Psychosomatics 44:407-411, October 2003
© 2003 The Academy of Psychosomatic Medicine
Accuracy of Referring Psychiatric Diagnosis on a Consultation-Liaison Service
Stephen L. Dilts, JR., M.D., M.B.A.,
Nancy Mann, R.N., M.S., and
John G. Dilts, M.P.P.
Received Aug 29, 2002; revision received Jan. 21, 2003; accepted Feb. 6, 2003. From York Hospital, York, Pa. Address reprint requests to Stephen L. Dilts, M.D., M.B.A., 1600 S. George St., York, PA 17403; diltsjs{at}aol.com (e-mail).
|
ABSTRACT
|
The authors determined the accuracy of the initial psychiatric diagnosis of primary medical providers requesting psychiatric consultation in a general medical inpatient setting. A retrospective review of 346 consecutive psychiatric consultations was conducted in which the initial diagnostic impression of primary medical providers was compared with the final psychiatric diagnosis. Accuracy rates for cognitive disorders, substance use disorders, and depressive disorders were 100%, 88.9%, and 53.6%, respectively. Thus, initial diagnoses of a cognitive or substance use disorder by primary medical providers are likely to be accurate, whereas an initial diagnosis of a depressive disorder will be inaccurate in approximately half of the cases.
|
INTRODUCTION
|
The prevalence and seriousness of psychiatric disorders in the general medical setting are well established.1–3 It is estimated that between 26.5% and 60% of general medical inpatients suffer psychiatric comorbidity,4–7 and psychiatric comorbidity has been associated with poorer outcomes. For example, delirium has been shown to predict increased length of stay, cost, morbidity, and mortality,8–12 and depression is associated with increased morbidity and mortality in patients with coronary artery disease.13–15 However, there is evidence that despite broad awareness of the scope of this problem, psychiatric disorders continue to be underdiagnosed and misdiagnosed in the general medical inpatient setting. Studies have examined the accuracy of emergency department physicians' assessment of delirium in the elderly,16–19 the misdiagnosis of delirium in the general medical setting,20–24 and the detection rate of depression in the general hospital and critical care unit.3,25 To date, however, no one has examined the accuracy of all initial referring diagnoses in those patients subsequently evaluated by a consulting psychiatrist. Psychiatric consultation rates are historically low,26,27 but when primary medical providers do suspect the presence of a psychiatric illness and refer accordingly, how accurate are the initial diagnoses?
We attempted to ascertain the accuracy of initial psychiatric diagnoses among patients referred for psychiatric consultation in the general medical inpatient setting. On the basis of impressions gathered on this service, as well as findings from other studies, we hypothesized that an initial diagnosis of depression would be less accurate than other psychiatric disorders and often actually represent a cognitive disorder.
|
METHOD
|
The setting for this study was York Hospital, a 450-bed regional trauma center located in York, Pa., and serving approximately 30,000 inpatients a year. The psychiatric consultation-liaison service at this center receives requests for an average of 2.9 new consultations a day for a wide variety of problems. All patients on this consultation-liaison service are examined by a board-certified or board-eligible psychiatrist who, with the assistance of a master's-level clinical nurse specialist, conducts a full psychiatric assessment to determine diagnosis and make recommendations for treatment. We performed a retrospective review of all consultation requests from Sept. 6, 2001, to Jan. 6, 2002, and calculated the accuracy of the initial diagnostic impression compared with the final psychiatric diagnosis as determined by the consulting psychiatrist. Ten different psychiatrists performed assessments during the study interval.
