The Relationship Between Anxiety and Depression: A Clinical Comparison of Generalized Anxiety Disorder, Dysthymic Disorder, Panic Disorder, and Major Depressive Disorder Molly M. Shores, Tarika Glubin, Deborah
S. Cowley,
Stephen
R. Dager, Peter P. Roy-Byrne,
and David L. Dunner This study examined the relationship between the chronic disorders, generalized anxiety disorder (GAD) and dysthymic disorder (DD), and the more acute disorders, panic disorder (PD) and major depressive disorder (MDD) in 110 psychiatric outpatients with diagnoses of either PD, MDD, GAD, or DD. Pure, mixed, and early-/late-onset forms of the chronic disorders were compared with each other and then with PD and MDD on clinical measures and psychiatric history. Minimal differences were found between pure GAD and mixed GAD or between pure DD and mixed DD. The chronic disorders, DD and GAD, had distinct clinical symptom profiles when compared with each
T
HE RELATIONSHIP between depression and anxiety has been the subject of extensive debate since the 193Os, when Lewis wrote his classic report on the topic.’ Studies on the relationship between depression and anxiety have emphasized one of three different conceptual models: anxiety and depression as a continuum in which anxiety and depressive symptoms overlap and blend together; anxiety and depression as distinct illnesses with clear boundaries; and a mixed anxiety-depressive disorder as a distinct disorder differing from the pure forms of these disorders.‘-’ Anxiety and depressive symptoms frequently coexist to varying degrees within major depression (MDD), dysthymic disorder (DD), panic disorder (PD), and generalized anxiety disorder (GAD). Studies examining the anxiety-depression relationship have primarily focused on MDD and PD. The relationship between anxiety and depression in the chronic anxiety and depressive disorders, GAD and DD, has been unclear. One study has shown that the overlap between depressive and anxiety symptoms is greater in symptomatic patients not clearly meeting criteria for an axis I mood or anxiety disorder.‘O This suggests that as severity of psychopathology increases, and finally reaches threshold for an axis I disorder, the presentation of anxiety and depression becomes more distinct. Thus, a greater overlap would be expected in patients with less severe, even subclinical disorders.” Comprehensive Psychiatry, Vol. 33, No. 4 (July/August),
other and appeared more closely related to their parent disorders than to each other. However, despite these similarities, there were significant differences between DD and MDD in contrast to the minimal differences between GAD and PD, providing less support for GAD as a valid diagnostic category separate from PD. Comparisons of early-/late-onset DD and GAD showed more severe symptoms in late-onset DD, in contrast to more severe symptoms in early-onset GAD. These varying patterns of symptom severity may warrant study for further syndromal delineation. Copyright 0 1992 by W.B. Saunders Company
The chronic disorders, GAD and DD, have been alternatively conceptualized as variants of the same disorder, as nonspecific indicators of psychiatric distress associated with chronicity and severity, as character disorders, as subaffective disorders, as prodromes to the major disorders, or as incomplete remissions from the major disorders. ‘Jo-” GAD is a prodrome to panic in a significant number of PD patients and a history of prior major depression in DD is common.14Jx-1” Studies have been complicated by the apparent heterogeneity present in DD. DSM-III-R has defined distinct DD groups based on early or late onset of symptoms. Akiskal has characterized early-onset dysthymia as a characterologic dysphoria and late-onset dysthymia as a residual chronic dysphoria occurring primarily after episodes of major depression.13 Akiskal divides early-onset dysthymia further into a character spectrum disorder and a subaffective disorder. The subaffective disorder patients generally have biologic correlates similar to MDD and are responsive to antidepressant medication. The character spectrum disorder patients tend to be
From the Center for Anxiety und Depression, Department of Psychiatry and Beha~kval Sciewes, Uni~wsity of Washington, Seattle. WA. Address reprint requests to Dal,id L. Dunner, M.D., 4225 Roosevelt Way, NE, Suite 306. Seattle, WA 98105. Copyright 0 1992 6~ W;B. Saunders Company OOIO-440X19213304-0005$03.0010
