Alzheimer’s & Dementia 3 (2007) 200 –203
The Frontal Systems Behavior Scale discriminates frontotemporal dementia from Alzheimer’s disease Paul Malloy*, Geoffrey Tremont, Janet Grace, Laura Frakey Brown Medical School, Providence, RI, USA
Abstract
Frontotemporal dementia (FTD) is characterized by pronounced changes in affect, self-regulation, and social conduct. These behaviors can predate significant cognitive changes and can be the most disabling aspect of FTD, yet there are few scales designed to assess such changes. The Frontal Systems Behavior Scale (FrSBe) is a 46-item behavior rating scale that is intended to measure behavior associated with damage to the frontal systems of the brain, with subscales measuring Apathy, Disinhibition, and Executive Dysfunction. Thirty-four FTD patients and 34 matched patients with dementia of the Alzheimer’s type (DAT) were compared on the FrSBe in the present study. Both groups displayed increases in apathy and dysexecutive behaviors after the onset of dementia, but the FTD group exhibited significantly greater change in disinhibition. A discriminant analysis with just two scores from the FrSBe and a memory test correctly classified 81% of the patients into diagnostic group. © 2007 The Alzheimer’s Association. All rights reserved.
Keywords:
Frontal lobes; Dementia; Frontotemporal dementia; Alzheimer’s disease; Neuropsychological assessment
1. Introduction Frontotemporal dementia (FTD) is a subtype of frontotemporal lobar degeneration (FTLD) characterized by pronounced changes in affect, self-regulation, and social conduct that can predate significant cognitive changes and typically has an earlier age of onset than Alzheimer’s disease (AD). Cognitive deficits caused by damage or disease affecting frontal lobe systems, including FTD, are well-known to clinical neuroscientists. There are a host of standardized measures of frontal cognitive changes, such as the Wisconsin Card Sorting Test [1] and the Stroop Test [2]. Noncognitive changes in behavior, however, are often more disruptive to adaptive functioning [3]. These changes commonly include apathy, disinhibition, and dysexecutive behaviors, and these three problems have been attributed to disruption of the medial, orbital, and dorsolateral frontal subsystems, respectively [4]. Despite the recognized importance of disruptive behaviors consequent to frontal lesions, only a few psychometric measures have been developed to measure these problems. These include the Behavior Rating Inventory of Executive Functions
[5], the Dysexecutive Questionnaire [6], the Iowa Rating Scales of Personality Change [7], the Frontal Behavior Inventory (FBI) [8], and the Frontal Systems Behavior Scale (FrSBe) [9]. A recently published review of these scales [10] documented that only the FBI has been studied in FTD. The FrSBe [9] is a 46-item behavior rating scale that is intended to measure behavior associated with damage to the frontal systems of the brain. The FrSBe has been demonstrated to have good reliability [9] and is sensitive to behavior change in focal frontal lesions when compared with patients with posterior lesions [11]. The FrSBe has also been shown to be sensitive to behavior changes in AD, Huntington’s disease, Parkinson’s disease, vascular dementia, and mild cognitive impairment [12–14]. We predicted that the FrSBe would detect the behavior changes that typify FTD, particularly the disinhibited behaviors commonly displayed by patients with this disorder. 2. Methods 2.1. Participants
*Corresponding author. Tel.: 401-301-0427; Fax: 401-455-6405. E-mail address:
[email protected]
Participants were selected from the patient populations of three memory clinics affiliated with Brown Medical
1552-5260/07/$ – see front matter © 2007 The Alzheimer’s Association. All rights reserved. doi:10.1016/j.jalz.2007.04.374
P. Malloy et al. / Alzheimer’s & Dementia 3 (2007) 200 –203
