Ucsur.pitt.edu

BRIEF REPORT
Generalized Anxiety
late-life GAD as a chronic disorder distinct from MDD.
(Am J Geriatr Psychiatry 2005; 13:77–80) Disorder in Late Life
Lifetime Course and
Generalized anxiety disorder (GAD) in elderly Comorbidity With Major
persons is highly prevalent1 and positively as- sociated with functional disability and healthcare Depressive Disorder
utilization.1 Late-life GAD is often concurrent withMDD;2 however, little is known about its naturalcourse. Controversy exists regarding whether gener- Eric J. Lenze, M.D.
alized anxiety symptoms in depressed patients rep- Benoit H. Mulsant, M.D.
resent GAD as a separate diagnosis.3 In depressed Jan Mohlman, Ph.D.
elderly patients, GAD is associated with greater se- M. Katherine Shear, M.D.
verity, including suicidality and poorer functioning.2 Mary Amanda Dew, Ph.D.
However, we are not aware of any study examiningthe lifetime course of late-life GAD. The present study Richard Schulz, Ph.D.
had two aims: 1) to examine the lifetime course and Mark D. Miller, M.D.
comorbidity of GAD in elderly persons; and 2) to de- Barbara Tracey, M.S.N.
termine whether the disorder could be differentiated, Charles F. Reynolds III, M.D.
in lifetime chronology, from MDD. We hypothesizedthat most subjects with GAD and MDD would havedifferent times of onset and offset of the disorders.
Objective: Generalized anxiety disorder (GAD) in el-
The overarching aim was to provide a characteriza- derly persons is highly prevalent, but little is known tion of late-life GAD that could inform future studies.
about its course, age at onset, and relationship to co- morbid major depressive disorder (MDD). The au- thors assessed the course and comorbidity of late-life GAD and MDD. Methods: Authors assessed elderly
subjects in anxiety or depression intervention studies who had a lifetime history of GAD, with current MDD To examine the characteristics of GAD in elderly per- (Nס57) or without (Nס46). Subjects’ lifetime course sons seeking mental health treatment (regardless of of illness was charted retrospectively. Results: The 103
whether subjects had comorbid MDD or other diag- subjects had a mean age of 74.1 years, and a mean noses), we evaluated data from all recent studies that age at onset of GAD of 48.8 years; 46% were late-onset. recruited subjects with GAD. The studies ran concur- GAD episodes were chronic, and 36% were longer than rently at the University of Pittsburgh Intervention Re- 10 years. Of the comorbid GAD–MDD patients, most search Center for the study of Late-life Mood Disor- had different times of onset and/or offset of the dis- ders (IRC/LLMD); two studies were of anxiety orders; typically, GAD preceded MDD. Conclusions:
disorders, and one was of MDD. Subjects were re- Elderly subjects with GAD tended to have chronic cruited via radio and newspaper advertisements (tar- symptoms lasting years-to-decades, without interrup- geting anxiety/depression), referrals, community tion, and many have late onset. Elderly persons with presentations, website, newsletter, and word-of- lifetime GAD and MDD tend to have different onset mouth. The first anxiety study was an open-label and offset of the two disorders. Findings characterize study of fluvoxamine in the treatment of anxiety dis- Received August 25, 2003; revised March 26, 2004, June 7, 2004; accepted June 9, 2004. From the Intervention Research Center in Late-Life MoodDisorders, Department of Psychiatry, University of Pittsburgh School of Medicine, Pittsburgh, PA (EJL,BHM,MKS,MAD,RS,MDM,BT,CFR), the VAPittsburgh (BHM), and the Department of Psychology, Syracuse University, Syracuse, NY (JM). Send correspondence and reprint requests to Dr.
Lenze, Western Psychiatric Institute and Clinic, Room E835, 3811 O’Hara Street, Pittsburgh, PA 15213. e-mail: [email protected] ᭧ 2005 American Association for Geriatric Psychiatry Am J Geriatr Psychiatry 13:1, January 2005 Lifetime GAD: Comorbidity With Depression orders.4 The second study, for which recruitment GAD by episode length and age at onset, and an ex- started soon after the end of the first, was a placebo- amination of the chronological order of lifetime ill- controlled evaluation of citalopram.5 These studies ness with respect to MDD in those with both diag- required subjects to be age 60-and-older, with an noses. We also compared early- and late-onset anxiety disorder, without current MDD, as diagnosed subjects in terms of baseline clinical and sociodemo- by the Structured Clinical Interview for DSM-IV Axis graphic data by use of chi-square and t-tests.