Because most referrals were not for a specific DSM-IV diagnosis but rather for a general diagnostic category such as "depression" or "dementia," diagnostic agreement was defined as general category agreement of the initial and final diagnoses. Thus, an initial diagnosis of "depression" was validated if the final diagnosis included major depression, depressive disorder not otherwise specified, adjustment disorder with depressed mood, bipolar depression, or depression secondary to a general medical condition or the effects of substance use. Although these disorders are quite heterogeneous, any could have been implied under the initial referring problem. Similarly, an initial diagnosis of a substance use disorder was considered validated if any substance use disorder was found to be the primary diagnosis on final assessment. Some patients received multiple final diagnoses. In such cases, the first primary diagnosis was used. Accuracy was calculated as the ratio of confirmed diagnoses to total referred diagnoses within each category. Chi-square and two-sided Fisher's exact tests were conducted to determine if differences in accuracy across initial diagnostic categories reflected statistically significant differences in the observed levels of agreement between initial and final diagnoses.
|
RESULTS
|
The consultation-liaison service saw 346 patients during the study interval. Approximately 9,380 patients were admitted to the hospital during this interval, yielding a psychiatric consultation rate of 3.7%. Three diagnostic categories—cognitive disorders, substance use disorders, and depressive disorders—dominated both the initial and final diagnostic impressions, accounting as a group for 56.6% and 80.6% of diagnoses, respectively (Table 1). An initial diagnostic impression of a cognitive disorder was 100% accurate, and a substance use diagnosis was accurate in 88.9% of cases (Table 2). A Fisher's exact test revealed no statistically significant difference between these accuracies.
View this table:
[in this window]
[in a new window]
|
TABLE 1. Final Psychiatric Diagnoses of 346 General Medical Inpatients Referred to a Consultation-Liaison Service Over a 4-Month Perioda
|
View this table:
[in this window]
[in a new window]
|
TABLE 2. Accuracy of Primary Medical Provider Psychiatric Referral Diagnoses of 346 General Medical Inpatients Referred to a Consultation-Liaison Service Over a 4-Month Period Compared With Final Diagnoses
|
An initial diagnosis of a depressive disorder was accurate in 53.6% of cases, and a primary cognitive disorder was diagnosed in 26.2% (N=22 of 84) of those patients referred for depression. Although constituting a small percentage of total referrals, bipolar disorder, anxiety disorders, and psychotic disorders were accurate in 42.8%, 40.0%, and 40.0% of cases, respectively. For these categories, the majority of missed diagnoses represented cognitive disorders. Chi-square tests revealed a statistically significant difference between the accuracy of an initial cognitive or substance use disorder versus a depressive disorder ( 2=36.0, df=1, p<0.0001). Two-sided Fisher's exact tests showed that the diagnostic accuracy was significantly greater for cognitive and substance use disorders than for anxiety disorder (p<0.001), psychotic disorder (p<0.01), or bipolar disorder (p<0.01).
|
CONCLUSIONS
|
The overall psychiatric consultation rate of 3.7% approximates the rate of consultation seen in other studies and may be slightly higher. In this study, we found that three diagnostic categories—cognitive disorders, substance use disorders, and depressive disorders—formed the large majority of both initial and final diagnoses. That these diagnoses represented a larger percentage of the final assessments than the initial is probably explained by the substantial number of referrals (33.8%) made for problem-based, nondiagnostic reasons such as "suicide attempt" or "competency." The majority of these patients were ultimately found to have depressive and cognitive disorders.