School. All patients had undergone comprehensive evaluations including neurologic examination, neuropsychological testing, and neuroimaging. Diagnosis was based on results of consensus conference of the memory clinic staffs, with the criteria of Neary et al [15,16] for diagnosis of FTD and with National Institute of Neurological and Communicative Disorders and Stroke (NINCDS) criteria for diagnosis of probable dementia of the Alzheimer’s type (DAT). Patients who met the Neary criteria for semantic dementia or progressive nonfluent aphasia were excluded because of the small number of these patients available in our sample and because these subtypes do not typically present with early significant behavioral disinhibition. Thirty-four FTD patients were identified, and 34 DAT patients were matched for gender and education from consecutive referrals to the Butler Hospital Memory Clinic. The DAT patients could not be matched for age, because earlier age of onset is a defining feature of FTD. However, FrSBe scores were transformed to T-scores, normatively corrected for age, education, and gender. 2.2. Measures The FrSBe consists of two rating forms, a Self Rating form to be completed by the patient and a Family Rating form to be completed by an informant who has regular contact with the patient. Each FrSBe form yields a total score and scores for subscales measuring Apathy, Disinhibition, and Executive Dysfunction. Scores are obtained on each scale for premorbid behavior and current behavior, allowing measurement of change consequent to neurologic disease or insult. All participants were administered the FrSBe Family form [9] as part of their routine evaluation in the memory clinics. The Family form is recommended for use in dementia patients, in whom anosognosia, abulia, and amnesia are likely to distort self-ratings. Informants (usually spouses or other family members) completed the FrSBe, and T-scores on each subscale were calculated with normative tables from the FrSBe manual [9]. The normative sample includes 436 healthy adults, with normative tables stratified for gender, age, and education. T-scores have a mean of 50 and standard deviation of 10, with higher scores on the FrSBe indicating greater pathology.
201
3. Results 3.1. Demographics Matching resulted in no differences between the groups in education (FTD mean, 13.88 years; DAT mean, 12.65 years; t(66) ⫽ 1.69, not significant) or gender (21 men and 13 women in each group). As would be expected, the mean age of the FTD group was significantly younger (67.6 years) than the mean age of the DAT group (80.9 years), t(66) ⫽ –7.39, P ⬍ .001). As noted above, FrSBe scores were age-corrected. 3.2. Comparison of diagnostic groups on FrSBe Table 1 presents the average scores for the two diagnostic groups on each FrSBe subscale. The FTD group had slightly higher T-scores for each of the behavioral scales at baseline than the AD group; however, it will be noted that the baseline, pre-dementia ratings of both groups were in the normal range for all subscales. The scores of both groups became elevated after onset of dementia, falling more than two standard deviations above normal (T ⬎ 70). The exception was Disinhibition, which remained in the normal range for the DAT group but not for the FTD group. Mixed model 2 ⫻ 2 (diagnostic group by time) analyses of variance (ANOVAs) were conducted for FrSBe total score and Apathy, Disinhibition, and Dysexecutive subscale scores. The ANOVA demonstrated a significant Group by Time (pre-post onset of dementia) interaction for FrSBe total score, F1,66 ⫽ 4.55, P ⬍ .037, indicating that the FTD group had greater abnormal behavior change overall. Indeed, the mean total FrSBe score for the FTD group was more than 90, greater than 4 standard deviations above the mean of the normative group. Comparison of scores on the individual subscales of the FrSBe revealed a significant Group by Time interaction for Disinhibition, F1,66 ⫽ 8.59, P ⬍ .005, but not for the Apathy or Dysexecutive subscales. These results indicate that the FTD group became significantly more disinhibited over time, but the AD group did not. 3.3. Discriminant analysis We next conducted a discriminant analysis to determine the accuracy with which individual diagnosis could be pre-
Table 1 Mean T-scores of FTD and DAT groups on the FrSBe (high scores are more abnormal) FTD
FrSBe score Apathy Disinhibition Dysexecutive Total
DAT
Pre-dementia
Post-dementia
Change
Mean
SD
Mean
SD
Mean
55.6 52.9 56.5 56.7
12.2 14.0 14.9 14.7
88.3 78.4 88.4 92.2
16.8 22.4 16.6 18.5
32.7 25.5 31.9 35.4
Pre-dementia
Post-dementia
Change
SD
Mean
SD
Mean
SD
Mean
SD
19.8 21.0 15.0 17.7
49.5 43.5 49.8 47.3
10.4 12.0 12.4 11.6
73.3 56.4 77.3 73.8
20.8 19.9 15.9 18.6
23.8 12.9 27.5 26.6
18.3 13.6 14.7 16.4
202
P. Malloy et al. / Alzheimer’s & Dementia 3 (2007) 200 –203