1992: pp 237.244
237
238
nonresponsive to medication and have a history of alcohol abuse and sociopathy. Klein et al. found that patients with earlyonset DD had more psychiatric comorbidity, more history of substance abuse and personality disorders, more severe depression, greater history of prior psychiatric treatment, and more relatives with a history of affective disorder than patients with late onset DD.‘* They proposed that early-onset DD is closely related to MDD and may even be a more severe form of affective disorder with a worse prognosis than MDD.20 However, McCullough et al. compared earlyand late-onset DD and found no significant differences between the two groups on symptom measures, onset patterns, or cognitive measures.21 In DSM-III-R, dysthymia is categorized further into primary dysthymia and dysthymia secondary to physical illness or a chronic, nonaffective psychiatric illness. This is in contrast to the conceptualization of MDD, which has moved away from a dichotomy based on endogenous versus exogenous etiology. One study has found an increased rate of suicide attempts in primary DD as opposed to secondary DD.22 GAD also appears to be a heterogenous group. Koehler et al. proposed dividing GAD into a “mixed” group and a pure group, with the mixed group having a history of panic attacks.23 Koehler et al. found that mixed GAD patients had increased autonomic lability and increased somatic symptoms as compared with pure GAD patients. Studies by Cowley et al. and Dager et al. have shown that patients with a history of chronic anxiety and infrequent panic attacks are more similar to PD patients than to controls or patients with pure GAD.24+25Cowley et al. showed that patients with chronic anxiety and infrequent panic attacks panicked with lactate infusions, similarly to PD patients. Dager et al. noted that patients with chronic anxiety and infrequent panic attacks had rates of mitral valve prolapse that were comparable to PD patients, but higher than pure GAD patients.25 Breslau found increased autonomic lability in patients who experienced GAD for greater than 24 months versus patients who had GAD for less than 24 monthsI Several studies have noted the high comorbidity of other psychiatric disorders with DD and
SHORES ET AL
GAD, with rates of pure GAD and DD ranging from 0.5% to 10% of the total sample.13,26 Breslau noted that GAD and DD either occurred after other psychiatric disorders had been diagnosed or coincided with them.17 Rarely did GAD and DD occur as the primary psychiatric diagnosis. Due to the low prevalence of pure DD and pure GAD, there have been few attempts to study these conditions. The purpose of this study was to examine the relationship between the chronic disorders, GAD and DD, and their relationship to PD and MDD. We sought to clarify the nature of these disorders by comparing pure, mixed, and early-/ late-onset DD and GAD with each other and then with MDD, PD, and nonpsychiatrically ill controls. METHOD The patients in this study were predominantly middleclass males and females participating in psychopharmacology studies at Harborview Medical Center between 1986 and 1988. The control group was selected from nonpsychiatrically ill patients participating in a sexual dysfunction drug study at Harborview Medical Center. All subjects participated in these projects on a voluntary basis with informed consent. The patients met one of the following DSM-III-R diagnoses: major depression (N = 20), panic disorder (N = 20), generalized anxiety disorder (N = 40), or dysthymic disorder (N = 30). Their diagnosis was established from a semistructured interview conducted by one of four faculty psychiatrists who had expertise in anxiety and depressive disorders. The diagnostic interviews also included data on treatment history, course of illness, psychosocial history, medical history, demographic factors, and family history. Family psychiatric history of first-degree relatives was obtained from all 110 patients using the same semistructured interview format. For unipolar and bipolar disorder, the morbid risk of illness in first-degree relatives was determined. The Stromgren method was used to calculate bezugziffers (BZs), and the morbid risk of illness was then determined for unipolar and bipolar disorders separately.