I disorders (SCID)6 and a Hamilton Rating Scale forAnxiety (Ham-A) score Ն17. From these two studies,we analyzed only those in a current GAD episode, as In the third study, Maintenance Therapies in Late- A total of 103 subjects with GAD were examined in life Depression (MTLD2),7 subjects age 69-and-older this analysis, 46 from the GAD-only group and 57 are required to be in a major depressive episode, as from the MDD-with-GAD group. Table 1 shows sub- diagnosed by the SCID, with a baseline 17-item Ham- jects’ baseline characteristics. The most common life- ilton Rating Scale for Depression (Ham-D) score Ն15.
time comorbidities were panic disorder (Nס24), dys- From this study, we analyzed only those in a current thymic disorder (Nס17), specific phobia (Nס14), GAD episode, creating an “MDD with GAD” group.
social phobia (Nס10), agoraphobia without panic All three studies excluded subjects with history of bi- disorder (Nס7), and alcohol abuse or dependence polar disorder, psychotic symptoms, or active alco- We evaluated age at first onset of GAD; this was SCID data were confirmed in consensus confer- obtainable for all but one subject (who had early- ences with at least two geriatric psychiatrists. Using adulthood onset). Mean (standard deviation [SD]) lifetime SCID data, we were able to examine duration age at GAD onset was 48.8 (30.0) years (median: 58; of GAD episodes, as well as age at onset of disorders.
range: 5–87). Thus, although some subjects had GAD We modified the SCID by removing the criterion re- since childhood or early adulthood, 47 (46%) had on- quiring that GAD symptoms be present outside of set later in life. We found no differences between the episodes of depression. Thus, we were able to iden-tify GAD when present only in the presence of de- Baseline Characteristics of Subjects
pression, necessary to determine how often GAD had GAD Without MDD
MDD With GAD
onset/offset simultaneous with MDD. We defined early-onset GAD patients as those with first onset of symptoms meeting full criteria for GAD before age 60, and late-onset patients as those with first onset at or after age 60. Additional baseline assessment data included sociodemographics and scores on the Mini- Mental State Exam and Ham-D. Ham-A scores were available for anxiety-study subjects only. Ham-A and Ham-D ratings were done by master’s-level clinicians supervised by investigators, with excellent interrater reliability (intraclass correlation coefficients of 0.93 All studies used the MMSE. We assessed functional disability with an Activities of Daily Living (ADL) scale; in this 15-item scale, a score of 30 indicates no Values are mean (standard deviation), unless otherwise disability, and lower scores indicate increasing dis- aFor those 14 subjects with history of MDD only.
ability. We measured physical illness with the Cu- Ham-D: Hamilton Rating Scale for Depression; Ham-A: Hamilton mulative Illness Rating Scale for Geriatrics (CIRS–G);8 Rating Scale for Anxiety; MMSE: Mini-Mental State Exam; CIRS–G:Cumulative Illness Rating Scale, Geriatrics; ADL: Activities of Daily higher scores indicate more severe medical burden.
Living; MDD: major depressive disorder.
Our analytic strategy was a characterization of Am J Geriatr Psychiatry 13:1, January 2005 early-onset and late-onset GAD groups in age (72.2 be delineated from MDD. Our data did not support [SD: 7.3] early versus 74.7 [6.0] late; t our hypothesis that disability, chronic medical illness, pס0.07), MMSE score (28.4 [2.0] early versus 27.9 and cognitive impairment were more common in late-onset than early-onset cases. Our negative find- ings are consistent with other work;9 however, our cardiovascular illness (37/52 early versus 36/44 late; negative findings may have been due to the measures 0.2; pס0.67) or cerebrovascular illness (19/52 used; for example, the MMSE may not be sufficiently score (26.8 [4.4] early versus 26.4 [4.0] late; t In terms of comorbidity of GAD with MDD, our data did support our hypothesis that GAD and MDD We evaluated duration of the GAD episode; this have different onset and offset. Even though we sus- information was obtainable for 98 of the 103 subjects pended the DSM-IV hierarchy rule for GAD, allow- (data not available for one GAD-only subject and four ing its diagnosis to be made when it occurred only MDD-with-GAD subjects). Mean episode duration within MDD episodes, simultaneous onset and offset was 16.7 (27.7) years (median: 3.1; range: 0.5–75).
for these disorders was the exception rather than the GAD episode duration was not different between rule; furthermore, GAD tended to be a single chronic those with comorbid MDD (16.5 [22.5] years) and episode, whereas MDD is typically recurrent. We found that the most common lifetime pattern of co- Only 18 of the 103 subjects had a “recurrent” pat- tern (i.e., episodes of GAD before the current one) and In terms of duration of GAD, we found a large only one subject had residual GAD (symptoms in re- number of late-onset GAD subjects, together with a mission or partial remission). Thus, 82% of subjects smaller number of early-onset subjects, who had ep- (84/103) had GAD continuously from first onset until isodes lasting several decades. We found that GAD they presented to our clinic for treatment of anxiety tends to persist, with spontaneous remission (without or depression. By contrast, 54% of subjects (31/57) treatment) unlikely even if comorbid MDD remits.