dicted. Because of the sample size, we took the relatively conservative approach of using just two predictors, one cognitive measure and one behavioral index that would highlight the early hallmarks of each diagnosis. This approach was also selected to illustrate that behavioral measures can enhance diagnostic accuracy over cognitive test scores. We chose delayed free recall on a verbal memory test (the Hopkins Verbal Learning Test-Revised [HVLT-R]) as the first predictor, because it should be sensitive to early memory impairment in AD. We chose Disinhibition on the FrSBe as the second predictor, because it should be sensitive to behavioral derangement in FTD. Four participants were excluded from this analysis because of missing HVLT-R scores (3 FTDB and 1 DAT patients). A stepwise discriminant analysis demonstrated a significant ability of the two variables to predict diagnosis, Wilks ⫽ 0.626, 2(2) ⫽ 28.62, P ⬍ .001. HVLT free recall entered in the first step, and FrSBe Disinhibition entered in the second step (significance of F to enter, P ⬍ .001 in each step). Standardized canonical discriminant function coefficients were 0.73 for HVLT and 0.69 for FrSBe, indicating comparable predictive values. This model successfully classified 81.3% of total sample, with 90.9% of the FTD patients and 71% of the AD patients being correctly classified. We cross-validated these results with a jackknife procedure, which indicated the findings were reliable with no change in classification accuracy. The “leave one out” or jackknife cross-validation procedure helps ensure that outliers do not unduly influence results. 4. Discussion When the behavioral reports of our participants were compared before and after dementia onset, patients with FTD displayed significant increases in noncognitive, problematic behaviors that could be measured efficiently with the Family form of the FrSBe. The FrSBe was also found to be sensitive to behavioral changes occurring in DAT patients, but the pattern of these changes was distinctive from the pattern in FTD patients. Apathy is the most common behavioral problem in DAT [17], and dysexecutive problems can be found in both DAT and FTD, particularly as the severity of the disorders increases [18]. Both of our groups displayed increases in apathy and dysexecutive behaviors after the onset of dementia, but the FTD group showed more dramatic change in disinhibition. The FrSBe Disinhibition scale contains items assessing behaviors commonly observed in FTD (and less often in DAT), such as swearing, silly or childlike behavior, impulsivity, inappropriate sexual behavior, and labile emotionality. Similar findings have been reported when the FrSBe has been used with other disorders involving disinhibition as a prominent feature. Spinella et al [19,20] have conducted a series of studies showing that high FrSBe scores were associated with polysubstance abuse, binge eating, and difficulty delaying gratification in simulated gambling tasks.
Previous research has shown that FrSBe behavior profiles differ in cortical (AD) and subcortical (Huntington’s disease, Parkinson’s disease) dementias [12]. Psychotic AD patients have been found to score higher on the Disinhibition scale of the FrSBe than nonpsychotic AD patients [21]. In addition to its utility in differential diagnosis, the FrSBe correlates with real-world outcomes in dementia patients. FrSBe Apathy scores predict performance on activities of daily living [22], whereas higher FrSBe Disinhibition and Executive scores are associated with caregiver burden [23,24]. The present study extends these findings to FTD and bolsters the notion that quantified measurement of behavioral change provides useful information in dementia diagnosis. The FrSBe might be a useful addition to the armamentarium of clinicians by allowing objective evaluation of behavior change over time. The FrSBe is not meant to replace a comprehensive, multidisciplinary evaluation. Rather, it can provide a standardized method for evaluating behavior change, complementing cognitive assessment, neuroimaging, and the neurologic examination. It is low in cost and in burden on the patient, requiring only about 15 minutes of caregiver time to complete. Only one frontal behavior scale has been previously validated in FTD, but there were limitations to this prior research. In the first of two studies, Kertesz et al [8] examined 12 patients with FTD, 16 patients with AD, and 11 patients with “depressive dementia.” FTD patient scores on the FBI were significantly higher