Referral Patterns and Recognition of DepressionAmong African-American and Caucasian Patients

Raphael J. Leo, M.D., Constance Sherry, R.N., M.S., and Amy W. Jones

Abstract: A retrospective review of psychiatric consultationswas conducted for African-American and Caucasian patientsfor a 2-year period. Reasons for referral, assigned diagnoses,accuracy rates, and discordance and concordance rates wereassessed. Referrals for depression comprised 24.6% of all con-sults for Caucasian and African-American inpatients. Only40.3% of patients referred for depression were diagnosed witha depressive disorder; 54.4% of patients diagnosed with depres-sive disorders were referred for other reasons. African-American patients were referred for evaluation of depressionand diagnosed with depressive disorders significantly less oftenthan Caucasian patients. No significant differences were ob-tained between African-Americans and Caucasians in the ac-curacy rates of patients referred for depression. Discordanceand concordance rates for the two groups were comparable.Diagnoses assigned to African-Americans and Caucasians in-correctly referred for depression did not differ significantly. Fordepressed African-Americans and Caucasians referred for rea-sons other than depression, the only difference noted was in thereferral rates for adjustment of psychotropics. The nonpsychi-atric staff fails to recognize depression and often refer depressedpatients inappropriately. Depressed patients are primarily re-ferred for suicide assessment and disruptive behaviors. Refer-rals for depression may be a secondary concern to nonpsychi-atric staff. In addition, cultural variables and racial differencesbetween hospital staff and patients may account for the differ-ences in referral patterns. Awareness of the needs of African-American patients is required. © 1998 Elsevier Science Inc.

Introduction

In the medically ill, depression is common as anaffective response to illness and hospitalization, asymptom of medical illness, and as a clinical syn-

drome [1]. Depression rates among hospitalizedmedically ill patients have been estimated to bebetween 22% and 32% [2–4]. Nonetheless, referralrates to psychiatric consultation-liaison (C-L) ser-vices for medical and surgical inpatients with de-pression remain quite low [5,6].

Depression is associated with higher morbidityand mortality rates among patients with medicalillnesses [7–11]. The presence of comorbid depres-sion among medically ill inpatients results in am-plification of somatic symptoms, increased func-tional disability, and decreased adherence tomedical regimens [12–14]. Hence, recognition andtreatment of depression among medically ill pa-tients may reduce patient distress, improve treat-ment compliance, and reduce morbidity and mor-tality associated with medical illnesses.

Unfortunately, depression often goes unrecog-nized and untreated in medical settings [15–19].Reasons for this include poor recognition of symp-toms, inadequate time spent evaluating patients,and inadequate trials of medications [16,19–22]. Inaddition, patient characteristics, such as ethnic andcultural factors, may impede the recognition of de-pression [19,23,24]. Research in a wide variety ofcultures and among ethnic groups in the UnitedStates suggests that cultural beliefs and practicesaffect the manner in which individuals both expressor manifest symptoms of depression, and experi-ence depression as a phenomenal state [25–27]. Therelationship between depressive symptoms and pa-tient variables, e.g., age, gender, and race have beenexamined in several studies, although few studieshave examined the relationship between ethnicityand depression in the patients referred to C-L ser-vices [28]. C-L psychiatry is uniquely positioned toevaluate cultural influences on psychiatric illnesses,e.g., the manner in which depression is manifested,

Department of Psychiatry, School of Medicine and BiomedicalSciences, State University of New York, Buffalo, New York

Address reprint requests to: Raphael J. Leo, M.D., Departmentof Psychiatry, School of Medicine and Biomedical Sciences, StateUniversity of New York, Erie County Medical Center, 462 GriderStreet, Buffalo, New York 14215

General Hospital Psychiatry 20, 175–182, 1998 175© 1998 Elsevier Science Inc. All rights reserved. ISSN 0163-8343/98/$19.00655 Avenue of the Americas, New York, NY 10010 PII S0163-8343(98)00019-X

beliefs about treatment, response to treatment, andrecovery issues [29].