*’ For substance abuse and anxiety disorders, the percentage of affected relatives was calculated. Subjects for each classification were selected from a consecutive series of patients. The DD and GAD groups were intentionally made larger to allow for further subdivision into subgroups of pure GAD and mixed GAD+ and pure DD and mixed DD+. DD+ and GAD+ subgroups consisted of patients having GAD or DD plus a lifetime, but no current, history of another psychiatric disorder. Pure GAD and pure DD were compared with GAD+ and DD+, respectively. Subsequent analyses included comparisons of pure DD and pure GAD, pure GAD and PD/GAD+, pure GAD and PD, and pure DD and MDD. Each of these analyses compared the groups on the age of
239
ANXIETY AND DEPRESSION
first treatment, prior treatment history (medication or psychotherapy), duration of illness, suicide attempts, past psychiatric history, family psychiatric history, substance abuse history, caffeine use, and revised Hamilton-Anxiety (HAM-A) and Hamilton-Depression (HAM-D) scores.28 Following these analyses, DD and GAD groups were divided into early- and late-onset groups and compared and contrasted on the variables listed above. Early-onset DD has been defined in DSM-III-R as onset of symptoms at age less than 21. However, prior studies defined early-onset DD as symptom onset at age less than 25.‘” Therefore, the analysis was performed twice with early-onset DD being defined as symptom onset at age less than 21 and then reanalyzed with early-onset being defined as symptom onset at age less than 25. To be consistent, GAD was also analyzed in this manner, as we were unaware of prior studies examining GAD by symptom onset. Analysis of variance (ANOVA) was calculated using Duncan, Tukey-HSD, and Scheffe post-hoc tests for threeway comparisons. The homogeneity of variances was analyzed with Cochran’s C and Bartlett’s Box F. Family history data were analyzed with chi-square. RESULTS
There were no significant differences between the control group and any of the psychiatric groups in regard to age, sex, and marital status. There were a greater number of women with MDD (N = 16) than PD (N = 8) (xl = 5.10; df= 1; P < .02). The mean age for each group was in the late 30s. Pure Vems Mixed disorders (DD v DD+; GAD v GAD+)
LX3 Versus GAD
The lifetime psychiatric diagnoses for DD+ and GAD+ patients are listed in Table 1. The most common lifetime diagnosis in the DD+ group was MDD and the most common lifetime psychiatric diagnosis in the GAD+ group was panic attacks. The comparisons of the mixed disorders with the pure disorders showed few significant differences. There were no significant differences in Table 1. Lifetime Psychiatric Diagnoses Complicating DO+ and GAD+
Infrequent panic attacks
DD+ (N = 16)
GAD+ (N = 25)
% N
% N
-
Major depression
94% (15)
Eating disorder
12% (2)
Generalized anxiety disorder
12% (2)
72% (18) 56% (14) -
Post traumatic stress disorder Hypomania
6% (I) -
any of the following variables: age, sex, marital status, age at first symptoms and treatment, duration of illness, history of suicide attempts, alcohol and drug use history, and revised total HAM-A and HAM-D scores. Most of the differences were inherent in how these groups were defined. DD+ patients were more likely to have had prior treatment (psychotherapy or medication) (F = 11.2; df = 1,28; P < .002), more frequently had a history of other psychiatric illnesses (F = 196; df= 1,28; P < .OOl), and had lower ratings for diurnal variation than the pure DD group (F = 4.07; df = 1,28; P < .05). GAD+ patients were more likely to be treated with medication at index presentation (F = 5.66; df = 1, 38; P < .02) and more frequently had a history of major depression (F = 18.1; df = 1,38; P < .OOOl) or other psychiatric illnesses (F = 45.1; df = 1,38; P < .OOl) than the GAD group. In contrast to previous studies, there were no StatisticalIy signi~cant differences between pure GAD and GAD+ on any of the items on the revised HAM-A or HAM-D scales, including items commonly associated with symptoms of panic attacks such as respiratory, cardiovascuIar. and autonomic symptoms.
4% i?)
