistics [NCHS], 1994). A related U.S. Public Health Service goal is to increase the proportion of primary care
providers who routinely review their patients' cognitive and emotional functioning to 60% (NCHS, 1996).
Given that primary care physicians are the most frequently visited health care providers for elderly persons, their first point of contact, and the de facto mental
health care system (Cohen & Cairl, 1996), we believe
that these physicians are in a unique position to help
achieve these national goals.
References
Adamek, M. E., & Kaplan, M. S. (1996a). Firearm suicide among older
men. Psychiatric Services, 47, 304-306.
Adamek, M. E., & Kaplan, M. S. (1996b). The growing use of firearms by
suicidal older women, 1979-1992: A research note. Suicide and LifeThreatening Behavior, 26, 71-78.
Badger, L. W., Ackerson, B., Buttell, F., & Rand, E. H. (1997). The case
for integration of social work psychosocial services into rural primary
care practice. Health and Social Work, 22, 20-29.
Badger, L. W., DeGruy, F. V., Hartman, J., Plant, M. A., Leeper, J., Anderson,
R., Ficken, R., Gaskins, S., Maxwell, A., Rand, E., & Tietze, P. (1994).
Patient presentation, interview content, and the detection of depression by primary care physicians. Psychosomatic Medicine, 56, 128135.
Baer, L, Jacobs, D. G., Cukor, P., O'Laughlen, J., Coyle, J., & Magruder,
K. (1995). Automated telephone screening survey for depression. Journal
of the American Medical Association, 273, 1943-1944.
Banazak, D. A. (1996). Late-life depression in primary care. How well are
we doing? Journal of General Internal Medicine, 11, 163-167.
Bell, J. R. (1997). Underdiagnosis of depression in primary care: By accident or design? [letter]. Journal of the American Medical Association,
277, 1433.
Butler, R., Lewis, M., & Sunderland, T. (1991). Aging and mental health:
Positive psychosocial and biomedical approaches. New York: Merrill.
Callahan, C. M., Hendrie, H. C , Dittus, R. S., grater, D. C , Hui, S. L., &
Tierney, W. M. (1994). Improving treatment of late life depression in
primary care: A randomized clinical trial. Formal of the American Geriatrics
Society, 42, 839-846.
Callahan, C. M., Hendrie, H. C , & Tierney, W. M. (1996). The recognition and treatment of late-life depression: A view from primary care.
International Journal of Psychiatry and Medicine, 26, 155-171.
Callahan, C. M., Hendrie, H. C , Nienaber, N. A., & Tierney, W. M.
(1996). Suicidal ideation among older primary care patients. Journal
of the American Geriatrics Society, 44, 1205-1209.
Cassata, D. M., & Kirkman-Liff, B. L. (1981). Mental health activities of
family physicians. Journal Family Practice, 12, 683-692.
Cohen, D., & Cairl, R. (1996). Mental health care policy in an aging
society. In B. L. Levin & J. Petrila (Eds.), Mental health services: A
public health perspective (pp. 301-324). New York: Oxford University Press.
The Gerontologist
Conwell, Y., Olsen, K., Caine, E., & Flannery, C. (1991). Suicide in late
life: Psychological autopsy findings. International Psychogeriatrics, 3,
56-66.
Cooper-Patrick, L, Crum, R. M., & Ford, D. E. (1994). Identifying suicidal
ideation in general medical practice. Journal of the American Medical
Association, 272, 1757-1762.
Courage, M. M., Codbey, K. L, Ingram, D. A., Schramm, L. L, & Hale,
W. E. (1993). Suicide in the elderly: Staying in control. Journal of
Psychosocial Nursing and Mental Health Services, 31, 26-31.
Depression Guideline Panel. (1993). Depression in primary care: Detection, diagnosis, and treatment (AHCPR Publication No. 93-0552). Rockville,
MD: U.S. Department of Health and Human Services.
Directory of medical specialists. (1996). Chicago: Marquis Who's Who.
Eisenberg, L. (1992). Treating depression and anxiety in the primary care
setting. New England Journal of Medicine, 326, 1080-1083.
Frank, E., Rothenberg, R., Brown, W. V., & Maibach, H. (1997). Basic
demographic and professional characteristics of Cognomen physicians.
Western Journal of Medicine, 166, 179-184.
Frierson, R. L. (1991). Suicide attempts by the old and the very old. Archives of Internal Medicine, 151, 141-144.
German, P. S., Shapiro, S., Skinner, E. A., Von Korff, M., Klein, L. E.,
Turner, R. W., Teitelbaum, M. L, Burke, J., & Bums, B. J. (1987).
