IMPORTANCE OF BEHAVIOR CHANGES IN FRONTAL DISORDERS
Cognitive deficits due to damage or disease affecting frontal lobe systems are well known to clinical neuroscientists. There are a host of standardized measures of frontal cognitive changes, such as the Wisconsin Card Sorting Test and the Stroop Test. Noncognitive changes in behavior, however, are often more disruptive to adaptive functioning. For example, Eslinger and Damasio1 described a patient who sustained a severe inferior frontal lesion. The patient performed quite normally on cognitive testing but experienced severe changes in social and emotional behavior that rendered him disabled. Despite the recognized importance of disruptive behaviors consequent to frontal lesions, only a few psychometric measures of behavior change in frontal disorders have been developed.
VARIETIES OF BEHAVIOR CHANGE IN FRONTAL DAMAGE
Much of the scientific literature on frontal lobe damage prior to 1970 was characterized by a morass of confusing and contradictory descriptions of frontal behavior change. More recent research has brought some order to this chaos and delineated a number of frontal subsystems and resultant behavioral syndromes. Alexander et al2 described a multiple circuit structure of the frontal lobes. This model includes five parallel, functionally segregated circuits having specific target regions in the frontal lobes: the supplementary motor area, the frontal eye fields, the dorsolateral prefrontal cortex, the orbitofrontal cortex, and the medial frontal cortex. Frontal lobe lesions are now known to result in syndromes attributable to dysfunction in these areas: motor dysfunction (eg, release of primitive reflexes, incontinence), executive dysfunction (eg, disorganization, inflexibility, loss of hypothesis generation and testing, impaired working memory), reduction in drive and motivation, and disinhibition (impulsivity, disinhibition, poor social judgment). Of course, naturally occurring frontal lobe lesions can disrupt multiple circuits because of their close proximity to one another. This is particularly true in strategic, subcortical lesions (eg, thalamic stroke), in which relatively small lesions can produce behavior changes comparable to large cortical frontal lesions. Recent conceptualizations have raised questions about anatomically or functionally segregated circuits and have described an open, interconnected model of basal ganglia-thalamocortical circuitry.3
Building upon the neuroanatomical work just described, Cummings and colleagues4,5 proposed a model linking the three clinically observable frontal behavioral syndromes to the three frontostriatothalamic circuits. The dorsolateral prefrontal circuit has been associated with executive cognitive dysfunction; the lateral orbital prefrontal circuit has been associated with disorders of self-regulation, such as the syndrome of frontal disinhibition; and the anterior cingulate circuit has been associated with disorders of activation, spontaneous behavior, and motivation, resulting in syndromes such as apathy.
Some of the scales reviewed below have taken a theoretical approach by developing subscales to measure the functions of these frontal systems. Others have been more clinical in their orientation, providing inventories of signs and symptoms commonly seen in frontal patients.
The current review will concentrate on several scales that were designed to measure frontal behavior change, and have research supporting their validity and utility. These include the Behavior Rating Inventory of Executive Functions (BRIEF), the Dysexecutive Questionnaire (DEX), the Frontal Systems Behavior Scale (FrSBe), and the Iowa Rating Scales of Personality Change (IRSPC). Two other scales (the Frontal Behavioral Inventory [FBI] and the Neuropsychiatric Inventory [NPI]) were originally designed to measure behavioral or neuropsychiatric changes in dementia but may be useful in other disorders affecting frontal systems. We describe each of these scales and review the literature on reliability, validity, and available norms. Finally, other scales purporting to measure specific aspects of frontal functioning, but having less research support, will be mentioned. Relative strengths and weaknesses of the scales are reviewed in Discussion.
Behavior Rating Inventory of Executive Functions (BRIEF)
The BRIEF6 is an inventory designed to measure executive dysfunction in children aged 5-18. It consists of two rating forms (a parent and a teacher questionnaire) designed to assess executive functioning in the home and school environments. The BRIEF produces two indexes composed of several clinical subscales. The Behavioral Regulation Index is made up of the Inhibit, Shift, and Emotional Control subscales. The Metacognition Index is made up of the Initiate, Working Memory, Plan/Organize, Organization of Materials, and Monitor subscales. There is also a Global Executive Composite score incorporating all eight clinical scales, which the authors recommend be used only when there is little variability in subscale scores. Validity scales measure Negativity and Inconsistency of Responses.