Pure DD and pure GAD groups had distinct symptom profiles as determined by total HAM-A and HAM-D scores (see Table 2). On the revised HAM-A scale, 47% of the items were significantly different, with GAD patients having more severe sensory, autonomic, gastrointestinal, muscular, and respiratory symptoms; and more severe somatic and psychic anxiety, anxious behavior, anxious mood, and agitation than DD patients. On the revised HAM-D scale, 40% of the items were significantly different, with DD patients having more severe depressed mood, guilt, retardation, and greater impairment in work and activities than the GAD patients. There was also a trend for more severe diurnal variation in the pure DD group than in the pure GAD group. Pure GAD and GAD+ patients were combined and then compared with a combined group of pure DD and DD+ patients. This was done to see if the distinctions between pure DD
240
SHORES ET AL
Table 2. Comparison of Pure GAD With Pure DD Pure GAD
Pure DD
(N = 15)
(N = 14)
Age at first symptoms (yr)
20.9
20.8
F = ,001; df = 1,28; P = .97
Age at first treatment (yr)
32.1
31.4
F=.O3:df=1,28;P=.86
Any prior treatment N(%)
3 (20%)
3 (21%)
Total revised Ham-A
26.3
14.4
F = 21.90; df = f,28; P < .Ol
Total revised Ham-D
6.8
11.0
F = 11.64; df=
Anxious mood
2.6
1.5
F = 8.13; df = 1,28; P < .Ol
Muscular symptoms
2.0
1.0
F = 5.43; df = 1,28; P < .03
Sensory symptoms
1.7
0.3
F = 14.24; df = 1.28; P < .OOl
Respiratory symptoms
1.3
0.3
F = 5.93; df = 1.28; P < -02
GI symptoms
1.9
0.6
F = 10.11; df = 1.28; P < .Ol
Autonomic symptoms
1.7
0.7
F = 7.87; c/f = 1.28; P < .Ol
Anxious behavior
2.3
1.1
F = 30.34: df = 1,28; P < .Ol
Agitation
1.4
0.5
F = 12.09; df = 1,28; P < .03
Psychic anxiety
2.2
1.6
F = 5.58; df = .28; P < .03
Somatic anxiety
2.0
0.9
F = 14.4: df = 1,28; P < .Ol
Statistics
F = ,008; df = 1,28; P = .93 1.28: P < .Ol
Specific revised Ham-A items
Specific revised Ham-D items Depressed mood
1.1
2.1
F = 8.99; df = 1‘28: P < .Ol
Guilt
0.7
1.6
F = 14.85; df = 1,28; P < ,001
impairment in work/activities
0.8
1.8
F = 15.48; df = 1,28: P < .OOl
Diurnal variation
0.3
0.7
F = 3.46; df = 1,28; P = .07
Retardation
0.4
0.9
F=4,39;&=1,28;P
S. Cowley,
Stephen
R. Dager, Peter P. Roy-Byrne,
and David L. Dunner This study examined the relationship between the chronic disorders, generalized anxiety disorder (GAD) and dysthymic disorder (DD), and the more acute disorders, panic disorder (PD) and major depressive disorder (MDD) in 110 psychiatric outpatients with diagnoses of either PD, MDD, GAD, or DD. Pure, mixed, and early-/late-onset forms of the chronic disorders were compared with each other and then with PD and MDD on clinical measures and psychiatric history. Minimal differences were found between pure GAD and mixed GAD or between pure DD and mixed DD. The chronic disorders, DD and GAD, had distinct clinical symptom profiles when compared with each
T
HE RELATIONSHIP between depression and anxiety has been the subject of extensive debate since the 193Os, when Lewis wrote his classic report on the topic.’ Studies on the relationship between depression and anxiety have emphasized one of three different conceptual models: anxiety and depression as a continuum in which anxiety and depressive symptoms overlap and blend together; anxiety and depression as distinct illnesses with clear boundaries; and a mixed anxiety-depressive disorder as a distinct disorder differing from the pure forms of these disorders.‘-’ Anxiety and depressive symptoms frequently coexist to varying degrees within major depression (MDD), dysthymic disorder (DD), panic disorder (PD), and generalized anxiety disorder (GAD). Studies examining the anxiety-depression relationship have primarily focused on MDD and PD. The relationship between anxiety and depression in the chronic anxiety and depressive disorders, GAD and DD, has been unclear. One study has shown that the overlap between depressive and anxiety symptoms is greater in symptomatic patients not clearly meeting criteria for an axis I mood or anxiety disorder.‘O This suggests that as severity of psychopathology increases, and finally reaches threshold for an axis I disorder, the presentation of anxiety and depression becomes more distinct. Thus, a greater overlap would be expected in patients with less severe, even subclinical disorders.” Comprehensive Psychiatry, Vol. 33, No. 4 (July/August),
