Psychosomatics 40:50-56, February 1999
© 1999 The Academy of Psychosomatic Medine
Panic Disorder Patients and Their Medical Care
Arthur J. Barsky, M.D.,
Beth A. Delamater, Ed.M., and
John E. Orav, Ph.D.
Received February 27, 1997; revised April 27, 1998; accepted May 8, 1998. From the Divisions of Psychiatry and Clinical Epidemiology, Brigham and Women's Hospital; the Department of Psychiatry, Harvard Medical School; and the Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts. Address correspondence and reprint requests to Dr. Barsky, Division of Psychiatry, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115.
|
ABSTRACT
|
The goal of the study was to examine the functional status and medical care of general medical outpatients with panic disorder. One hundred patients completed self-report questionnaires and a diagnostic interview for panic disorder. They were compared with a random sample of patients without panic disorder. Medical morbidity was assessed from the medical record, and the patients' clinic physicians completed a questionnaire about them. The prevalence of current (1 month) panic disorder was 6.7%–8.3%. The panic disorder patients had fewer serious medical diagnoses, but more medical utilization and more role impairment than the comparison group. The clinic physicians rated the panic patients as more anxious, more depressed, more hypochondriacal, and more difficult to care for. Sixty-one percent of the panic disorder patients recalled receiving an anxiety disorder diagnosis. These findings add to a growing body of evidence that panic disorder imposes a significant burden on those with this illness and that it is a seriously underdiagnosed condition in primary care practice.
Key Words: Panic Disorder • Medical Outpatients • Functional Status
|
INTRODUCTION
|
Substantial research has been devoted to the medical care and medical outcomes of patients in primary care practice with psychiatric disorders.1–4 In particular, panic disorder has recently begun to attract such interest.5 In community samples, panic disorder is associated with pervasive general health consequences, as well as with impairment of occupational, social, and marital function and poorer perceived health.6–9 Health maintenance organization patients with panic disorder have disability levels comparable to those of patients with chronic medical conditions such as congestive heart failure and diabetes.10 They are also high users of medical services, including general medical, specialty mental health, and emergency room visits.6,11–16 Yet only a minority of panic disorder cases in general medical practice are detected and treated.10,17
In this study, conducted in 1994, we examined a rigorously selected sample of panic disorder patients from a primary care setting, and we compared them to a random sample of patients without panic disorder from the same setting. We focused particularly on their functional status and on their medical care.
|
METHODS
|
Subjects and Setting
The study was conducted in a large general medicine clinic of an academic teaching hospital. Written informed consent was obtained after the study was fully explained to the participants in accordance with hospital procedures. For the panic disorder sample, all patients who were literate, spoke English, and were free of major communication deficits were eligible. For the comparison sample, only the patients of the attending physicians were studied, and they were eligible only if they had been followed at the hospital for at least 24 months and had at least one prior visit to the same primary physician.
Design and Procedure
Panic Disorder Sample.
Consecutive clinic attenders on randomly selected days completed a self-report screening questionnaire for panic disorder (discussed next). A cutoff score was chosen to maximize sensitivity at the expense of specificity; we were willing to accept a relatively large number of false-positive diagnoses to minimize the number of false-negatives. Those patients exceeding the cutoff were telephoned at home to determine whether their panic symptoms had a medical basis. If not, they were asked to return to the hospital to undergo the research battery. This included a structured diagnostic interview for panic disorder, and the interview-positive patients constitute the panic disorder sample.
Comparison Sample.
The comparison group was derived from another study that was conducted concurrently in the same setting.18–20 Consecutive visitors to the clinic were screened with a self-report hypochondriasis questionnaire. All patients scoring above a predetermined cutoff, and a random sample of those below the cutoff, were asked to participate. For the purposes of this report, we have weighted these two groups proportionately so as to reconstitute a representative sample of the entire clinic.
Research Battery.
The research interview was the same for both samples. All instruments were administered by trained, Master's-level research assistants. After the research interview, each patient's medical record was audited blindly (AJB) to assess medical morbidity. In addition, the patient's primary physician completed a 10-item questionnaire about the patient. The subjects received $40 for participating.
Variables and Their Measurement
Panic Disorder.
The 15-item Screener for Panic Disorder was used.21 This questionnaire was developed by expert consensus panel for use in medical settings with medically ill respondents. The instrument has high sensitivity and acceptable specificity and has been validated against the Anxiety Disorders Interview Schedule.21 A positive response to five or more questions was used as the cutoff in the present study.