The BRIEF manual reports high internal consistency (Cronbach α = 0.80-0.98) and test-retest reliability (r = 0.81 for parents and 0.87 for teachers). There are relatively low interrater reliabilities between teacher and parent ratings (mean r = 0.32), as might be expected owing to differences in behavior across settings.
To enhance content validity, items were initially generated from parent and teacher interviews and then subjected to scale assignment by a panel of experts. Concurrent validity was assessed through correlations with other rating scales for children. A series of factor analyses in normal and clinical samples supported the two-factor model of executive functions used in scoring the BRIEF. Principal factor analysis with oblique rotation using the data from a normative sample on the parent form of the BRIEF resulted in factor loadings ranging from 0.48 to 0.97 for the Metacognition subscales and 0.46 to 0.99 for the Behavioral Regulation subscales. With use of data from a clinical sample, the factor loadings ranged from 0.58 to 0.97 for the Metacognition subscales and from 0.68 to 0.93 for the Behavioral Regulation subscales. The Inhibit and Shift subscales appeared to display lower and less consistent loadings than other subscales. Similar results were found using the teacher form of the BRIEF. The factor analysis on the parent form accounted for 74% of the variance in normal subjects and 76% in the clinical sample. The factor analysis of the teacher form accounted for 83% of the variance in normal subjects and 79% in clinical subjects.
The heterogeneous clinical sample described in the manual included children with developmental disorders or acquired neurologic disorders including reading disorder, attention deficit-hyperactivity disorder (ADHD) subtypes, traumatic brain injury (TBI), Tourette disorder, mental retardation, localized brain lesions, and high-functioning autism. Twenty-one children with magnetic resonance imaging-confirmed focal frontal lesions and 11 with nonfrontal lesions were included. Inspection of Table 42 of the manual indicates that these two groups did not differ on any BRIEF subscale, although both lesion groups differed from normal control subjects.
The BRIEF Working Memory and Inhibit Scales differentiate among ADHD subtypes. Gioia and colleagues7 compared parent ratings on the BRIEF of children with inattentive (ADHD-I) and combined (ADHD-C) types of ADHD, autistic spectrum disorders (ASDs), moderate and severe TBI, reading disabilities (RDs), and controls. Profile analysis revealed significant differences in global elevations and in profiles between diagnostic groups. ASD, ADHD-I, and ADHD-C groups exhibited greater elevations across the BRIEF scales than did RD and severe TBI groups, who were in turn more elevated than moderate TBI and control groups. The ADHD-C group had particularly abnormal ratings of inhibitory deficits, whereas the ASD group was distinguishable by its deficits in flexibility.
Norms presented in the BRIEF manual are based on child ratings from 1419 parents and 720 teachers from rural, suburban, and urban areas, reflecting 1999 US census estimates for socioeconomic status, ethnicity, and gender distribution. Separate normative tables for both the parent and teacher forms provide T scores, percentiles, and 90% confidence intervals for four developmental age groups (5-18 years) by gender of the child.
Dysexecutive Questionnaire (DEX)
The DEX is part of the Behavioural Assessment of the Dysexecutive Syndrome (BADS). The BADS is a battery of six tests aimed at predicting everyday problems arising from the “dysexecutive syndrome.” The authors noted that traditional neuropsychological tests are short and highly structured and hence do not make sufficient demands on organizational and persistence problems. Yet it is these deficits that cause patients with executive deficits to have difficulty in everyday life. The six primary tests that make up the BADS are cognitive in nature (eg, navigating to locations on a map) and are used to calculate a profile score. The DEX is a 20-item questionnaire that supplements the primary BADS test and is not used in the profile calculation.
The DEX was designed to measure four areas of change: emotional or personality changes, motivational changes, behavioral changes, and cognitive changes. The 20 items purport to assess the following characteristics: abstract thinking problems, impulsivity, confabulation, planning problems, euphoria, temporal sequencing problems, lack of insight, apathy, disinhibition, disturbed impulse control, shallow affective responses, aggression, lack of concern, perseveration, restlessness, inability to inhibit responses, knowledge-response dissociation, distractibility, loss of decision-making ability, and unconcern for social rules. Each item is scored on a 5-point Likert scale ranging from “never” to “very often.” There are forms of the DEX to be given to the patient and to an informant who knows the patient well. This allows the calculation of a discrepancy score, thought to indicate a lack of insight or awareness in the patient.