The results of this study suggest that when primary medical providers in the general medical inpatient setting suspect a cognitive or substance use disorder, they are likely to be correct. The accuracy of the initial diagnostic impression for both these disorders in our study was quite good. However, there is a significant likelihood of inaccuracy when primary medical providers suspect depression, bipolar disorder, psychosis, or anxiety. Primary medical providers, while accurate when they do suspect a cognitive disorder, appear to frequently mistake symptoms of a cognitive disorder for a depressive or, less commonly, a bipolar, anxiety, or psychotic disorder. Primary providers are prone to miss the significance of cognitive impairment, referring for a mood disorder instead. Since delirium is a well-known predictor of mortality, increased length of stay, and increased cost, this failure to accurately appreciate the prevalence of cognitive disorders has significant implications for patient care. In addition to delays in treating covert general medical problems, it can result in the initiation of inappropriate therapy, such as antidepressant medication in the delirious patient. Primary providers diagnosed only 22 of 93 patients ultimately found to have a cognitive disorder, or 23.7% (compared to 90.9% [N=80 of 88] for substance use disorders and 45.9% [N=45 of 98] for depressive disorders). This result confirms the difficulty in identifying cognitive disorders that has been seen in other studies.16–23 The quality of care issues that this persistent deficiency represents have been reviewed elsewhere.28–32
It is interesting that consulting psychiatrists did not discover any new cases of a substance use disorder in those patients initially diagnosed with depression. This may be because primary medical providers are good at appreciating substance use disorders and referring them as such or because the relatively older age of the inpatient medical population and high prevalence of cognitive disorders obscured the detection of substance use disorders even by psychiatrists. It is also possible that the consulting psychiatrists identified new substance use disorders as secondary problems, which were not reflected in our data. While it would have been preferable to have the final psychiatric diagnosis validated by a structured clinical interview, such a method is difficult in the general medical inpatient setting, where a large number of patients are unable (because of cognitive impairment, pain, general malaise and debilitation, or demands of care) to participate in a lengthy structured interview. Since the diagnostic impression of a psychiatrist is in practice the ultimate determinant of diagnosis and care, this study does reflect the current clinical climate but does not validate final diagnosis beyond assuming the expertise of the consulting psychiatrist.
The absence of axis II pathology as a primary diagnosis also merits discussion. Axis II pathology may have been underdiagnosed because of the lack of an ongoing physician-patient relationship, which is often necessary to appreciate axis II pathology. Also, the consulting psychiatrists may have felt that any axis II pathology was overridden by more pressing axis I concerns in the general medical inpatient setting.
This study is biased by other factors. First, although the data were accumulated from a large number of referring and consulting physicians, they represent a sample from a single practice setting, and so results may not generalize to other centers. Second, because most referrals were not for specific DSM-IV diagnoses, but rather for general problem categories, we were only able to validate to this level of agreement. Thus, for example, a nonspecific referral for a depressive disorder was credited as accurate if the final primary diagnosis was any depressive disorder. Assuming it were possible to collect a sufficient sample of specific initial diagnoses to render meaningful data, primary medical providers would likely have performed substantially worse. We considered two nondiagnostic symptoms, "confusion" and "psychosis," on referral to mean "cognitive disorder" and "psychotic disorder," respectively. The first assumption rendered a 100% accuracy in favor of the primary medical provider diagnosis even though it is not clear that providers referring for confusion were even familiar with the concept of a cognitive disorder (we received no consult requests for "delirium," for example). The assumption that a referral for "psychosis" implied an initial impression of a psychotic disorder may have caused the accuracy for psychotic disorders to read artificially low, since primary providers may have correctly assumed that other disorders (such as cognitive, substance use, and affective disorders) can exhibit psychotic symptoms.
The retrospective nature of the study may have undermined our ability to determine what referring providers thought was the initial diagnosis. Perhaps a referral for depression, for example, was simply intended to describe an observed behavioral problem and not to indicate an attempt at diagnosis. While we have duplicated the naturalistic method of assessing referrals used in other studies,17,18,23,31 it provides an inexact picture of what the referring physician intended. Correcting this difficulty in a prospective study, perhaps by contacting referring providers to obtain a specific diagnosis, would improve our understanding of the accuracy of primary provider diagnosis. A prospective trial would also increase the possibility of standardizing psychiatric consultation responses, such as by directing them to designate whether the axis I or axis II diagnosis is the primary problem. It should be noted, however, that the retrospective nature of this study served to blind the consulting psychiatrists so that their assessments were not biased by the study itself.