compared with control groups. In a second study, Kertesz et al [25] administered the FBI to caregivers of 26 patients with FTD and 82 patients with other dementing disorders (vascular dementia, AD, primary progressive aphasia, or depressive disorder). Discriminant analysis with “leave one out” or jackknife cross-validation was used to evaluate the ability of the FBI to discriminate FTD from other diagnoses. The hit rates for the initial validation were not reported. The “leave one out” results correctly classified 92.6% of FTD patients versus all other patients in the study. However, it appears that all 24 items from the FBI were available for entry in the prediction equation. The “leave one out” cross-validation procedure helps ensure that outliers do not unduly influence results, but it does not address the issue of high predictor to observation ratio. In comparison, the present study used only two predictors with 68 subjects. We also showed the incremental utility of the FrSBe by demonstrating that it added to the prediction of a memory test alone. Some limitations of the present study should be noted. Examination of the premorbid ratings of our AD and FTD groups showed that the FTD groups had higher T-scores at baseline. However, both groups fell within the normal range before dementia onset. There are two plausible explanations for why the FTD patients were rated as having more problem behaviors premorbidly. One possibility is that the FTD patients might have actually exhibited early changes in
P. Malloy et al. / Alzheimer’s & Dementia 3 (2007) 200 –203
behavior that predated the development of significant cognitive change and dementia. Another possibility is that caregiver ratings might have been subject to a “halo effect,” whereby current behavioral disturbances influenced retrospective ratings of premorbid behavior. In either case, it is important to note that covariance analyses confirmed premorbid differences in T-scores did not account for postdementia differences on any FrSBe scale. Disinhibition is more likely to be seen in FTD than in the other FTLD subtypes in which language deficits are more often seen early in the disease process. We compared the scores of our FTD sample with a small sample (n ⫽ 7) of patients with semantic dementia or primary progressive aphasia who were seen in our memory clinic. Results indicated that although both groups had similar T-scores premorbidly, the FTD group showed more significant behavioral pathology on all the FrSBe scales after dementia onset. FTLD is a relatively rare disorder, and our sample was therefore small, despite recruitment from three regional dementia centers. All consecutive cases of FTD were included in the present sample, and AD patients were matched to the FTD sample. Consensus diagnosis by expert clinicians with research diagnostic criteria was chosen as the gold standard for accurate diagnosis for this study. The only more accurate method of diagnosis of which we are aware is autopsy. These data were unfortunately not available in this cohort. We hope to investigate the FrSBe further with a larger sample of FTLD patients and to correlate scores with pathology and/or neuroimaging changes in future studies. The focus of the present article was the utility of the FrSBe in distinguishing FTD from AD, the most common etiology of dementia. However, the diagnostic utility of any instrument is dependent in part on the base rates of disorders in the population and the homogeneity of the diagnoses. Future research should therefore evaluate the FrSBe (1) in larger, mixed samples including patients with other commonly occurring etiologies, such as cerebrovascular disease and Lewy body dementia; (2) in patients with AD with and without small vessel pathology; and (3) in other subtypes of FTLD (eg, semantic dementia, progressive aphasia). Future research could also investigate changes in patterns and severity of behavioral disturbance in FTD as the disease progresses. For example, does FTD continue to diverge behaviorally from AD during the course of the illness, and can these changes be predicted in individual patients by using the FrSBe? References [1] Heaton RK, Chelune GJ, Talley JL, Kay GG, Curtiss G. Wisconsin Card Sorting Test manual: revised and expanded. Odessa, FL: Psychological Assessment Resources, 1993. [2] Stroop JR. Studies of interference in serial verbal reactions. J Exp Psychol 1935;18:643– 62.