Reliability was not reported in the BADS manual.
A factor analysis presented in the DEX manual that “accounted for more than 50% of the variance, suggested that the questionnaire measures change in at least three potentially dissociable areas, which together make up the DEX” (p. 18). These factors were termed Behavior (eight items loading), Cognition (five items), and Emotion (three items). Analyses revealed that all the BADS cognitive tests correlated with the Behavior factor, three correlated with the Cognition factor, and none with the Emotion factor. The Total profile score was correlated with all three factors. It should be noted that this factor analysis was based on a “modest” but unspecified number of patients. A confirmatory analysis was planned for the future.
The DEX has not been shown to discriminate frontal from nonfrontal patients. Indeed, the authors eschew this approach, indicating that “specification of a syndrome in terms of localization is unfortunate and potentially misleading” (p. 1). Although normal and patient groups were tested on the BADS, comparisons between the groups on DEX scores were not reported.
Bogod and colleagues8 compared measurement of awareness by the DEX “self vs. other” rating with the Self-Awareness of Deficits Interview (SADI), a semistructured interview measure. Evaluation of awareness by these measures was compared with tests of executive functioning and IQ. Results indicated significant, albeit marginal relationships between the two measures, and better correlation of the SADI with measures of frontal lobe functioning.
Evans et al9 administered the BADS and DEX to 31 schizophrenic patients, 35 brain-injured patients, and 26 healthy control subjects. The two patient groups were more impaired than control subjects on the BADS but did not differ from each other. The two patient groups scored significantly worse on the DEX than control subjects when caregiver ratings were compared. When patient ratings were compared, the brain-injured patients were significantly worse than the schizophrenic subjects but not the control subjects.
The manual describes a sample of 216 normal subjects stratified by estimated IQ and age and 78 patients with diverse diagnoses including closed head injury (59%), encephalitis (6.5%), dementia (13%), stroke (8.5%), and other acquired insult (13%). The authors did not indicate whether any of these patients had specific damage to frontal systems. The manual provides clinical cutoffs for the BADS profile score but not for the DEX. Percentile scores are provided for the DEX, presumably based on the normative sample described above. The authors “anticipate that it [the DEX] will supplement the information provided by performance on the battery, primarily through the provision of additional qualitative information” (p. 16).
However, information about the performance of normal subjects may be crucial to correct interpretation of results. Chan10 tested 93 normal participants on the DEX and a set of clinical tests of executive function. A factor analysis resulted in a five-factor solution: Inhibition (factor 1), Intentionality (factor 2), Knowing-Doing Dissociation (factor 3), In-Resistance (factor 4), and Social Regulation (factor 5). Correlation was established among the derived factors and tests of executive function. This study provided empirical evidence that a nonclinical sample may encounter dysexecutive behaviors in daily life.
Frontal Behavioral Inventory (FBI)
The FBI11 is designed to optimize diagnostic accuracy for frontal lobe dementia. It rates behavior change via caregiver interview by trained professionals using structured questions. Items were based on behaviors described in diagnosis consensus reports by the Lund and Manchester groups12 and Neary et al13 for frontotemporal dementia (FTD). Twenty-four items represent both negative and positive behaviors: apathy, aspontaneity, emotional indifference, inflexibility, concreteness, personal neglect, distractibility/disorganization, inattention, loss of insight, logopenia, verbal apraxia, alien hand, perseveration, disinhibition/irritability, jocularity, irresponsibility/poor judgment, social inappropriateness, impulsivity, euphoria/restlessness, aggression, hyperorality, hypersexuality, utilization behavior, and incontinence. Behaviors are rated on a 4-point scale of frequency ranging from “none” to “severe or most of the time.”
Interrater reliability (Cohen κ = 0.90) and item consistency (Cronbach α = 0.89) were both high in a study comparing patients with FTD with patients having other dementias or depression.14
There have been two validation studies with the FBI. In the first study, Kertesz et al11 studied 12 patients with FTD, 16 patients with Alzheimer disease (AD), and 11 patients with depressive dementia. FTD patient scores on the FBI were significantly higher compared with control groups. Item analysis showed loss of insight, indifference, distractibility, personal neglect, and apathy as the most frequent negative symptoms. Perseveration, disinhibition, inappropriateness, impulsivity, and irresponsibility were the most significant positive symptoms. A cutting score of 27 resulted in only one false positive in the depressive group and none among the AD group, indicating little overlap between patient groups, and good discrimination with the FBI.