The implications of this naturalistic study for consulting psychiatrists remain. Regardless of whether primary providers intended a referral for depression to indicate an initial formal diagnosis of depression, this study shows that psychiatrists should approach such a referral with a high index of suspicion for cognitive disorders, while they may be more comfortable when the referral is for a cognitive or substance use problem.
In conclusion, we found that initial diagnoses of a cognitive or substance use disorder by primary medical providers in the general medical inpatient setting are highly accurate but that this is not the case with depressive, anxiety, psychotic, or bipolar disorders. Many of these referrals are likely to represent cognitive disorders instead. Because of the high prevalence of cognitive, substance use, and depressive disorders in the general medical inpatient setting, consultation-liaison teaching should devote most of its energies toward the identification and management of these disorders, with particular emphasis on the accurate diagnosis of cognitive disorders. Future studies should attempt to replicate these findings in other practice sites, to discern the accuracy of specific DSM-IV referring diagnoses rather than problem categories, and to validate consultation diagnosis by psychiatrists in the general medical setting.
|
REFERENCES
|
- Yates WR: Epidemiology of psychiatric disorders in the medically ill, in Psychiatric Treatment of the Medically Ill Patient. Edited by Robinson RG, Yates WR. New York, Marcel Dekker, 1999, pp 41–64
- Crespo D, Gil A, Porras Chavarino A: Prevalence of depressive disorders in consultation-liaison psychiatry. Actas Esp Psiquiatr 2001; 29:75–83[Medline]
- Rincon HG, Granados M, Unutzer J, Gomez G, Duran R, Badiel M, Salas C, Martinez J, Mejia J, Ordonez C, Florez N, Rosso F, Echeverri P: Prevalence, detection and treatment of anxiety, depression, and delirium in the adult critical care unit. Psychosomatics 2001; 42:391–396[Abstract/Free Full Text]
- Silverstone PH: Prevalence of psychiatric disorders in medical inpatients. J Nerv Ment Dis 1996; 184:43–51[CrossRef][Medline]
- Hansen MS, Fink P, Frydenberg M, Oxhoj M, Sondergaard L, Munk-Jorgensen P: Mental disorders among internal medical inpatients: prevalence, detection, and treatment status. J Psychosom Res 2001; 50:199–204[CrossRef][Medline]
- Seltzer A: Prevalence, detection and referral of psychiatric morbidity in general medical patients. J R Soc Med 1989; 82:410–412[Abstract]
- Martucci M, Balestrieri M, Bisoffi G, Bonizzato P, Covre MG, Cunico L, De Francesco M, Marinoni MG, Mosciaro C, Piccinelli M, Vaccari L, Tansella M: Evaluating psychiatric morbidity in a general hospital: a two-phase epidemiological survey. Psychol Med 1999; 29:823–832[CrossRef][Medline]
- Inouye SK, Rushing JT, Foreman MD, Palmer RM, Pompei P: Does delirium contribute to poor hospital outcomes? a three-site epidemiologic study. J Gen Intern Med 1988; 13:234–242
- McCusker J, Cole M, Abrahamowicz M, Primeau F, Belzile E: Delirium predicts 12-month mortality. Arch Intern Med 2002; 162:457–463[Abstract/Free Full Text]
- Stevens LE, de Moore GM, Simpson JM: Delirium in hospital: does it increase length-of-stay? Aust NZ J Psychiatry 1998; 32:805–808[Medline]
- Marcantonio ER, Flacker JM, Michaels M, Resnick NM: Delirium is independently associated with poor functional recovery after hip fracture. J Am Geriatr Soc 2000; 48:618–624[Medline]
- Kishi Y, Iwasaki Y, Takezawa K, Kurosawa H, Endo S: Delirium in critical care unit patients admitted through an emergency room. Gen Hosp Psychiatry 1995; 17:371–379[CrossRef][Medline]
