Comparison of primary and secondary panic disorder: a preliminary report

Comparison of primary and secondary panic disorder: a preliminary report

Journal ofAffectice Disorders, 27 (1993) 81-86 0 1993 Elsevier Science Publishers B.V. All rights reserved 81 0165-0327/93/$06.00 JAD 00963 Compari...

502KB Sizes 0 Downloads 45 Views

Journal ofAffectice Disorders, 27 (1993) 81-86 0 1993 Elsevier Science Publishers B.V. All rights reserved

81 0165-0327/93/$06.00

JAD 00963

Comparison of primary and secondary panic disorder: a preliminary report 1 Vladan

Starcevic

a, E.H. Uhlenhuth

a, Robert

Kellner

a and Dorothy

Pathak

b

Departments of a Psychiatry and b Family Practice and Community Medicine, University of New Mexico School of Medicine, Albuquerque, New Mexico, USA (Received 5 June 1992) (Revision received 29 September 1992) (Accepted 8 October 1992)

Summary

We examined the onsets of comorbid psychiatric disorders in patients with panic disorder (PD) (N = 54). In 42 patients (77.8%), PD was preceded by another psychiatric condition (secondary PD group), while in 12 patients (22.2%), PD occurred first or was the only diagnosed mental disorder (primary PD group). Patients with primary and secondary PD did not differ with respect to demographic variables, mean ages of onset of PD, mean duration of PD, number of patients with a personality disorder, and number of patients with different DSM-III-R subtypes of PD. Except for the anger and sleep scales of the Hopkins Symptom Checklist 99, patients with primary PD had significantly less self-rated psychopathology; they aisodisplayed less extensive phobic avoidance, and had a lower rate of current psychiatric comorbidity. These findings are discussed in light of the value of the concepts of comqrbidity and primary/ secondary dichotomy.

Key words: Primary panic disorder; Secondary panic disorder; Comorbidity

Introduction

The primary/ secondary distinction proved to be valid and useful for major depression. The chronological order in which depression develops

Correspondence to: Vladan Starcevic, Department of Psychiatry, University of New Mexico School of Medicine, 2400 Tucker, N.E., Albuquerque, NM 87131, USA. t Portions of this work were presented at the Annual Meeting of the American Psychiatric Association in Washington, DC, May 1992.

in relation to other mental disorders was found to be associated with specific differences pertaining to clinical, family, and laboratory features, as well as to course and treatment (Grove et al., 1987a,b; Winokur, 1990). The issue of primary/ secondary dichotomy has not been directly explored in anxiety disorders. There was only one recent report (Garvey et al., 1991), which concluded that the concept of primary/secondary anxiety disorder did not appear useful when applied to the study of comorbidity in psychiatric inpatients. Outpatients with pri-

82

mary and secondary diagnoses of specific anxiety disorders have not been studied systematically. The purpose of this pilot study was to compare patients with the chronologically primary and secondary panic disorder (PD) on the basis of their demographic and psychological test characteristics. We also sought to examine in a preliminary manner whether classification of PD in terms of the primary/secondary distinction might be conceptually or clinically useful. We hypothesized that the clinical presentation of the secondary PD patients might be colored by the presence, complications and/or impairment of the primary disorder(s), and that therefore, secondary PD might be a more severe condition in comparison with primary PD. Subjects and methods The characteristics of patients in our sample and details of the method have been described elsewhere (Starcevic et al., 1992a). A summary of the methods and description of subjects is as follows: The subjects for this study were outpatients evaluated for possible participation in a drug treatment study of PD. They were recruited through newspaper advertisements, referrals from other physicians, flyers, and word of mouth. The inclusion criterion for participation was presence of the DSM-III-R PD (with or without agoraphobia (AG)). Patients were excluded if they had a current, principal diagnosis of major depression or alcoholism, or a lifetime history of schizophrenia, bipolar disorder, delusional disorder, any other psychotic disorder, or organic mental syndrome. Diagnoses were established by the means of the Structured Clinical Interview for DSM-III-R Disorders, Upjohn Version - Revised (SCID-UPR; Spitzer and Williams, 1988). Interviews were performed by two of the authors (VS. and E.H.U.) at the time of the initial evaluation (screening) for participation in the drug treatment study. The ages of onset of PD and comorbid disorders were obtained as accurately as possible. The age of onset was defined as the age when, retrospectively determined, all criteria for a given diagnosis had been met. The age of onset of PD did not necessarily coincide with the time