Detection and management of mental health problems of older patients by primary care providers. Journal of the American Medical Association, 257, 489-493.
Glasser, M., & Gravdal, J. A. (1997). Assessment and treatment of geriatric depression in primary care settings. Archives of Family Medicine,
6, 433-438.
Gonzalez, G. M., Costello, C. R., La Tourette, T. R., Joyce, L. K., & Valenzuela,
M. (1997). Bilingual telephone-assisted computerized speech-recognition assessment: Is a voice-activated computer program a culturally
and linguistically appropriate tool for screening depression in English
and Spanish? Cultural Diversity and Mental Health, 3, 93-111.
Greene, M. G., & Adelman, R. D. (1996). Psychosocial factors in older
patients' medical encounters. Research on Aging, 18, 84-102.
Grumbach, K., Becker, S. H., Osborn, E. M. H., & Bindman, A. B. (1995).
The challenges of defining and counting generalist physicians: An analysis
of Physician Masterfile data. American Journal of Public Health, 85,
1402-1407.
Henderson, A. S. (1990). The social psychiatry of later life. British Journal
of Psychiatry, 56, 645-653.
Kamerow, D. B., Pincus, H. A., & MacDonald, D. I. (1986). Alcohol abuse,
other drug abuse, and mental disorders in medical practice: Prevalence, costs, recognition, and treatment. Journal of the American Medical
Association, 255, 2054-2057.
Kaplan, M. S., Adamek, M. E., & Johnson, S. (1994). Trends in firearm
suicide among older American males: 1979-1988. The Cerontologist,
34, 59-65.
Katon, W., Von Korff, M. E., Lin, E., Bush, T., & Ormel, J. (1992). Adequacy and duration of antidepressant treatment in primary care. Medical
Care, 30, 67-76.
Katon, W., Von Korff, M., Lin, E., Simon, G., Walker, E., Bush, T., &
Ludman, E. (1997). Collaborative management to achieve depression
treatment guidelines. Journal of Clinical Psychiatry, 58, 20-23.
Loonier, D. C , & Clancy, C. M. (1996). The changing interface of primary and specialty care. Journal of Family Practice, 42, 303-305.
Lazarus, A. (1995). The role of primary care physicians in managed mental health care. Psychiatric Services, 46, 343-345.
Lebowitz, B. D. (1996). Community services for elderly people. Current
Opinion Psychiatry, 9, 294-297.
Leggatt, M. (1996). Prevention of anxiety and depression in vulnerable
groups by Joanna Murray [book review]. Health Promotion International, 11, 249-250.
Lichtenberg, P. A., Gibbons, T. A., Manna, M., & Blumenthal, F. (1993).
Physician detection of depression in medically ill elderly. Clinical Cerontologist, 13, 81-90.
Lin, E., Van Korff, M. E., & Wagner, E. H. (1989). Identifying suicide
potential in primary care. Journal of General Internal Medicine, 4, 1 6.
Mclntosh, J. L., Santos, J. F., Hubbard, R. W., & Overholser, J. C. (1994).
Elder suicide: Research, theory, and treatment. Washington, DC: American
Psychological Association.
Mechanic, D. (1989). Mental health and social policy. Englewood Cliffs,
NJ: Prentice Hall.
Morrow-Howell, N., & Proctor, E. K. (1992). The use of logistic regression
in social work research. In D. F. Gillespie & C. Glisson (Eds.), Quantitative methods in social work: State of the art (pp. 87-104). New
York: Haworth Press.
Mort, E. A., Edwards, J. N., Emmons, D. W., Covery, K., & Blumenthal,
D. (1996). Physician response to patient insurance status and ambulatory care. Medical Care, 34, 783-793.
Mossey, J. M., Knott, K. A., Higgins, M., & Talerico, K. (1996). Effective-
Vol. 39, No. 4, 1999
425
ness of a psychosocial intervention, interpersonal counseling, for subdysthymic depression in medically ill elderly. Journal of Gerontology:
Medical Sciences, 51, 172-178.
Murphy, G. E. (1986). The physician's role in suicide prevention. In
A. Roy (Ed.), Suicide (pp. 171-179). Baltimore: Williams & Wilkins.
National Center for Health Statistics. (1994). Healthy people 2000: National health promotion and disease prevention objectives (DHHS publication PHS 94-1232-1). Hyattsville, MD: U.S. Public Health Service.
National Center for Health Statistics. (1996, July). Healthy people 2000
Progress review: Mental health and mental disorders. Hyattsville, MD:
U.S. Public Health Service.