Given that ethnic variables impact on the mannerin which symptoms of depression are experiencedand/or presented, or influence the degree to whichhelp is sought, African-American patients may beat risk for underrecognition of affective disorders,as compared with Caucasian patients [30–32]. In aretrospective review of C-L consult requests in ageneral hospital, African-American inpatients werereferred for evaluation of depression and suicideless often than were Caucasian inpatients [33]. In areview of C-L referrals among geriatric medicalinpatients, Caucasians were referred for psychiatricconsultation for mood disorders, specifically de-pression, more often than African-American pa-tients; and significantly less often than African-Americans for psychosis [34]. This pilot studyexamines rates of referral for depression and ratesof diagnosis of depressive disorders amongAfrican-American and Caucasian patients of allages referred to the C-L service in our facility.

Methods

This study was conducted at the Erie County Med-ical Center (ECMC) in Buffalo, New York, a 600-bedtertiary-care hospital, including a skilled nursingfacility. A retrospective review of psychiatric con-sultations completed for patients during a consec-utive 24-month period from January 1, 1994 to De-cember 31, 1995 was undertaken. The consults onfile with the C-L service were abstracted and codedfor analysis. Consultations included in the studywere those completed for patients referred for as-sessment of depression or depressive symptoms orwhich indicated that a diagnosis of a depressivedisorder was assigned to the patient. Depressivedisorders consisted of major depression, dysthymicdisorder, depressive disorder NOS, depression dueto general medical conditions or medication use,adjustment disorder with depressed mood, adjust-ment disorder with depressed and anxious mood,bipolar I disorder—depressed, and dementia withdepressive features. Data abstracted for analysisincluded patient demographic information, reasonfor consultation request (if other than for assess-ment of depression), and psychiatric diagnoses as-signed at the time of consultation. Chi-square anal-yses and Fisher’s Exact tests were performed todetermine whether significant racial differences ex-ist. DSM-III-R diagnoses were assigned during thestudy period in 1994 and DSM-IV diagnostic crite-

ria were assigned during the study interval in 1995.For study purposes, all diagnoses were convertedto DSM-IV standards.

Results

A total of 21,740 patients (7841 African-Americansand 12,678 Caucasians) were admitted to nonpsy-chiatric services at ECMC during the 2 years of thisstudy. Psychiatric consultations were provided for1523 patients1, 410 (26.9%) African-American pa-tients [males 210 (51.2%); females 200 (48.8%)] and1039 (68.2%) Caucasian patients [males 561 (54%);females 478 (46%)]. Significantly more Caucasianpatients (8.2%, 1039/12,678) than African-Americanpatients (5.2%, 410/7841) were referred for psychi-atric consultation, x2 5 64.5, df 5 1, p , 0.001.

Depression as a Reason for Referral

Patients referred for the assessment of depressionincluded 357 Caucasian and African-American pa-tients, representing 1.6% of all hospitalized pa-tients, and 24.6% of all African-American and Cau-casian patients referred for consultation.Significantly fewer African-American patients(1.1%, 84/7841) were referred for depression thanCaucasian patients (2.2%, 273/12,678), x2 5 32.55,df 5 1, p , 0.001. Females and males referred forassessment of depression were 175 (49%) and 182(51%), respectively, which did not differ signifi-cantly. Regarding age, Caucasian patients (mean 551.7 6 18.6 years) referred for depression weresignificantly older than African-American patients(mean 5 42.1 6 16.4 years), t(355) 5 4.3, p ,0.001.

Rates of Diagnosis of Depression

Three hundred sixteen (21.8%, 316/1449) Caucasianand African-American inpatients were diagnosedwith a depressive disorder by the C-L service, withsignificantly fewer African-Americans (18.3%, 75/410) than Caucasians (23.2%, 241/1039), x2 5 3.86,df 5 1, p , 0.05. Females and males diagnosedwith depressive disorders were 165 (52.2%) and 151(47.8%), respectively, which did not differ signifi-cantly. Caucasian patients (mean 5 49.4 6 18.4years) diagnosed with depressive disorders were

1 This value includes 61 Hispanic, 7 Native-American and 6Asian-American patients. Due to the small sample sizes, thesegroups were excluded from the present study.

R. Leo et al.

176

significantly older than African-American patients(mean 5 35.3 6 12.5 years), t(314) 5 6.2, p , 001.

Accuracy Rates of Patients Referred forDepression

Among the 357 patients referred for depression, 144(40.3%) were diagnosed with a depressive disorder(see Table 1). Of the patients referred for depressionand ultimately diagnosed with a depressive disor-der, 43.2% (118/273) were Caucasian and 31% (26/84) were African-American. The accuracy rateswere not significantly different ( p 5 0.06) betweenAfrican-American and Caucasian patients.