Kertesz and colleagues14 administered the FBI to caregivers of 108 patients with FTD, vascular dementia (VaD), AD, primary progressive aphasia (PPA), and depressive disorder (DD). The mean scores of FTD patients were significantly above those of all other groups. Scores in VaD were also higher than in AD, PPA, and DD. FTD patients had highest scores on ratings of perseveration, indifference, inattention, inappropriateness, and loss of insight. Apathy, aspontaneity, inflexibility, disorganization, impulsivity, personal neglect, and poor judgment were also significantly higher in FTD. Discriminant function correctly classified 92.7% of FTD patients versus all other patients in the study. A total of 18.8% of VaD patients were misclassified as FTD. Indifference, alien hand, and inappropriateness contributed to the highest discriminant function.
One study of leukotomized schizophrenic subjects did not provide strong support for use of the FBI in other frontal disorders. Black and colleagues15 collected measures of frontal behavior (FBI), psychopathology positive and negative syndrome scale (PANSS), neurologic examinations, and computed tomography scans in 19 chronically institutionalized, elderly schizophrenic patients who had undergone orbitofrontal leukotomy and 11 control subjects matched for age, length of hospitalization, education, and diagnosis. There was a tendency for the leukotomized group to have fewer indexes of frontal behavioral dysfunction. There were no significant differences between leukotomized patients and controls on Folstein Mini-Mental State score, utilization behavior, Luria alternating written and motor sequences, verbal fluency, imitation behavior, motor impersistence, primitive reflexes, or psychopathology. Significant differences were found on clock drawing and on a go/no-go test. Both leukotomized and nonleukotomized schizophrenic patients showed varying degrees of distractibility, difficulty in set shifting, poor planning and organization, susceptibility to interference, primitive reflexes, and signs of global cognitive impairment.
There are no norms available for the FBI to date.
Frontal Systems Behavior Scale (FrSBe)
The FrSBe16 is a 46-item behavior rating scale that is intended to measure behavior associated with damage to the frontal systems of the brain. According to the manual, the FrSBe was designed (a) to be brief, reliable, and valid; (b) to assess adult behavior before (ie, premorbid baseline) and after frontal systems damage occurs; and (c) to permit multiple observers to provide behavior ratings. It 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, such as spouse, child, relative, or significant other. Professionals (eg, rehabilitation staff) seeking to track patient behavior over time also can use the family form to rate a patient's current behavior. Each FrSBe form yields a Total score and scores for subscales measuring Apathy, Disinhibition, and Executive Dysfunction. Scores are obtained on each scale for baseline behavior and current behavior. Thus, behavior change can be indexed by comparing T scores based on normative data of the ratings of prior and current patient behavior.
High internal consistency has been demonstrated in multiple studies. The FrSBe manual16 reports α coefficients of 0.92, 0.78, 0.80, and 0.87 for the Total, Apathy, Disinhibition, and Executive scales of the family form and 0.88, 0.72, 0.75, and 0.79 for the self-form in a normative sample. Stout and colleagues17 found α values were 0.94, 0.87, 0.84, and 0.91 for the Total, Apathy, Disinhibition, and Executive scales in a large neurologic sample. Velligan et al18 found α values were 0.94, 0.88, 0.86, and 0.91 for Total scale, Apathy, Disinhibition, and Executive subscales, respectively, in schizophrenia patients. They also found test-retest reliability after 3 months was 0.78 for the Total score and 0.68, 0.65, and 0.65 for Apathy, Disinhibition, and Executive Dysfunction subscales, respectively.
The strongest evidence of construct validity of the FrSBe was addressed by Grace et al.19 They studied 24 patients with frontal lobe brain damage, 15 patients without frontal lobe brain damage, and 48 healthy control subjects. Construct validity was examined by comparing the FrSBe Total scores of patients with frontal systems lesions with (a) their behavior prior to the lesions, (b) behavior of normal control participants, and (c) behavior of patients with nonfrontal lesions. A significant behavior change on the FrSBe was found when premorbid (before) baseline scores of frontal lesion patients were compared with their postlesion behavior. Frontal lesion patients scored significantly higher on the FrSBe than the nonfrontal lesion and healthy control groups.