The criterion standard diagnosis of panic disorder was made with the Diagnostic Interview Schedule (DIS), Version 3-R.22 This widely used, highly structured interview generates the major Axis I diagnoses, both current and lifetime, by using operationalized DSM-III-R criteria. Its reliability, validity, and psychometric properties have been demonstrated. The modules covering the anxiety disorders, depressive disorders, and somatoform disorders (somatization disorder) were administered. The panic disorder section was modified slightly, as discussed by Katon and colleagues,23,24 to minimize the false-negatives that are more likely among highly somatizing patients.
The patients were also directly asked about the diagnosis of panic disorder: "Has any health professional ever told you that you have an anxiety disorder, anxiety attacks, or panic attacks?" and "Have you ever had medical treatment for an anxiety disorder or anxiety attacks?"
Medical Morbidity.
Medical morbidity was assessed with a blind medical-record audit. This method was used in prior studies.25 Each medical diagnosis was assigned a weight from 1 to 4 depending upon its severity, using explicit, preestablished criteria (1=major, 2=moderate, 3=minor, 4=nonspecific complaint), and the number of diagnoses in each category was obtained. This method was shown in pilot work to have satisfactory interrater reliability. The patient's physician also rated current aggregate morbidity and future prognosis on a five-point ordinal scale. These two methods of assessing morbidity are intercorrelated (r=0.52, P=0.0001).25
Functional Status.
Disability was assessed with the Functional Status Questionnaire (FSQ), a valid and reliable self-report instrument developed for use in ambulatory medical populations.26 We used the 31 items that comprise four of its six subscales: basic activities, intermediate activities of daily living, social activities, and work performance. FSQ scores range from 1 to 100 (maximal functional ability).
Medical Care.
Medical utilization was obtained by asking the patients directly about their use of services in the preceding 12 months.
Satisfaction with medical care was measured with three items taken from the FSQ, scored on an ordinal scale from one (strongly agree) to five (strongly disagree): The intrascale reliability of these three items was 0.76 (Cronbach's  ). Satisfaction with overall health was measured with the single item, "How do you feel about your own health?" scored on an ordinal scale from one (very satisfied) to five (very dissatisfied).
Each patient's primary care physician completed a questionnaire about him/her. The questions included "How depressed do you think this patient is?" "How anxious do you think this patient is?" "How effective are you in treating this patient's somatic symptoms?" "How demanding, dependent, and clinging do you find this patient?" "How much does this patient thwart your attempts to help?" and "How emotionally taxing and frustrating do you find this patient?" Responses were scored on a five-point ordinal scale from 1 (not at all) to 5 (a great deal).
Data Analysis
The data for the subjects in the comparison sample were weighted to reconstitute a sample that is representative of the clinic as a whole. The prevalence rates previously reported are 6.5% for DSM-III-R hypochondriasis, 2.9% for subthreshold hypochondriasis, and 90.6% for nonhypochondriasis.18 The cases were weighted so that each subgroup of patients accounted for the appropriate proportion of the total sample (N=188). The patients with hypochondriasis (n=60) and subthreshold hypochondriasis (n=28) were oversampled, and consequently were given weights of 0.21 and 0.20, respectively. The nonhypochondriacal patients (n=100) were relatively underrepresented, and therefore weighted 1.7. Multiplying the number of subjects in each subgroup by the corresponding weight yields a weighted sample of 188 patients. Nine patients who met criteria for lifetime panic disorder were then deleted. The resultant group presented here therefore reflects a weighted sample of 179 subjects.
The clinical characteristics and medical care of patients with panic disorder (n=100) were then compared with those of the comparison patients. Another series of analyses explored the differences between those patients with panic disorder who reported receiving the diagnosis of "panic" or "anxiety" from a medical care professional (n=61) and the panic disorder patients who denied receiving the diagnosis (n=39). Weighted t-tests were used as the test of significance for comparisons involving continuous variables, and a weighted chi-square was used for categorical dependent variables. When age was included as a covariate, analysis of variance with covariates was used.