Discordance and Concordance Rates

For the 316 African-American and Caucasian pa-tients diagnosed with depressive disorders by theC-L team, 172 patients (54.4%) were referred to theC-L service for reasons other than assessment ofdepression (false negatives). In addition, 213 pa-tients referred for depression were diagnosed withsome other disorders (false positives). It was possi-ble, therefore, to calculate discordance rates, i.e., thesum of false positives and false negatives dividedby the total sample size for the individual groups.These are summarized in Table 1.

The discordance rate for all the subjects was26.6%; for African-Americans 26.1%; and for Cau-casians 26.8%. The discordance and concordancerates were comparable among African-Americansand Caucasians.

Reasons for Referral

Among patients diagnosed with depressive disor-ders, 49% (118/241) of Caucasians and 34.7% (26/75) of African-Americans were referred for assess-ment of depression. Significantly more depressedCaucasians were referred for a suspected depres-sion than were depressed African-Americans, x2 54.15, df 5 1, p , 0.05.

The reasons for referral for patients ultimatelydiagnosed as depressed but referred for reasonsother than depression are summarized in Table 2.More than one reason for referral may have beenapplied to a patient. Of the false negative referrals,the majority were referred for suicide assessment(68.2%). Comparisons among African-Americanand Caucasian patients revealed that only rates ofreferrals for adjustment of psychotropics were sig-nificantly different between the two groups, Fish-er’s Exact test2, p , 0.05.

Diagnoses Assigned to Patients Referred forDepression (False Positives)

The diagnoses assigned to patients incorrectly re-ferred for depression and ultimately diagnosedwith another psychiatric condition are summarizedin Table 3. Over one-half of the patients referred forsuspected depression were diagnosed with sub-stance use disorders, other adjustment disorders,and bereavement. No significant differences were

2 Fisher’s Exact tests were performed when sample sizes weretoo small to allow for Chi-square analyses.

Table 1. Frequencies of patients referred for depression and diagnosed with depressive disorders

Referred forTotals

(%)False

positivesFalse

negatives Discordance/ConcordanceDepression (%) Other reasons (%)

Entire sample 213 172 26.6%/73.4%Depressed 144 (40.3) 172 (15.8) 316 (21.8)Not depressed 213 (59.7) 920 (84.2) 1133 (78.2)Total 357 (100) 1092 (100) 1449 (100)

African-Americans 58 49 26.1%/73.9%Depressed 26 (31) 49 (15) 75 (18.3)Not depressed 58 (69) 277 (85) 335 (81.7)Total 84 (100) 326 (100) 410 (100)

Caucasians 155 123 26.8%/73.2%Depressed 118 (43.2) 123 (16.1) 241 (23.2)Not depressed 155 (56.8) 643 (83.9) 798 (76.8)Total 273 (100) 766 (100) 1039 (100)

Referrals for and Recognition of Depression

177

observed in the frequencies of diagnoses assignedto African-American and Caucasian patients re-ferred for depression but who were found to haveanother primary psychiatric condition.

Discussion

General Trends

Nearly 25% of all Caucasian and African-Americanpatients referred to the C-L service in the presentstudy were referred for assessment of depression.This rate of referral is comparable to rates reportedpreviously [3,35,36]. If assessments of accuracy, dis-cordance, and concordance are assumed to reflecthow well nonpsychiatric staff recognize depressionamong inpatients, then we can say that they do sopoorly.

Remarkably, 60% of patients referred for a sus-pected depression were not diagnosed with a de-pressive disorder (false positives). Previous inves-tigations have suggested that nonpsychiatric staffreferring patients for depression were inaccurate in40%–78% of cases [35,37]. A tendency for patients topresent with vague complaints, to somatize, or toexpress psychosocial stressors has led to patientsbeing mislabeled as depressed [15]. The nonpsychi-atric staff may have been misled by somatic symp-toms of medical illnesses, e.g., fatigue, sleep andappetite disturbances, etc., which can mimic symp-toms of depression [4,38–40]. However, there is

concern when the presumption of depression onthe part of nonpsychiatric staff can delay evaluationand treatment of other conditions, some of whichmay be life-threatening [35,37]. Thus, as can be seenin Table 3, conditions requiring different treatmentapproaches, e.g., delirium or alcohol/drug with-drawal, may be delayed or mismanaged becausethe nonpsychiatric staff suspected depression in-stead.