other and appeared more closely related to their parent disorders than to each other. However, despite these similarities, there were significant differences between DD and MDD in contrast to the minimal differences between GAD and PD, providing less support for GAD as a valid diagnostic category separate from PD. Comparisons of early-/late-onset DD and GAD showed more severe symptoms in late-onset DD, in contrast to more severe symptoms in early-onset GAD. These varying patterns of symptom severity may warrant study for further syndromal delineation. Copyright 0 1992 by W.B. Saunders Company
The chronic disorders, GAD and DD, have been alternatively conceptualized as variants of the same disorder, as nonspecific indicators of psychiatric distress associated with chronicity and severity, as character disorders, as subaffective disorders, as prodromes to the major disorders, or as incomplete remissions from the major disorders. ‘Jo-” GAD is a prodrome to panic in a significant number of PD patients and a history of prior major depression in DD is common.14Jx-1” Studies have been complicated by the apparent heterogeneity present in DD. DSM-III-R has defined distinct DD groups based on early or late onset of symptoms. Akiskal has characterized early-onset dysthymia as a characterologic dysphoria and late-onset dysthymia as a residual chronic dysphoria occurring primarily after episodes of major depression.13 Akiskal divides early-onset dysthymia further into a character spectrum disorder and a subaffective disorder. The subaffective disorder patients generally have biologic correlates similar to MDD and are responsive to antidepressant medication. The character spectrum disorder patients tend to be
From the Center for Anxiety und Depression, Department of Psychiatry and Beha~kval Sciewes, Uni~wsity of Washington, Seattle. WA. Address reprint requests to Dal,id L. Dunner, M.D., 4225 Roosevelt Way, NE, Suite 306. Seattle, WA 98105. Copyright 0 1992 6~ W;B. Saunders Company OOIO-440X19213304-0005$03.0010
1992: pp 237.244
237
238
nonresponsive to medication and have a history of alcohol abuse and sociopathy. Klein et al. found that patients with earlyonset DD had more psychiatric comorbidity, more history of substance abuse and personality disorders, more severe depression, greater history of prior psychiatric treatment, and more relatives with a history of affective disorder than patients with late onset DD.‘* They proposed that early-onset DD is closely related to MDD and may even be a more severe form of affective disorder with a worse prognosis than MDD.20 However, McCullough et al. compared earlyand late-onset DD and found no significant differences between the two groups on symptom measures, onset patterns, or cognitive measures.21 In DSM-III-R, dysthymia is categorized further into primary dysthymia and dysthymia secondary to physical illness or a chronic, nonaffective psychiatric illness. This is in contrast to the conceptualization of MDD, which has moved away from a dichotomy based on endogenous versus exogenous etiology. One study has found an increased rate of suicide attempts in primary DD as opposed to secondary DD.22 GAD also appears to be a heterogenous group. Koehler et al. proposed dividing GAD into a “mixed” group and a pure group, with the mixed group having a history of panic attacks.23 Koehler et al. found that mixed GAD patients had increased autonomic lability and increased somatic symptoms as compared with pure GAD patients. Studies by Cowley et al. and Dager et al. have shown that patients with a history of chronic anxiety and infrequent panic attacks are more similar to PD patients than to controls or patients with pure GAD.24+25Cowley et al. showed that patients with chronic anxiety and infrequent panic attacks panicked with lactate infusions, similarly to PD patients. Dager et al. noted that patients with chronic anxiety and infrequent panic attacks had rates of mitral valve prolapse that were comparable to PD patients, but higher than pure GAD patients.25 Breslau found increased autonomic lability in patients who experienced GAD for greater than 24 months versus patients who had GAD for less than 24 monthsI Several studies have noted the high comorbidity of other psychiatric disorders with DD and