|
RESULTS
|
A total of 1,754 consecutive patients on randomly selected days completed the panic disorder screening questionnaire. Of these, 248 (14.1%) exceeded the cutoff score, 120 (6.8%) were ineligible or returned incomplete screens, and 1,386 were below the cutoff. Of the 248 patients who exceeded the cutoff score, 3 patients (1.2%) could not be reached, 47 (19.0%) did not complete the interview because their panic symptoms had medical explanations, 11 (4.4%) failed to meet inclusion criteria, and 52 (27.4%) refused to be interviewed. One hundred thirty-five patients (71.0% of the 190 eligible patients) underwent the research battery. One hundred patients (74.1%) in this group received a diagnosis of lifetime DSM-III-R panic disorder and comprise the sample studied here. Ninety-four of these patients had current (within the last 6 months) panic disorder. In 74, panic disorder was present within the last month. The details of the accrual of the comparison group have been presented in previous publications.25
If we assume that the rate of panic disorder among the screen-positive patients who were ineligible could not be contacted or refused to participate (n=113) was the same as it was among those eligible screen-positive patients who completed the research battery (n=135), then the prevalence of lifetime panic disorder in our clinic is 11.2% and the prevalence of current (1 month) panic disorder is 8.3%. If none of the patients who exceeded the screening score and had medical explanations for their symptoms (n=47) had panic disorder, then the prevalence of lifetime panic disorder is 9.1% and 6.7% for current (1 month) panic disorder.
Table 1 compares the sociodemographic characteristics of both samples. The panic disorder patients were significantly younger, more likely to be single, and were of lower socioeconomic status.
The panic patients had significantly fewer major medical problems (mean ± SD: 0.31 ± 0.62 vs. 0.92 ± 1.26, P=0.006) and significantly fewer moderate medical problems (0.98 ± 1.11 vs. 2.83 ± 2.07, P=0.0001). These differences remained significant after controlling for age. Psychiatric comorbidity was prevalent in the panic disorder sample. Thirteen percent of the patients met diagnostic criteria for current (6 month) somatization disorder, 57% had current major depression, 12% had current obsessive-compulsive disorder, and 14% had dysthymia.
Patient self-reports of functional status are compared in Table 2. The panic disorder patients were significantly more disabled with respect to basic, intermediate, and social activities than the comparison group. The former group also reported significantly more days spent in bed and significantly more restricted activity days.
The panic disorder patients used more medical care despite being medically healthier. In the preceding year, they made significantly more physician visits (mean ± SD: 10.6 ± 12.2 vs. 4.4 ± 3.0, P=0.0001), emergency room visits (2.0 ± 3.6 vs. 0.6 ± 0.9, P=0.0001), and mental health visits (9.7 ± 17.6 vs. 3.7 ± 14.7, P=0.0065) than the comparison group. The panic disorder patients also reported more hospital admissions (0.5 ± 0.9 vs. 0.05 ± 0.2, P=0.0001). At the same time, they were more dissatisfied with their medical care (2.48 ± 0.94 vs. 1.91 ± 0.87, P<0.0001) and with their general health (3.42 ± 1.17 vs. 2.29 ± 1.03, P<0.0001).
The primary care physicians considered their panic disorder patients significantly more hypochondriacal, more depressed, and more anxious than those without panic disorder (Table 3). The physicians also felt less effective in treating their patients' symptoms and found them more demanding, more frustrating, and more help-rejecting. Sixty-one of the panic patients recalled their doctors having told them that they had an "anxiety disorder," "anxiety attacks," or "panic attacks" (Table 4). Forty-eight reported receiving treatment for an anxiety disorder. Those aware of their diagnosis had significantly more panic attack symptoms than the undiagnosed patients and made significantly more emergency room visits. The diagnosed and undiagnosed patients did not differ significantly with respect to medical morbidity, age, disability, or medical care utilization. Undiagnosed patients, however, were rated as less anxious, less hypochondriacal, and less demanding by their physicians than those with diagnosed panic disorder.
|
DISCUSSION
|
The prevalence of current (1 month) panic disorder meeting full diagnostic criteria is between 6.7% and 8.3% in this primary care setting and that of lifetime disorder is between 9.1% and 11.2%. This is comparable to the rates of 6%–13% reported in other general medical settings5,10,27 and in ambulatory cardiology practices.28 A somewhat higher prevalence (17.5%) was reported among emergency room visitors with chest pain.15
Panic disorder patients, although younger and medically healthier than other primary care patients in the same setting, reported significantly more role impairment and disability, used significantly more medical care, and were more dissatisfied with their care and their general health. The magnitudes involved here are clinically meaningful: the panic disorder patients averaged 10.6 physician visits per year and 2 emergency room or walk-in visits per year, both very high figures. Their physicians distinguished them from other patients, rating them more hypochondriacal, more depressed, and more difficult to care for. These findings are compatible with the literature, which indicates that panic disorder patients in general medical practice have elevated levels of physician visits and emergency room visits,6,12,14 as well as impaired social, marital, and occupational functioning.6–8 They have also been found to have more medically unexplained symptoms, poorer subjective ratings of physical health, and more minor medical illness (but not more major diagnoses) than other general practice patients.6,14,27 Some of these differences may be attributable not purely to panic disorder, but to the other comorbid psychiatric conditions that so commonly accompany the panic disorder. However, the clinical findings remain that primary care patients with panic disorder differ in these characteristics from those without panic disorder in the same setting.