Among patients diagnosed with depressive dis-orders, over 54% were referred for reasons otherthan the evaluation/assessment of depression (falsenegatives). It is conceivable that nonpsychiatricstaff may recognize that a patient is distressed ordisplaying a change in behaviors. Due to the har-ried pace within the hospital and with insufficienttime to gather patient data, the patient may be

Table 2. Reasons for referral of patientsdiagnosed with depression(false negatives)

African-Americansa

(%)Caucasiansb

(%) Total (%)

Suicide assessment 36 (69.2) 86 (67.7) 122 (68.2)Behavior problems 7 (13.5) 10 (7.9) 17 (9.5)Anxiety 5 (9.6) 7 (5.5) 12 (6.7)Adjustment of

psychotropics 0 (0) 10 (7.9) 10 (5.6)Psychosis 0 (0) 7 (5.5) 7 (3.9)Capacity 2 (3.85) 4 (3.1) 6 (3.4)Confusion 2 (3.85) 1 (0.8) 3 (1.7)Transfer to

psychiatry 0 (0) 2 (1.6) 2 (1.1)Total 52 (100) 127 (100) 179 (100)

a Three patients had two reasons for referral.b Four patients had two reasons for referral.

Table 3. Diagnoses assigned to patients referredfor depression (false positives)

African-Americans

(%)Caucasians

(%) Total (%)

Substance usedisorders 16 (27.6) 41 (26.5) 57 (26.7)

Adjustmentdisorders 12 (20.7) 24 (15.5) 36 (16.9)

Bereavement 10 (17.2) 24 (15.5) 34 (16)Dementia 7 (12.1) 17 (11) 24 (11.3)Delirium 4 (6.9) 15 (9.7) 19 (8.9)No Axis I

disorder 3 (5.2) 9 (5.8) 12 (5.6)Schizophrenia

and otherpsychoses 5 (8.6) 7 (4.5) 12 (5.6)

Anxietydisorders 0 (0) 7 (4.5) 7 (3.3)

Bipolardisorder 1 (1.7) 2 (1.3) 3 (1.4)

MentaldisorderNOS 0 (0) 3 (1.9) 3 (1.4)

Sleepdisorders 0 (0) 2 (1.3) 2 (0.9)

Psychosomaticdisorders 0 (0) 2 (1.3) 2 (0.9)

Pain disorder 0 (0) 1 (0.6) 1 (0.5)Medication-

inducedmovementdisorder 0 (0) 1 (0.6) 1 (0.5)

Total 58 (100) 155 (100) 213 (100)

R. Leo et al.

178

referred to the C-L service for a variety of reasonsother than depression. Yet, for other patients notedby the nonpsychiatric staff to display depressivesymptoms, attributions that these symptoms are anexpected reaction to illness or hospitalization mayinterfere with appropriate referral to the C-L ser-vice [41].

Review of the reasons for referral of the falsenegatives (see Table 2) indicates that most de-pressed patients were referred for issues pertainingto “management” or treatment issues in the hospi-tal. Concerns about patient safety in the hospital(i.e., referrals for suicide assessment or confusion)or patient cooperation with treatment (i.e., referralsfor behavior problems and capacity) though impor-tant, may have been more of a concern to nonpsy-chiatric staff than diagnosis or treatment of depres-sion. There is a relative constancy in the types ofconsultation requests over time. Previous studieshave likewise demonstrated that referring servicesfocus on behavior problems/disruptive behaviorsand disposition issues rather than identifying andtreating underlying psychiatric disorders [36,42].

Racial Differences

Regarding race, African-American patients were re-ferred for depression significantly less often thanCaucasian patients. If the manner in which symp-toms of depression are expressed by African-American patients differs from what health careproviders expect, the staff will be less likely torecognize these symptoms of depression[25,32,43,44]; hence, referral rates will be reduced.African-Americans may be less likely to seek pro-fessional help for depression than are Caucasians[45–47]. Elder African-Americans may deny symp-toms of distress and may decline psychiatric eval-uation if those symptoms are not experienced asinterfering with activities of daily living or recoveryfrom medical illness [48]. Whether this trend existsamong younger African-Americans has, as yet, tobe clarified. Sussman et al. [47] suggest that fearsamong African-Americans about treatment, partic-ularly fears of hospitalization, may prevent themfrom seeking professional help. In the nonpsychi-atric inpatient setting, that fear may be translatedinto a concern over a prolonged hospitalization,and consequently a prolonged separation fromfamily and other social supports. Finally, the per-ception of the availability and/or efficacy of treat-ment for depressive symptoms may impact on re-ferral. African-Americans tend to rely on social

supports and are less likely than Caucasians to seekout professional assistance under times of duress[45,47].