of the first panic attack, because the presence of panic attacks does not automatically denote the presence of the diagnosis of PD. In addition, the retrospective method is not suitable for examination of the psychopathology at the level of subthreshold diagnoses. When PD and another condition had an onset in the same year (at the same age), we made further inquiries to determine chronological primacy. Because of the striking nature of the first panic attack, most patients were able to recall the time of its occurrence in temporal relation to other emotional problems and disorders starting in the same year. On the basis of the age of onset of PD in relation to any comorbid disorder, patients were divided into a primary PD group and a secondary PD group. In patients with primary PD, panic preceded any associated condition by at least two months, or was the only diagnosed mental disorder. Secondary PD was diagnosed if PD started at least two months after another psychiatric condition. Of 54 patients who participated in the study, there were 12 (22.2%) with primary PD and 42 (77.8%) with secondary PD. There were 6 patients in whom AG developed prior to PD, but in these patients AG had in turn been preceded by other conditions, so that they were all classified in the secondary PD group. Other instruments used for comparisons of the primary and secondary PD patients were the Structured Clinical Interview for DSM-III-R Personality Disorders (SCID-II; Spitzer et al., 1987), the Hopkins Symptom Checklist 90 (HSCL; Lipman et al., 1979), and the Fear Questionnaire (FQ; Marks and Mathews, 1979). For comparisons of means on the HSCL and FQ scales between groups of patients with primary and secondary PD, we used the MannWhitney test because of the skewed distribution of the data. The chi-square test was used for analyses for all categorical variables in order to compare distributions of proportions between the groups (Zar, 1984). Results Patients with primary and secondary PD did not differ with respect to demographic variables,

83 TABLE

1A

TABLE

Demographic characteristics secondary panic disorder

Mean current age (in years) % Female % Never married % Whites % Employed % Educated above the high school level

of patients

with

primary

Primary panic disorder (N= 12)

Secondary panic disorder (N = 42)

37.0 (SD = 8.4) 66.7 8.3 75.0 66.7

35.0 (SD = 8.7) 52.4 31.0 61.9 78.6

66.7

54.8

and

2

HSCL scores disorder

in patients

panic

Primary panic disorder (N= 12) mean (SD)

Secondary panic disorder (N= 42) mean (SD)

P *

Somatization

1.21 (0.39) 0.82 (0.80) 0.48 (0.54) 1.Ol (0.86) 1.53 (0.81) 0.61 (0.70) 0.85 (0.79) 0.43 (0.42) 0.33 (0.35) 1.47 (1.31)

1.72 (0.74) 1.74 (1.05) 1.41 (0.91) 1.72 (0.94) 2.42 (0.79) 0.86 (0.78) 1.69 (1.19) 0.98 (0.78) 1.06 (0.91) 2.23 (1.30)

0.02

0.89 (0.52)

1.59 (0.68)

0.001

Sensitivity Depression

TABLE

and secondary

HSCL Scales

Obsessive-Compulsiveness

including age, sex, marital status, ethnic origin, current employment status, and educational level (Table 1A). There was no significant difference in the mean duration of PD between the two groups, and the mean ages of onset of PD in the primary and secondary PD groups were virtually identical (Table 1B). Patients with primary and secondary PD did not differ significantly in terms of the DSM-III-R subtypes of PD and character pathology (Table lB), although there was a trend for secondary PD patients to be more agoraphobic, and there were more secondary PD patients who had a diagnosis of a personality disorder. Patients with primary PD had less self-rated psychopathology, as measured by the HSCL; except for the anger and sleep scales, differences on all the scales were significant (Table 2). The primary PD patients also scored lower on the avoidance scales for blood-injury phobia, social phobia, and overall phobia (Table 3).

with primary

Anxiety Anger Phobia Psychoticism Paranoia Sleep

Total

* P-values based on the N.S. = non-significant.

Mann-Whitney

0.01 0.002 0.02 0.003 N.S. 0.03 0.009 0.006 N.S.

test with

df = 52;

The comorbidity data for the whole sample of panic patients were presented in detail elsewhere (Starcevic et al., 1992b,c). The most common

lb

Age of onset of panic disorder, duration primary and secondary panic disorder

of panic

disorder,

Mean age of onset of panic disorder (in years) Mean duration of panic disorder (in months) % Patients without agoraphobia % Patients with mild agoraphobia % Patients with moderate and severe agoraphobia % Patients with a personality disorder Differences

between

primary

and secondary

panic disorder

personality

disorders,

and subtypes

of panic

disorder

Primary panic disorder (N= 12)

Secondary (N=42)

28.8 (SD = 14.5) 97.6 (SD = 137.5) 50.0 33.3 16.7 25.0

28.9 (SD = 11.2) 66.7 (SD = 80.3) 31.0 21.4 47.6 42.9

were not significant

in patients

panic disorder

for any of the above variables.

with

84 TABLE

3

Fear questionnaire (FQ) scores secondary panic disorder

FQ Scales

Agoraphobia Blood-Injury

Phobia

Social Phobia

Overall

Phobia

in patients

with primary

Primary panic disorder (N=12) mean

Secondary panic disorder (N= 42) mean

(SD)

(SD)

12.00 (11.70) 8.17 (5.04) 8.50 (5.60)

16.26 (11.38) 16.14 (8.98) 17.67 (9.09)

28.67 (18.07)

50.07 (23.25)

* P-values based on the N.S. = non-significant.