Netting, F. E., & Williams, F. G. (1996). Case manager-physician collaboration: Implications for professional identity, roles, and relationships.
Health and Social Work, 21, 216-224.
Olfson, M., Weissman, M. M., Leon, A. C , Edmund S., Higgins, E. S.,
Barrett, J. E., & Blacklow, R. S. (1995). Psychological management by
family physicians. Journal of Family Practice, 41, 543-550.
Orleans, C. T., George, L. K., Houpt, J. L., & Brodie, K. H. (1985). How
primary care physicians treat psychiatric disorders: A national survey
of family practitioners. American Journal of Psychiatry, 142, 52-57.
Pearson, J. L., Conwell, Y., & Lyness, J. M. (1997). Late life suicide and
depression in the primary care setting. New Directions in Mental Health
Services, 76, 13-38.
Price, J. H., Bedell, A. W., Everett, S. A., & Oden, L. (1997). Training in
firearm safety counseling in family practice residency programs. Journal of Community Health, 22, 91-99.
Price, J. H., Clause, M., & Everett, S. A. (1995). Patients' attitudes about
the role of physicians in counseling about firearms. Patient Education
and Counseling, 25, 163-170.
Rapp, S. R., & Davis, K. M. (1989). Geriatric depression: Physicians' knowledge, perceptions, and diagnostic practices. The Gerontologist, 29, 252257.
Rogers, W. H., Wells, K. B., Meredith, L. S., Sturm, R., & Burnam, A.
(1993). Outcomes for adult outpatients with depression under prepaid or fee-for-service. Archives of General Psychiatry, 50, 517-525.
Rosen, J., Rogers, J. C , Marin, R. S., Mulsant, B. H., Shanar, A., Reynolds
III, C. F. (1997). Control-relevant intervention in the treatment of minor and major depression in a long-term care facility. American Journal of Geriatric Psychiatry, 5, 247-257.
Scogin, F., & McElreath, L. (1994). Efficacy of psychosocial treatments for
geriatric depression: A quantitative review. Journal of Consulting and
Clinical Psychology, 62, 69-74.
Semla, T. P., Schwartz, A., Koch, H., & Nelson, C. (1993). Patterns of
drug prescribing. In J. F. Van Nostrand, S. E. Furner, & R. Suzman
(Eds.), Health data on older Americans: United States, 1992, DHHS
Publication PHS 93-1411, pp. 187-193. Washington, DC: U.S. Government Printing Office.
Solomon, R. (1996). Coping with stress: A physicians' guide to mental
health in aging. Geriatrics, 51, 4 6 - 5 1 .
Spitzer, R. L., Williams, J. B., Kroenke, K., Linzer, M., deGruy, F. V.,
Hahn, S. R., Brady, D., & Johnson, J. G. (1994). Utility of a new
procedure for diagnosing mental disorders in primary care. Journal of
the American Medical Association, 272, 1749-1756.
Stults, B. M. (1984). Preventive health care for the elderly. Western Journal of Medicine, 141, 832-845.
Tobias, C. R., Pary, R., & Lippmann, S. (1992). Preventing suicide in
older people. American Family Physician, 45, 1707-1713.
Uniitzer, J., Patrick, D. L., Simon, G., Gremeowski, D., Walker, E., Rutter,
C , & Katon, W. (1997). Depressive symptoms and the cost of health
services in HMO patients aged 65 years and older: A 4-year prospective study. Journal of the American Medical Association, 277, 16181623.
Valente, S. M. (1997). Preventing suicide among elderly people. The American
Journal of Nurse Practitioners, 1, 15-31.
Waitzkin, H., Britt, T., & Williams, C. (1994). Narratives of aging and
social problems in medical encounters with older persons. Journal of
Health and Social Behavior, 35, 322-348.
Waxman, H. M., & Camer, E. A. (1984). Physicians' recognition, diagnosis, and treatment of mental disorders in elderly medical patients. The
Gerontologist, 24, 593-597.
Wells, K. B., Hays, R. D., Burnam, A., Greenfield, S., & Ware Jr., J. E.
(1989). Detection of depressive disorders for patients receiving prepaid for fee-for-service care: Results from the medical outcome study.
Journal of the American Medical Association, 262, 3298-3302.
Wells, K. B., Sturm, R., Sherbourne, C. D., & Meredith, L. S. (1996).
Caring for depression. Cambridge, MA: Harvard University Press.
Willcox, S. M., Himmelstein, D. U., & Woolhandler, S. (1994). Inappropriate drug prescribing for the community-dwelling elderly. Journal of
the American Medical Association, 272, 292-296.
Received July 30, 1998
Accepted April 9, 1999