The identification of panic disorder by primary care physicians is particularly important since it can now be effectively treated with pharmacotherapy and/or cognitive–behavioral therapy. Furthermore, it has been shown that accurate detection of anxiety and depressive disorders by general practitioners is associated with improved psychiatric status and social role functioning.17 Although we were unable to establish definitively the extent of underdiagnosis of panic disorder in this sample, 61% of the patients reported having been told of an anxiety disorder, 53% had seen a mental health professional in the preceding year, and almost 50% reported having an anti-anxiety medication prescribed. This is comparable to the literature suggesting that about one-half of panic disorder cases in general practice go or remain undiagnosed.10,17 Thus, the problem of underdiagnosis or misdiagnosis appears to be substantial. We attempted to probe this by comparing those whose panic disorder had and had not been recognized. There were few surprising differences: the fomer were more symptomatic; made more emergency room visits; and appeared more anxious, more hypochondriacal, and more demanding to their physicians. The most likely interpretation is simply that patients with more severe disorder are more likely to be recognized, but it is also possible that physicians are more ready to affix a psychiatric label to patients whom they find more problematic and difficult to care for.
This study has several limitations. First, our setting may not be representative of primary care practice in general, and the 27% refusal rate raises the possibility of selection bias. Since we do not have any information on the patients who declined to participate, we are unable to estimate the magnitude of this possible bias. Second, there were minor differences in the eligibility criteria for the panic disorder and comparison patients. Third, a small fraction of these patients had lifetime panic but did not have active current disorder at the time of the study. Fourth, as noted earlier, we have not controlled for Axis I psychiatric comorbidity; therefore, it is possible that some of the properties of the panic disorder sample are nonspecific characteristics of psychiatric disorder in general rather than being associated more narrowly and more specifically with panic disorder. Finally, our measure of physician recognition of panic disorder was the patient's report, and this may not be a valid reflection of the physician's actual awareness, or even of what the physician told the patient.
Nonetheless, these findings add to a growing body of evidence that panic disorder imposes a considerable and clinically significant burden on those with the disorder and presents an important challenge to the medical care system. Further investigation is needed to better understand the underdiagnosis of panic disorder in medical settings.
|
FOOTNOTES
|
This work was supported by a research grant from the National Institute of Mental Health (Grant No. MH–40487).
|
REFERENCES
|
-
Wells KB, Stewart A, Hays RD, et al: The functioning and well-being of depressed patients. JAMA 1989; 262:914–919[Abstract]
-
Spitzer RL, Kroenke K, Linzer M, et al: Health-related quality of life in primary care patients with mental disorders. JAMA 1995; 274:1511–1517[Abstract]
-
Frasure-Smith N, Lesperance F, Talajic M: Depression following myocardial infarction. JAMA 1993; 270:1819–1825[Abstract]
-
Carney RM, Rich MW, Freedland KE: Major depressive disorder predicts cardiac events in patients with coronary artery disease. Psychosom Med 1988; 50:627–633[Abstract/Free Full Text]
-
Shear MK, Schulberg HC: Anxiety disorders in primary care. Bull Menninger Clin 1995; 59:A73–A85
-
Markowitz JS, Weissman MM, Ouellette R, et al: Quality of life in panic disorder. Arch Gen Psychiatry 1989; 46:984–992[Abstract]
-
Edlund MH, Swann AC: The economic and social costs of panic disorder. Hosp Community Psychiatry 1987; 38:1277–1279, 1288[Free Full Text]
-
Sherbourne CD, Wells KB, Judd LL: Functioning and well-being of patients with panic disorder. Am J Psychiatry 1996; 153:213–218[Abstract/Free Full Text]
-