203
[3] Eslinger PJ, Damasio AR. Severe disturbance of higher cognition after bilateral frontal lobe ablation: patient EVR. Neurology 1985;35: 1731– 41. [4] Cummings JL. Frontal-subcortical circuits and human behavior. Arch Neurol 1993;50:873– 80. [5] Gioia GA, Isquith PK, Guy SC, Kenworthy L. Behavior rating inventory of executive function. Neuropsychol Dev Cogn Sect C Child Neuropsychol 2000;6:235– 8. [6] Wilson BA, Alderman N, Burgess PW, Emslie H, Evans J. Behavioural Assessment of the Dysexecutive Syndrome (BADS). Bury St Edmunds, England: Thames Valley Test Company, 1996. [7] Barrash J, Tranel D, Anderson SW. Acquired personality disturbances associated with bilateral damage to the ventromedial prefrontal region. Dev Neuropsychol 2000;18:355– 81. [8] Kertesz A, Davidson W, Fox H. Frontal behavioral inventory: diagnostic criteria for frontal lobe dementia. Can J Neurol Sci 1997;24:29 –36. [9] Grace J, Malloy PF. Frontal Systems Behavior Scale (FrSBe): professional manual. Lutz, FL: Psychological Assessment Resources, 2001. [10] Malloy P, Grace J. A review of rating scales for measuring behavior change due to frontal systems damage. Cogn Behav Neurol 2005;18: 18 –27. [11] Grace J, Stout JC, Malloy PF. Assessing frontal lobe behavioral syndromes with the Frontal Lobe Personality Scale. Assessment 1999;6: 269 – 84. [12] Paulsen JS, Stout JC, DelaPena J, Romero R, Tawfik-Reedy Z, Swenson MR, et al. Frontal behavioral syndromes in cortical and subcortical dementia. Assessment 1996;3:327–37. [13] Zawacki TM, Grace J, Paul R, Moser DJ, Ott BR, Gordon N, et al. Behavioral problems as predictors of functional abilities in vascular dementia patients. J Neuropsychiatry Clin Neurosci 2002;14:1–7. [14] Ready RE, Ott BR, Grace J, Cahn-Weiner DA. Apathy and executive dysfunction in mild cognitive impairment and Alzheimer disease. Am J Geriatr Psychiatry 2003;11:222– 8. [15] Neary D, Snowden JS, Gustafson L, Passant U, Stuss D, Black S, et al. Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology 1998;51:1546 –54. [16] Neary D. Overview of frontotemporal dementias and the consensus applied. Dement Geriatr Cogn Disord 1999;10(Suppl 1):6 –9. [17] Boyle PA, Malloy PF. Treating apathy in Alzheimer’s disease. Dement Geriatr Cogn Disord 2004;17:91–9. [18] Bozeat S, Gregory CA, Ralph MA, Hodges JR. Which neuropsychiatric and behavioural features distinguish frontal and temporal variants of frontotemporal dementia from Alzheimer’s disease? J Neurol Neurosurg Psychiatry 2000;69:178 – 86. [19] Spinella M, Lyke J. Executive personality traits and eating behavior. Int J Neurosci 2004;114:83–93. [20] Spinella M, Lester D, Yang B. Gambling and delaying rewards as a function of frontal system dysfunction: a study in neuroeconomics. Percept Mot Skills 2004;99(Pt 1):993– 4. [21] Paulsen JS, Ready RE, Stout JC, Salmon DP, Thal LJ, Grant I, et al. Neurobehaviors and psychotic symptoms in Alzheimer’s disease. J Int Neuropsychol Soc 2000;6:815–20. [22] Norton LE, Malloy PF, Salloway S. The impact of behavioral symptoms on activities of daily living in patients with dementia. Am J Geriatr Psychiatry 2001;9:41– 8. [23] Rymer S, Salloway S, Norton L, Malloy P, Correia S, Monast D. Impaired awareness, behavior disturbance, and caregiver burden in Alzheimer disease. Alzheimer Dis Assoc Disord 2002;16:248 –53. [24] Davis J, Tremont G. Impact of frontal systems behavioral functioning in dementia on caregiver burden. J Neuropsychiatry Clin Neurosci 2007;19:43–9. [25] Kertesz A, Nadkarni N, Davidson W, Thomas AW. The Frontal Behavioral Inventory in the differential diagnosis of frontotemporal dementia. J Int Neuropsychol Soc 2000;6:460 – 8.