- Lesperance F, Frasure-Smith N, Juneau M, Theroux P: Depression and 1-year prognosis in unstable angina. Arch Intern Med 2000; 160:1354–1360[Abstract/Free Full Text]
- Frasure-Smith N, Lesperance F, Talajic M: Depression following myocardial infarction: impact on 6-month survival. JAMA 1993; 270:1819–1825; correction, 1994; 271:1082
- Legault S, Joffe R, Armstrong P: Psychiatric morbidity during the early phase of coronary care for myocardial infarction: association with cardiac diagnosis and outcome. Can J Psychiatry 1992; 37:316–325[Medline]
- Elie M, Rousseau F, Cole M, Primeau F, McCusker J, Bellavance F: Prevalence and detection of delirium in elderly emergency department patients. Can Med Assoc J 2000; 163:977–981[Abstract/Free Full Text]
- Hustey FM, Meldon SW: The prevalence and documentation of impaired mental status in elderly emergency department patients. Ann Emerg Med 2002; 39:248–253[CrossRef][Medline]
- Hustey FM, Meldon S, Palmer R: Prevalence and documentation of impaired mental status in elderly emergency department patients. Acad Emerg Med 2000; 7:1166[Medline]
- Lewis LM, Miller DK, Morely JE, Nork MJ, Lasater LC: Unrecognized delirium in ED geriatric patients. Am J Emerg Med 1995; 13:142–145[CrossRef][Medline]
- Milisen K, Foreman MD, Wouters B, Driesen R, Godderis J, Abraham IL, Broos PL: Documentation of delirium in elderly patients with hip fracture. J Gerontol Nurs 2002; 28:23–29
- Harwood DM, Hope T, Jacoby R: Cognitive impairment in medical inpatients, II: do physicians miss cognitive impairment? Age Ageing 1997; 26:37–39
- Farrell KR, Ganzini L: Misdiagnosing delirium as depression in medically ill elderly patients. Arch Intern Med 1995; 155:2459–2464[Abstract]
- Armstrong SC, Cozza KL, Watanabe KS: The misdiagnosis of delirium. Psychosomatics 1997; 38:433–439[Abstract/Free Full Text]
- Margolis RL: Nonpsychiatrist house staff frequently misdiagnose psychiatric disorders in general hospital inpatients. Psychosomatics 1994; 35:485–491[Abstract/Free Full Text]
- Boland RJ, Diaz S, Lamdan RM, Ramchandani D, McCartney JR: Overdiagnosis of depression in the general hospital. Gen Hosp Psychiatry 1996; 18:28–35[CrossRef][Medline]
- Rothenhausler H-B, Ehrentraut S, Kapfhammer H-P: Changes in patterns of psychiatric referral in a German general hospital: results of a comparison of two 1-year surveys 8 years apart. Gen Hosp Psychiatry 2001; 23:205–214[CrossRef][Medline]
- Tsai SJ, Lee YC, Yang CH, Sim CB: Psychiatric consultations in obstetric inpatients. J Obstet Gynaecol Res 1996; 22:603–607[Medline]
- Sanders AB: Missed delirium in older emergency department patients: a quality-of-care problem. Ann Emerg Med 2002; 39:338–341[CrossRef][Medline]
- Inouye SK: Delirium: a barometer for quality of hospital care. Hosp Pract (Off Ed) 2001; 36:15–16, 18
- Inouye SK, Schlesinger MJ, Lydon TJ: Delirium: a symptom of how hospital care is failing older persons and a window to improve quality of hospital care. Am J Med 1999; 106:565–573[CrossRef][Medline]
- Naughton BJ, Moran MB, Kadah H, Heman-Ackah Y, Longano J: Delirium and other cognitive impairment in older adults in an emergency department. Ann Emerg Med 1995; 25:751–755[CrossRef][Medline]
- Sanders AB: Recognition of cognitive problems in older adults by emergency department personnel. Ann Emerg Med 1995; 25:831–833[Medline]
Get information about faster international access.
Privacy Policy
Copyright © 2003
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