Klerman GL, Weissman MW, Ouellette R, et al: Panic attacks in the community: social morbidity and health care utilization. JAMA 1991; 265:742–746[Abstract]
-
Fifer SK, Mathias SD, Patrick DL, et al: Untreated anxiety among adult primary care patients in a health maintenance organization. Arch Gen Psychiatry 1994; 51:740–750[Abstract]
-
Boyd JH: Use of mental health services for the treatment of panic disorder. Am J Psychiatry 1986; 143:1569–1574[Abstract/Free Full Text]
-
Siegel L, Jones WC, Wilson JO: Economic and life consequences experienced by a group of individuals with panic disorder. J Anxiety Disord 1990; 4:201–211
-
Simon G: Psychiatric disorders and functional somatic symptoms as predictors of health care use. Psychiatric Medicine 1992; 10:49–59[Medline]
-
Simpson RJ, Kazmierczak T, Power KG, et al: Controlled comparison of the characteristics of patients with panic disorder. Br J Gen Pract 1994; 44:352–356[Medline]
-
Yingling KW, Wulsin LR, Arnold LM, et al: Estimated prevalences of panic disorder and depression among consecutive patients seen in an emergency department with acute chest pain. J Gen Intern Med 1993; 8:231–235[Medline]
-
Kennedy BL, Schwab JJ: Utilization of medical specialists by anxiety disorder patients. Psychosomatics 1997; 38:109–112[Abstract/Free Full Text]
-
Ormel J, Van Den Brink W, Koeter MWJ, et al: Recognition, management and outcome of psychological disorders in primary care: a naturalistic follow-up study. Psychol Med 1990; 20:909–923[Medline]
-
Barsky AJ, Wyshak G, Klerman GL, et al: The prevalence of hypochondriasis in medical outpatients. Soc Psychiatry Psychiatr Epidemiol 1990; 25:89–94[Medline]
-
Barsky AJ, Wyshak G, Klerman GL: Psychiatric comorbidity in DSM-III-R hypochondriasis. Arch Gen Psychiatry 1992; 49:101–108[Abstract]
-
Barsky AJ, Barnett MC, Cleary PD: Hypochondriasis and panic disorder: boundary and overlap. Arch Gen Psychiatry 1994; 51:918–925[Abstract]
-
Rost K, Smith GR, Hollenberg J: Mental health outcomes, module development and testing: validation of the panic outcomes mode. Kalamazoo, MI, The Upjohn Co., 1994
-
Robins LN, Helzer JE, Croughan J, et al: National Institute of Mental Health Diagnostic Interview Schedule: its development, evolution, and use. Soc Psychiatry Psychiatr Epidemiol 1981; 38:381–389
-
Katon W: Panic Disorder in the Medical Setting. Washington, DC, American Psychiatric Press, Inc., 1991
-
Katon W, Vitaliano PP, Russo J, et al: Panic disorder spectrum of severity and somatization. J Nerv Ment Dis 1987; 175:12–19[Medline]
-
Barsky AJ, Wyshak G, Klerman GL: The relationship between hypochondriasis and medical illness. Arch Intern Med 1991; 151:84–88[Abstract]
-
Jette AM, Davies AR, Cleary PD, et al: The Functional Status Questionnaire: reliability and validity when used in primary care. J Gen Intern Med 1986; 1:143–149[Medline]
-
Katon W, Vitaliano PP, Russo J, et al: Panic disorder: epidemiology in primary care. J Fam Pract 1986; 23:233–239[Medline]
-
Goldberg R, Morris P, Christian F, et al: Panic disorder in cardiac outpatients. Psychosomatics 1990; 31:168–173[Abstract/Free Full Text]
This article has been cited by other articles:
|
|
|
|
 
W. Rief, A. Martin, A. Klaiberg, and E. Brahler
Specific Effects of Depression, Panic, and Somatic Symptoms on Illness Behavior
Psychosom Med,
July 1, 2005;
67(4):
596 - 601.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
P. P. Roy-Byrne, M. G. Craske, M. B. Stein, G. Sullivan, A. Bystritsky, W. Katon, D. Golinelli, and C. D. Sherbourne
A Randomized Effectiveness Trial of Cognitive-Behavioral Therapy and Medication for Primary Care Panic Disorder
Arch Gen Psychiatry,
March 1, 2005;
62(3):
290 - 298.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
B. L. Felker, R. F. Barnes, D. M. Greenberg, E. F. Chaney, M. M. Shores, L. Gillespie-Gateley, M. K. Buike, and C. E. Morton
Preliminary Outcomes From an Integrated Mental Health Primary Care Team
Psychiatr Serv,
April 1, 2004;
55(4):
442 - 444.
[Abstract]
[Full Text]
[PDF]
|
|
|
Get information about faster international access.
Privacy Policy
Copyright © 1999
Academy of Psychosomatic Medicine.
All rights reserved.
Home
| Search
| Current Issue
| Past Issues
| Subscribe
| All APPI Journals
| Help
| Contact Us
|
TOP
ABSTRACT
INTRODUCTION