Mann-Whitney

and

P *

N.S. 0.003

comorbidity rate among the primary PD patients was low (25%) in comparison with the lifetime comorbidity rate among the patients with secondary PD, which was, by definition, 100%. The subset of nine primary panic patients without any lifetime comorbidity had a later mean age of onset of PD (33.1 & 13.9 years); the mean age of onset of PD was most delayed in the six primary PD patients without comorbid conditions and without AG (38.0 k 10.2 years). Discussion

< 0.001

0.003

test with

df = 52;

comorbid diagnosis was generalized anxiety disorder (51.9%). The lifetime rates of comorbidity for other common diagnoses were 48.1% for major depressive episodes, 44.4% for simple phobia, 40.7% for social phobia, 37.0% for alcohol abuse, and 31.5% for drug abuse. All comorbid conditions, except for major depression, tended to precede the onset of PD. The most common primary conditions in the secondary PD patients were simple phobia (13 cases; 30.9%) and social phobia (11 cases; 26.2%). Less frequently, the very first diagnoses that patients with secondary PD received in retrospect, were drug abuse (5 patients), alcohol abuse (4 patients), generalized anxiety disorder (4 patients), obsessive-compulsive disorder (2 patients), major depression (2 patients), and dysthymia (1 patient). Of the total number (N = 142) of comorbid diagnoses in the secondary PD group, only 3.5 (24.6%) diagnoses were made after the onset of PD. Of 42 patients with secondary PD, 22 (52.4%) developed another disorder after the onset of PD. The most common among these disorders was major depression - in 14 (63.6%) of 22 such patients. Only 3 patients with primary PD subsequently developed other disorders, so that the lifetime

The results of this study should be interpreted with caution, because of possible errors in recalling the ages of onset of PD and comorbid conditions, and the small sample of patients with primary PD. Also, this was a selected population (volunteers for a drug treatment study), and the findings are not necessarily applicable to panic patients in general. The cross-sectional design of the study is its main limitation, because we were unable to follow primary PD patients for a sufficient period of time, during which they could develop specific comorbid illnesses, on the basis of which they could then be compared more adequately to secondary PD patients. Studies of this nature would be better designed if primary PD patients with a specific secondary condition could be compared to secondary PD patients whose primary condition had been the same comorbid disorder. On the other hand, there were relatively few psychiatric conditions appearing after the onset of PD - both when PD was a primary and secondary illness. This finding may indicate that except for major depression (which tended to follow the onset of PD more frequently), PD is rarely conducive to other anxiety disorders and substance abuse. Thus, PD appears to be more likely to be preceded by another psychiatric disorder than to be a chronologically primary condition. This finding implies that some primary disorders (e.g., simple phobia, social phobia, substance abuse) may represent a specific predisposition for the development of PD. We have indicated that such etiologic relatedness may exist for primary substance abuse only (Starcevic et al., 1992~).

85

The secondary PD patients displayed a more diffuse psychopathology and appeared more ill. The severity of the self-rated psychopathology and severity of phobic avoidance in secondary PD appear to be a consequence of those disorders that preceded the onset of PD, and were still present at the time of evaluation. This is in accordance with the finding that in many cases, primary disorders were social phobia and simple phobia. The severity of illness in the secondary PD patients may also reflect current presence of other disorders that preceded PD, or that appeared after the onset of PD. Thus, the currently high comorbidity rate rather than the sequence of appearance of PD, may be the likely explanation for the severity of illness in patients with secondary PD. Likewise, the low level of the selfrated psychopathology and fewer phobic features among the primary PD patients may be related more to the low comorbidity rate in this group (and weaker association with social phobia and simple phobia) than to the fact that PD was a primary condition. The causes of the low comorbidity rate among the primary PD patients are not clear. A somewhat lower prevalence of AG among these patients might have been a factor (Starcevic et al., 199213). The age of onset of PD may also play a role here. For example, when primary PD developed later in life, it tended to present as a relatively ‘isolated’ disorder (without the subsequent appearance of other psychiatric conditions), especially in those patients without AG. Thus, the more ‘pure’ primary PD is (without AG and without comorbid conditions), the more likely it is to have begun at a later age. Another reason for the low comorbidity rate in patients with a lateonset, primary PD may be that these patients have already passed through a life period of greatest probability for the onset of many comorbid conditions (e.g., for simple phobia, social phobia and substance abuse). Patients who develop PD later in life, and who do not have an associated AG and other Axis I conditions, may suffer from a distinct subtype of PD or from a distinct disorder. Such a disorder, or a variant of PD, may be etiologically distinct, but its etiology does not have to be related to the primary nature of PD. It was recently reported