Significantly fewer African-American patientswere diagnosed with depressive disorders. Previ-ous literature has suggested that bias impacts uponpsychiatric diagnoses assigned to African-American and Caucasian patients [30]. In general,mood disorders may be overlooked and psychoticdisorders overdiagnosed in African-Americans,particularly if the diagnostician is of a different race[23,25,30,32,34,44,45,49–51]. Nonetheless, recent re-search suggests that rates of depression amongAfrican-Americans and Caucasians were compara-ble when demographic, sociocultural, and socioeco-nomic factors were controlled [52–54].

Concerns are raised that depression in African-Americans can be missed or overlooked by nonpsy-chiatric staff. Depressed Caucasian patients werereferred to the C-L service for depression signifi-cantly more often than depressed African-American patients. African-Americans diagnosedwith depressive disorders come to the attention ofthe C-L service less often by referrals for assessmentof depression and more often for other reasons.

Patient age may have impacted upon the referralrates for depression as well as the rates of diagnosisof depressive disorders in our samples. Caucasianpatients referred for depression and those diag-nosed with depressive disorders were significantlyolder than African-Americans. With advanced age,elder medically ill Caucasians may have experi-enced greater disability, and presumably, greaterdistress than younger African-American patients,prompting higher rates of referral as well as diag-nosis of depressive disorders.

In a previous study, we examined all psychiatricreferrals made for geriatric African-American andCaucasian patients. African-American elders werereferred for psychosis significantly more often thanCaucasian patients, and were referred less often formood disorders, particularly depression [34]. Con-sequently, we expected to find that in the presentstudy, fewer African-Americans than Caucasianswere referred for depression and that depressedAfrican-Americans would be referred for a varietyof reasons other than for a suspected depression.Furthermore, we expected that accuracy and con-cordance rates would be lower for African-Americans referred for depression as comparedwith Caucasians. Indeed, accuracy rates forAfrican-Americans referred for depression werelower, albeit not statistically significantly different

Referrals for and Recognition of Depression

179

from those of Caucasian patients referred for de-pression. Concordance rates were surprisinglyhigher for African-Americans than Caucasians.

It is possible that given the current climate ofmanaged care with its emphasis on expeditiousdischarge, patients are referred primarily for “man-agement” issues. Thus, behaviors that raise con-cerns about a psychiatric problem and that mightdelay treatment or discharge, would prompt psy-chiatric referral. It is conceivable that such issuesarose with Caucasian patients more often than theydid with African-Americans and that this ac-counted for the disparities noted here. Overall, sig-nificantly more Caucasian patients were referred tothe C-L service than were African-Americans. Onthe other hand, when compared for African-Americans and Caucasians, the reasons for referralof the false negatives did not differ significantly,with the exception of referrals for adjustment ofpsychotropic medications. The reasons for the dis-parity in the referrals for medication adjustmentcannot be determined at the present time.

Nonpsychiatric staff appear to detect depressionpoorly, inappropriately suspecting it in a majorityof patients referred for depression and failing torecognize it in a majority of patients referred forother reasons. Therefore, there are concerns thatsymptoms of depression may be overlooked or mis-interpreted. Racial disparities in symptom expres-sion of depression, help-seeking for depression,and the potential impact of racial biases complicatethe issue. There is a risk that African-Americans,and patients of other nonwhite racial groups, maybe particularly vulnerable to underreferral for andtreatment of depression in the inpatient setting.

Unfortunately, the retrospective nature of thisstudy does not permit cross validation of the diag-noses assigned to patients. Standardized scales orstructured interviews were not employed and mea-sures of interrater reliability were not possible. Inaddition, even though a reason for a psychiatricreferral is obtained at the time of consult request, itmay be possible that little thought went into thereported reasons for referral. Consequently, wecannot always be certain of the validity of the doc-umented reasons for referral in this study. As to theissue of racial differences in patterns of referral, weare unable by means of a chart review to determinewhat factors determine whether a psychiatric con-sult is requested in general, and whether these fac-tors differ for African-Americans and Caucasiansspecifically. Additionally, comparisons of African-American and Caucasian subjects matched for age,

gender, and severity of illness were not possible.Larger prospective studies, with comparablymatched subjects, would be necessary to increasethe generalizability of this data.