SHORES ET AL
GAD, with rates of pure GAD and DD ranging from 0.5% to 10% of the total sample.13,26 Breslau noted that GAD and DD either occurred after other psychiatric disorders had been diagnosed or coincided with them.17 Rarely did GAD and DD occur as the primary psychiatric diagnosis. Due to the low prevalence of pure DD and pure GAD, there have been few attempts to study these conditions. The purpose of this study was to examine the relationship between the chronic disorders, GAD and DD, and their relationship to PD and MDD. We sought to clarify the nature of these disorders by comparing pure, mixed, and early-/ late-onset DD and GAD with each other and then with MDD, PD, and nonpsychiatrically ill controls. METHOD The patients in this study were predominantly middleclass males and females participating in psychopharmacology studies at Harborview Medical Center between 1986 and 1988. The control group was selected from nonpsychiatrically ill patients participating in a sexual dysfunction drug study at Harborview Medical Center. All subjects participated in these projects on a voluntary basis with informed consent. The patients met one of the following DSM-III-R diagnoses: major depression (N = 20), panic disorder (N = 20), generalized anxiety disorder (N = 40), or dysthymic disorder (N = 30). Their diagnosis was established from a semistructured interview conducted by one of four faculty psychiatrists who had expertise in anxiety and depressive disorders. The diagnostic interviews also included data on treatment history, course of illness, psychosocial history, medical history, demographic factors, and family history. Family psychiatric history of first-degree relatives was obtained from all 110 patients using the same semistructured interview format. For unipolar and bipolar disorder, the morbid risk of illness in first-degree relatives was determined. The Stromgren method was used to calculate bezugziffers (BZs), and the morbid risk of illness was then determined for unipolar and bipolar disorders separately.*’ For substance abuse and anxiety disorders, the percentage of affected relatives was calculated. Subjects for each classification were selected from a consecutive series of patients. The DD and GAD groups were intentionally made larger to allow for further subdivision into subgroups of pure GAD and mixed GAD+ and pure DD and mixed DD+. DD+ and GAD+ subgroups consisted of patients having GAD or DD plus a lifetime, but no current, history of another psychiatric disorder. Pure GAD and pure DD were compared with GAD+ and DD+, respectively. Subsequent analyses included comparisons of pure DD and pure GAD, pure GAD and PD/GAD+, pure GAD and PD, and pure DD and MDD. Each of these analyses compared the groups on the age of
239
ANXIETY AND DEPRESSION
first treatment, prior treatment history (medication or psychotherapy), duration of illness, suicide attempts, past psychiatric history, family psychiatric history, substance abuse history, caffeine use, and revised Hamilton-Anxiety (HAM-A) and Hamilton-Depression (HAM-D) scores.28 Following these analyses, DD and GAD groups were divided into early- and late-onset groups and compared and contrasted on the variables listed above. Early-onset DD has been defined in DSM-III-R as onset of symptoms at age less than 21. However, prior studies defined early-onset DD as symptom onset at age less than 25.‘” Therefore, the analysis was performed twice with early-onset DD being defined as symptom onset at age less than 21 and then reanalyzed with early-onset being defined as symptom onset at age less than 25. To be consistent, GAD was also analyzed in this manner, as we were unaware of prior studies examining GAD by symptom onset. Analysis of variance (ANOVA) was calculated using Duncan, Tukey-HSD, and Scheffe post-hoc tests for threeway comparisons. The homogeneity of variances was analyzed with Cochran’s C and Bartlett’s Box F. Family history data were analyzed with chi-square. RESULTS