that patients with a late onset of panic attacks had fewer panic symptoms and displayed less agoraphobic avoidance (Sheikh et al., 1991). Therefore, such patients may be more amenable to treatment, particularly if PD was their first psychiatric condition. Inasmuch as the observed differences between patients with primary and secondary PD might be better accounted for by other factors, such as higher comorbidity rates in patients with secondary PD, the results of this pilot study do not suggest that the primary/ secondary dichotomy has much value in PD, unlike the primary/ secondary distinction in major depression. The subtyping of PD on the basis of the presence/ absence of AG, and age of onset of PD, may then be more meaningful and clinically relevant than the classification on the basis of the chronological appearance of PD. Future studies should explore issues such as homogeneity of the concepts of primary and secondary PD; a related question pertains to possible differences between subgroups of secondary PD patients who had different preexisting conditions. Patients with primary and secondary PD should also be compared in terms of their family constellations and the course of illness. Studies should be broadened to include patients with other principal diagnoses, who have a comorbid PD - both as a primary and secondary condition. Finally, the impact of the primary/secondary dichotomy on the patterning of comorbidity in PD remains to be examined in more detail. Acknowledgements This work was supported by The Upjohn Company, Kalamazoo, Michigan. Anonymous reviewers provided useful comments on an earlier version of the paper. References Garvey, M., Noyes, R., Anderson, D. and Cook, B. (1991) Examination of comorbid anxiety in psychiatric inpatients. Compr. Psychiatry 32, 277-282. Grove, W.M., Andreasen, N.C., Clayton, P.J., Winokur, G. and Coryell, W.H. (1987a) Primary and secondary affective disorders: baseline characteristics of unipolar patients. J. Affect. Disord. 13, 249-257.

86 Grove, W.M., Andreasen, N.C., Winokur, G., Clayton, P.J., Endicott, J. and Coryell, W.H. (1987b) Primary and secondary affective disorders: unipolar patients compared on familial aggregation. Comr. Psychiatry 28, 113-126. Lipman, R.S., Covi, L. and Shapiro, A.K. (1979) The Hopkins symptom checklist (HSCL): factors derived from the HSCL-90. J. Affect. Disord. 1, 9-24. Marks, I.M. and Mathews, A.M. (1979) Brief standard selfrating for phobic patients. Behav. Res. Ther. 17, 263-267. Sheikh, J.I., King, R.J. and Taylor, C.B. (1991) Comparative phenomenology of early-onset versus late-onset panic attacks: a pilot survey. Am. J. Psychiatry 148, 1231-1233. Spitzer, R.L. and Williams, J.B.W. (1988) Structured Clinical Interview for DSM-III-R, Upjohn Version, Revised (SCID-UP-R). New York State Psychiatric Institute, New York, NY. Spitzer, R.L., Williams, J.B.W. and Gibbon, M. (1987) Structured Clinical Interview for DSM-III-R Personality Disor-

ders (SCID-II). New York State Psychiatric Institute, New York, NY. Starcevic, V., Kellner, R., Uhlenhuth, E.H. and Pathak, D. (1992a) Panic disorder and hypochondriacal fears and beliefs. J. Affect. Disord. 24, 73-85. Starcevic, V., Uhlenhuth, E.H., Kellner, R. and Pathak, D. (1992b) Patterns of comorbidity in panic disorder and agoraphobia. Psychiatry Res. 42, 171-183. Starcevic, V., Uhlenhuth, E.H., Kellner, R. and Pathak, D. (1992~) Comorbidity in panic disorder: II. Chronology of appearance and pathogenic comorbidity. Psychiatry Res. (in press). Winokur, G. (1990) The concept of secondary depression and its relationship to comorbidity. Psychiatr. Clin. N. Amer. 13, 567-583. Zar, J.H. (1984) Biostatistical Analysis, 2nd edn. Prentice-Hall, Englewood Cliffs, NJ.