Conclusions

With increasing medical comorbidity, the severityof depression likewise increases [3,12,55,56].African-Americans have more medical problems(e.g., hypertension, coronary artery disease, cere-brovascular accidents, diabetes mellitus) and com-plications from those disorders than Caucasians[57–59]. Disparities between the two groups alsoinclude access to health care resources [60–65].Hence, it is anticipated that hospitalized African-Americans may present with increased medical dis-ability. As poor health is a powerful risk factor fordepression among African-Americans [66], the fac-tors mentioned above, when taken together, willlikely increase the risk of depression among thisgroup. Our findings suggest African-Americansmay be particularly vulnerable to being overlookedfor the referral, diagnosis, and treatment of depres-sion. This is especially important as the African-American population is projected to increase morethan twice the Caucasian population’s annual rateof growth in the next 50 years [67]. Heightenedawareness of comorbid depression, arising duringor exacerbated by hospitalization, is needed. Rec-ognition and treatment of comorbid depressionamong the medically ill can reduce patient distress,improve treatment compliance, and reduce morbid-ity and mortality associated with those medicalconditions. The psychiatrist, working collabora-tively with primary care physicians and physiciansin other specialties, can improve the care renderedto depressed patients with comorbid medical con-ditions and reduce readmission/recurrence rates[68].

The authors thank Jennifer Batterman-Faunce, Ph.D. and DonaldPollock, Ph.D. for their suggestions in the preparation of this manu-script.

References

1. Rodin G, Voshart K: Depression in the medically ill:an overview. Am J Psychiatry 143:696–705, 1986

2. Cavanaugh S: The prevalence of emotional and cog-nitive dysfunction in a general medical population:using the MMSE, GHQ, and BDI. Gen Hosp Psychi-atry 5:15–24, 1983

3. Moffic HS, Paykel ES: Depression in medical inpa-tients. Br J Psychiatry 126:346–353, 1975

R. Leo et al.

180

4. Schwab JJ, Bialow M, Brown JM, Holzer CE: Diag-nosing depression in medical inpatients. Ann InternMed 67:695–707, 1967

5. Clarke DM, McKenzie DP, Smith GC: The recognitionof depression in patients referred to a consultation-liaison service. J Psychosom Res 39:327–334, 1995

6. Clarke DM, Minas IH, McKenzie DP: Illness behav-iour as a determinant of referral to a psychiatricconsultation/liaison service. Aust N Z J Psychiatry25:330–337, 1991

7. Aromaa A, Raita Salo R, Reunanen A, et al: Depres-sion and cardiovascular disease. Acta PsychiatrScand 377(Suppl):77–82, 1994

8. Bruce ML, Leaf PJ: Psychiatric disorders and 15-month mortality in a community sample of olderadults. Am J Public Health 79:727–730, 1989

9. Lustman PJ, Griffith LS, Clouse RE, Cryer PE: Psy-chiatric illness in diabetes mellitus: relationship tosymptoms and glucose control. J Nerv Ment Dis 174:736–742, 1986

10. Murphy E, Smith R, Lindesay J, et al: Increased mor-tality rates in late life depression. Br J Psychiatry152:347–353, 1988

11. Walker EA, Gelfand MD, Gelfand AN, Creed F, Ka-ton WJ: The relationship of current psychiatric disor-der to functional disability and distress in patientswith inflammatory bowel disease. Gen Hosp Psychi-atry 18:220–229, 1996

12. Katon W: The epidemiology of depression in medicalcare. Int J Psychiatry Med 17:93–112, 1987

13. Katon W: The impact of major depression on chronicmedical illness. Gen Hosp Psychiatry 18:215–219,1996

14. Katon W, Sullivan MD: Depression and chronic med-ical illness. J Clin Psychiatry 51(Suppl):3–11, 1990

15. Gerber PD, Barrett J, Barrett J, et al: Recognition ofdepression by internists in primary care: a compari-son of internist and “gold standard” psychiatric as-sessments. J Gen Intern Med 4:7–13, 1989