There were no significant differences between the control group and any of the psychiatric groups in regard to age, sex, and marital status. There were a greater number of women with MDD (N = 16) than PD (N = 8) (xl = 5.10; df= 1; P < .02). The mean age for each group was in the late 30s. Pure Vems Mixed disorders (DD v DD+; GAD v GAD+)
LX3 Versus GAD
The lifetime psychiatric diagnoses for DD+ and GAD+ patients are listed in Table 1. The most common lifetime diagnosis in the DD+ group was MDD and the most common lifetime psychiatric diagnosis in the GAD+ group was panic attacks. The comparisons of the mixed disorders with the pure disorders showed few significant differences. There were no significant differences in Table 1. Lifetime Psychiatric Diagnoses Complicating DO+ and GAD+
Infrequent panic attacks
DD+ (N = 16)
GAD+ (N = 25)
% N
% N
-
Major depression
94% (15)
Eating disorder
12% (2)
Generalized anxiety disorder
12% (2)
72% (18) 56% (14) -
Post traumatic stress disorder Hypomania
6% (I) -
any of the following variables: age, sex, marital status, age at first symptoms and treatment, duration of illness, history of suicide attempts, alcohol and drug use history, and revised total HAM-A and HAM-D scores. Most of the differences were inherent in how these groups were defined. DD+ patients were more likely to have had prior treatment (psychotherapy or medication) (F = 11.2; df = 1,28; P < .002), more frequently had a history of other psychiatric illnesses (F = 196; df= 1,28; P < .OOl), and had lower ratings for diurnal variation than the pure DD group (F = 4.07; df = 1,28; P < .05). GAD+ patients were more likely to be treated with medication at index presentation (F = 5.66; df = 1, 38; P < .02) and more frequently had a history of major depression (F = 18.1; df = 1,38; P < .OOOl) or other psychiatric illnesses (F = 45.1; df = 1,38; P < .OOl) than the GAD group. In contrast to previous studies, there were no StatisticalIy signi~cant differences between pure GAD and GAD+ on any of the items on the revised HAM-A or HAM-D scales, including items commonly associated with symptoms of panic attacks such as respiratory, cardiovascuIar. and autonomic symptoms.
4% i?)
Pure DD and pure GAD groups had distinct symptom profiles as determined by total HAM-A and HAM-D scores (see Table 2). On the revised HAM-A scale, 47% of the items were significantly different, with GAD patients having more severe sensory, autonomic, gastrointestinal, muscular, and respiratory symptoms; and more severe somatic and psychic anxiety, anxious behavior, anxious mood, and agitation than DD patients. On the revised HAM-D scale, 40% of the items were significantly different, with DD patients having more severe depressed mood, guilt, retardation, and greater impairment in work and activities than the GAD patients. There was also a trend for more severe diurnal variation in the pure DD group than in the pure GAD group. Pure GAD and GAD+ patients were combined and then compared with a combined group of pure DD and DD+ patients. This was done to see if the distinctions between pure DD
240
SHORES ET AL
Table 2. Comparison of Pure GAD With Pure DD Pure GAD
Pure DD
(N = 15)
(N = 14)
Age at first symptoms (yr)
20.9
20.8
F = ,001; df = 1,28; P = .97
Age at first treatment (yr)
32.1
31.4
F=.O3:df=1,28;P=.86
Any prior treatment N(%)
3 (20%)
3 (21%)
Total revised Ham-A
26.3
14.4
F = 21.90; df = f,28; P < .Ol
Total revised Ham-D
6.8
11.0
F = 11.64; df=
Anxious mood
2.6
1.5
F = 8.13; df = 1,28; P < .Ol
Muscular symptoms
2.0
1.0
F = 5.43; df = 1,28; P < .03
Sensory symptoms
1.7
0.3
F = 14.24; df = 1.28; P < .OOl
Respiratory symptoms
1.3
0.3
F = 5.93; df = 1.28; P < -02
GI symptoms
1.9
0.6
F = 10.11; df = 1.28; P < .Ol
Autonomic symptoms
1.7
0.7
F = 7.87; c/f = 1.28; P < .Ol
Anxious behavior
2.3
1.1
F = 30.34: df = 1,28; P < .Ol
Agitation
1.4
0.5
F = 12.09; df = 1,28; P < .03
Psychic anxiety
2.2
1.6
F = 5.58; df = .28; P < .03
Somatic anxiety
2.0
0.9
F = 14.4: df = 1,28; P < .Ol
Statistics
F = ,008; df = 1,28; P = .93 1.28: P < .Ol
Specific revised Ham-A items
Specific revised Ham-D items Depressed mood
1.1
2.1
F = 8.99; df = 1‘28: P < .Ol
Guilt
0.7
1.6
F = 14.85; df = 1,28; P < ,001
impairment in work/activities
0.8
1.8
F = 15.48; df = 1,28: P < .OOl
Diurnal variation
0.3
0.7
F = 3.46; df = 1,28; P = .07
Retardation
0.4
0.9
F=4,39;&=1,28;P