with comorbid MDD had a recurrent pattern of MDD This finding enhances the rationale for long-term 22.1; p Ͻ0.001 for comparison of recurrence treatment of GAD; however, it should be noted that rate between GAD and MDD); thus, in this sample, our data may not be reflective of all elderly patients MDD was more of a recurrent illness than GAD.
with GAD; given the treatment setting, our study se- Finally, we evaluated the temporal pattern of GAD lected a sample likely to have current, unremitted and MDD in those 71 individuals who had a history GAD. It would be unlikely for individuals with re- of both disorders (14 subjects from anxiety studies mission of symptoms to present to research centers; and 57 from the MDD-with-GAD group). We saw the rather, these findings would be expected to be reflec- following patterns: 20 subjects met GAD criteria only tive of treatment-seeking elderly persons with cur- during MDD episodes; 31 had onset of GAD before the onset of MDD (15 early-onset GAD, 16 late-onset Our subjects were probably more likely than com- GAD), 16 had onset of GAD after onset of MDD (4 munity-dwelling elderly persons, in general, to have early-onset GAD, 12 late-onset GAD), and four had MDD comorbidity because we advertised for both simultaneous onset of GAD and MDD (with GAD depression and anxiety and because comorbidity persisting after resolution of depression).
tends to be higher in treatment-seeking populationsthan in the general community. We found that manysubjects had panic disorder or social phobia, which isnot surprising, because these are also the most com- DISCUSSION
mon anxiety disorders comorbid to GAD in youngadults. We detected very little alcohol and substance To our knowledge, this is the first report to provide comorbidity in our sample, possibly because current a lifetime perspective on the course and comorbidity alcohol or substance abuse was an exclusion criterion of GAD. In this evaluation of treatment-seeking el- for treatment in our studies. Another potential limi- derly patients, GAD was typically chronic and could tation is recall bias; subjects may not be able to recall Am J Geriatr Psychiatry 13:1, January 2005 Lifetime GAD: Comorbidity With Depression remissions—perhaps because of aging or psychiatric disorders. Together with previous studies showing illness, and perhaps because of the often-insidious that late-life GAD is highly prevalent and associated onset and waxing and waning course of GAD—in- with burden such as increased functional disability stead, describing GAD as lifelong. In this study, and healthcare utilization, our study depicts late-life though, these same subjects were able to recall the GAD as a chronic disorder that can be delineated episodic nature of MDD. Thus, it is likely that the from MDD. Because late-life GAD is poorly recog- greater chronicity of GAD, versus MDD, is real.
nized and undertreated, and there are few treatment In summary, GAD in treatment-seeking elderly studies, we hope that our findings will strengthen the populations seems to be a combination of cases per- argument that the treatment of this disorder needs sisting since adulthood (or even childhood) and late- onset cases. Cross-sectionally, patients with early-on-set and late-onset GAD look more alike than This work was supported by NIMH grants K23 different. Also, GAD and MDD usually follow inde- MH64196, P30 MH 52247, K01 MH01613, R37 pendent lifetime courses in elderly patients with both MH43832, R01 MH37869, and T32 MH19986. References
1. Beekman AT, Bremmer MA, van Balkom AJ, et al: Anxiety disor- sults from an 8-week, placebo-controlled trial. Am J Psychiatry ders in later life: a report from the Longitudinal Aging Study, Amsterdam. Int J Geriatr Psychiatry 1998; 13:717–726 6. First M, Spitzer RL, Gibbon M, et al: Structured Clinical Interview 2. deBeurs E, Beekman ATF, van Balkom AJLM, et al: Consequences for DSM-IV Axis I Disorders (SCID), Clinical Version: Administra- of anxiety in older persons: its effect on disability, well-being, tion Booklet. Washington, DC, American Psychiatric Press, 1996 and use of health services. Psychol Med 1999; 29:583–593 7. Szanto K, Mulsant BH, Houck P, et al: Occurrence and course of 3. Lenze EJ, Mulsant BH, Shear MK, et al: Comorbid anxiety disor- suicidality during short-term treatment of late-life depression.
ders in depressed elderly patients. Am J Psychiatry 2000; 8. Miller MD, Paradis CF, Houck PR, et al: Rating chronic medical 4. Wylie ME, Miller MD, Shear MK, et al: Fluvoxamine pharmaco- illness burden in geropsychiatric practice and research: applica- therapy of anxiety disorders in later life: preliminary open-trial tion of the Cumulative Illness Rating Scale. Psychiatry Res 1992; data and a review of the literature. J Geriatr Psychiatry Neurol 9. Beck JG, Stanley MA, Zebb BJ: Characteristics of generalized anxi- 5. Lenze EJ, Mulsant BH, Shear MK, et al: Efficacy and tolerability ety disorder in older adults: a descriptive study. Behavior Res of citalopram in the treatment of late-life anxiety disorders: re- Am J Geriatr Psychiatry 13:1, January 2005

Source: http://www.ucsur.pitt.edu/files/schulz/ajgplenze05.pdf