16. Hirschfeld RMA, Keller MB, Panico S, et al: The na-tional depressive and manic-depressive associationconsensus statement on the undertreatment of de-pression. JAMA 277:333–340, 1997

17. Sliman RJ, Donohue TA, Jarjoura D, Ognibene AJ:Recognition of depression by internal medicine resi-dents. J Community Health 17:143–152, 1992

18. U.S. Department of Health and Human Services: De-pression in Primary Care. AHCPR Publication No.93-0550, Rockville, Maryland, 1993

19. Williams-Russo P: Barriers to diagnosis and treat-ment of depression in primary care centers. Am JGeriatr Psychiatry 4(Suppl):84–90, 1996

20. Docherty JP: Barriers to the diagnosis of depressionin primary care. J Clin Psychiatry 58(Suppl):5–10,1997

21. Montano CB: Recognition and treatment of depres-sion in a primary care setting. J Clin Psychiatry55(Suppl):18–34, 1994

22. Rabins PV: Barriers to diagnosis and treatment ofdepression in elderly patients. Am J Geriatr Psychia-try 4(Suppl):79–83, 1996

23. Fabrega H Jr, Mulsant BM, Rifai AH, et al: Ethnicityand psychopathology in an aging hospital-based

population: a comparison of African-American andAnglo-European patients. J Nerv Ment Dis 182:136–144, 1994

24. Kleinman AM: Depression, somatization and the“new cross-cultural psychiatry.” Soc Sci Med 11:3–10,1977

25. Adebimpe VR, Hedlund JL, Cho DW, Wood JB:Symptomatology of depression in black and whitepatients. J Natl Med Assoc 74:185–190, 1982

26. Beiser M: A study of depression among traditionalAfricans, urban north Americans, and southeastAsian refugees. In Kleinman A, Good B (eds), Cultureand Depression. California, University of CaliforniaPress, 1985, pp, 299–324

27. Kleinman A: The social origins of distress and dis-ease. Yale University Press, 1986

28. Rosen DH, Gregory RJ, Pollock D, Schiffmann A:Depression in patients referred for psychiatricconsultation: a need for a new diagnosis. Gen HospPsychiatry 9:391–397, 1987

29. Kleinman A: Clinical relevance of anthropologicaland cross-cultural research: concepts and strategies.Am J Psychiatry 135:427–431, 1978

30. Adebimpe VR: Overview: white norms and psychi-atric diagnosis on black patients. Am J Psychiatry138:279–285, 1981

31. Fabrega H, Mezzich J, Ulrich RF: Black-white differ-ences in psychopathology in an urban psychiatricpopulation. Compr Psychiatry 29:285–297, 1988

32. Jones BE, Gray BA: Problems in diagnosing schizo-phrenia and affective disorders among blacks. HospCommunity Psychiatry 37:61–65, 1986

33. Collins D, Dimsdale JE, Wilkins D: Consultation/liaison psychiatry utilization patterns in different cul-tural groups. Psychosom Med 54:240–245, 1992

34. Leo RJ, Narayan DA, Sherry C, Michalek C, PollockD: Geropsychiatric consultation for African-Ameri-can and Caucasian patients. Gen Hosp Psychiatry19:216–222, 1997

35. Boland RJ, Diaz S, Lamdan RM, Ramchandani D,McCartney JR: Overdiagnosis of depression in thegeneral hospital. Gen Hosp Psychiatry 18:28–35, 1996

36. Ries RK, Bokan JA, Kleinman A, Schuckit MA: Psy-chiatric consultation-liaison service: patients, re-quests, and functions. Gen Hosp Psychiatry 3:204–212, 1980

37. Margolis RL: Nonpsychiatric house staff frequentlymisdiagnose psychiatric disorders in general hospitalinpatients. Psychosomatics 35:485–491, 1994

38. Cavanaugh S: Diagnosing depression in the hospital-ized patient with chronic medical illness. J Clin Psy-chiatry 45:13–16, 1984

39. Cavanaugh S, Clark DC, Gibbons RD: Diagnosingdepression in the hospitalized medically ill. Psycho-somatics 24:809–815, 1983

40. Plumb M, Holland J: Comparative studies of psycho-logical function in patients with advanced cancer: I.Self-reported depressive symptoms. Psychosom Med39:348–359, 1969

41. Rifkin A: Depression in physically ill patients. Post-grad Med 92:147–154, 1992

42. Craig TJ: An epidemiologic study of a psychiatricliaison service. Gen Hosp Psychiatry 4:131–137, 1982

Referrals for and Recognition of Depression

181

43. Bazargan M, Barbre AR: The effects of depression,health status, and stressful life-events on self-reported memory problems among aged blacks. Int JAging Hum Dev 38:351–362, 1994

44. Lawson WB: Racial and ethnic factors in psychiatricresearch. Hosp Community Psychiatry 37:50–54, 1986

45. Adebimpe VR: Race, racism, and epidemiologicalsurveys. Hosp Community Psychiatry 45:27–31, 1994

46. Broman CL: Race differences in professional helpseeking. Am J Community Psychol 15:473–489, 1987

47. Sussman LK, Robins LN, Earls F: Treatment-seekingfor depression by black and white Americans. Soc SciMed 24:187–196, 1987

48. Harper MS: Elderly issues in the African-Americancommunity. In Braithwaite RL, Taylor SE (eds),Health Issues in the Black Community. San Francisco,Jossey-Bass, 1992

49. Strakowski SM, Shelton RC, Kolbrener ML: The ef-fects of race and comorbidity on clinical diagnosis inpatients with psychosis. J Clin Psychiatry 54:96–102,1993

50. Elixhauser A, Harris DR, Coffey RM: Trends in hos-pital diagnoses for black patients and white patients:1980–87 (AHCPR Publication No. 95-0048). ProviderStudies Research Note 24, Agency for Health CarePolicy and Research, Rockville, Maryland: PublicHealth Service, 1995

51. Hanson B, Klerman GL: Interracial problems in theassessment of clinical depression: concordance differ-ences between white psychiatrist and black and whitepatients. Psychopharmacol Bull 10:65–66, 1974

52. Comstock MS, Helsing KJ: Symptoms of depressionin two communities. Psychol Med 6:551–563, 1976

53. Neff JA, Husaini BA: Race, socioeconomic status, andpsychiatric impairment: a research note. J Commu-nity Psychol 8:16–19, 1980

54. Roberts RE, Stevenson JM, Breslow L: Symptoms ofdepression among blacks and whites in an urbancommunity. J Nerv Ment Dis 169:774–779, 1981

55. Katon W, Schulberg H: Epidemiology of depression

in primary care. Gen Hosp Psychiatry 14:237–247,1992

56. Stewart MA, Drake F, Winokur G: Depression amongmedically ill patients. Dis Nerv Syst 26:479–485, 1965

57. Health United States, 1990. Hyattsville, Maryland,National Center for Health Statistics, 1991

58. Manton KG, Patrick CH, Johnson KW: Health differ-entials between blacks and whites: recent trends inmortality and morbidity. Milbank Q 65:129–199, 1987

59. Task Force on Black and Minority Health, Vol 1:Executive Summary. Washington, DC, Departmentof Health and Human Services, 1985

60. Blendon RJ, Aiken LH, Freeman HE, Corey CR: Ac-cess to medical care for black and white Americans.JAMA 261:278–281, 1989

61. Council on Ethical and Judicial Affairs: Black-whitedisparities in health care. JAMA 263:2344–2346, 1990

62. Fichtenbaum R, Gyimah-Brempong K: The effects ofrace on the use of physicians’ services. Int J HealthServ 27:139–156, 1997

63. Freeman HE, Blendon RJ, Aiken LH, et al: Americansreport on their access to health care. Health Aff 6:6–18, 1987

64. Phillips RS, Hamel MB, Teno JM, et al: Race, resourceuse, and survival in seriously ill hospitalized adults.J Gen Intern Med 11:387–396, 1996

65. Weissman JS, Stern R, Fielding SL, Epstein AM: De-layed access to health care: risk factors, reasons, andconsequences. Ann Intern Med 114:325–331, 1991

66. Brown DR, Ahmed F, Gary LE, Milburn NG: Majordepression in a community sample of African Amer-icans. Am J Psychiatry 152:373–378, 1995

67. U.S. Bureau of the Census: Population projections ofthe United States by age, sex, race, and Hispanicorigin: 1995 to 2050. Current Population Reports, P25-1130, U.S. Government Printing Office, Washington,DC, 1996

68. Katon W, Von Korff M, Lin E, et al: Collaborativemanagement to achieve depression treatment guide-lines. J Clin Psychiatry 58(Suppl):20–23, 1997

R. Leo et al.

182