Stout et al17 conducted a factor analysis in a sample of 324 neurologic patients and research participants, of whom about 63% were diagnosed with neurodegenerative diseases (Huntington disease, Parkinson disease, and AD). A three-factor solution confirmed a factor structure consistent with the three subscales proposed on the theoretical basis of the frontal systems. Most items (83%) from the FrSBe subscales of Apathy, Disinhibition, and Executive Dysfunction loaded on three corresponding factors. The exploratory principal factor analysis suggests that some revision or elimination of specific items may be warranted to refine the scale and enhance the validity of the subscales.
Convergent validity was demonstrated by Norton et al20 in a study of the FrSBe and NPI in 30 dementia patients and their caregivers. Significant correlations were found between the NPI Total and FrSBe Total scores, NPI and FrSBe Apathy, and NPI and FrSBe Disinhibition. Convergent validity was also supported by Velligan et al,18 who found differential relationships for the Apathy and Disinhibition subscales with symptoms rated from the Brief Psychiatric Rating Scale and with cognitive variables.
Cahn-Weiner and colleagues21 demonstrated the discriminant validity of the Apathy subscale, showing it did not correlate significantly with the Geriatric Depression Scale in Alzheimer and Parkinson patients. Ready et al22 also found that negative mood (eg, sadness, pessimism) was not significantly correlated with FrSBe Apathy, whereas items that measured loss of interest and motivation (eg, loss of interest, loss of energy) were significantly correlated with FrSBe Apathy. These findings suggest that apathy and depression are two distinguishable syndromes (see also the review of the Apathy Evaluation Scale below).
The FrSBe has been employed to characterize dementias with different underlying neuropathology. For example, research has shown that FrSBe behavior profiles differed in meaningful ways for cortical (AD) and subcortical (Huntington disease, Parkinson disease) dementias.21,23 Specifically, the subcortical dementia patients were found to display greater levels of apathy. Further, FrSBe subscales have been found to relate to AD progression when frontal systems involvement increases.22,24 In addition, psychotic AD patients have been found to score higher on the Disinhibition scale of the FrSBe than nonpsychotic AD patients.25
Although not ideal groups for validation of the scale, weaker evidence of the validity of the FrSBe has been provided by studies of psychiatric patients who may have frontal system dysfunction, such as schizophrenic patients. The most comprehensive study using the FrSBe in a psychiatric population was conducted by Velligan et al,18 who compared 131 schizophrenic patients with 51 education-matched control subjects. Schizophrenic patients had significantly greater impairment on all FrSBe scales than control subjects. They also found that the FrSBe was correlated with two measures of adaptive functioning. Cognitive tests had differential and meaningful relationships to FrSBe behavioral profiles. For example, Disinhibition scores correlated with Trails B errors and with false alarm errors on Continuous Performance Testing. In contrast, Apathy scores correlated with Verbal Fluency and Trails B Time. Spinella26 recently found that polysubstance users showed greater dysfunction on the Disinhibition subscale than nonpolysubstance users. Other psychiatric applications are reviewed in the professional manual.16
Studies of the ecologic validity of the FrSBe have been conducted. For example, Executive Cognitive Dysfunction and Apathy measured by the FrSBe are associated with decline in instrumental activities of daily living (ADLs) in dementia patients.20,27 Further, the FrSBe Disinhibition subscale has been shown to relate to caregiver burden in AD.28 The FrSBe accounted for variance in ADLs and burden even after controlling for executive cognitive dysfunction.
Pre- and postsurgical assessment with the FrSBe has been reported in Parkinson disease for both posteroventral pallidotomy (PVP) and deep brain stimulation (DBS) of the internal globus pallidus (GPi) or subthalamic nucleus (STN).29-32 Behavioral dyscontrol was documented by the FrSBe in post-PVP or bilateral STN or GPi-DBS for 25% of patients postoperatively, and behavior change persisted in some patients at the 9- to 12-month follow-up evaluation.
Behavioral consequences of multiple sclerosis (MS) have been assessed with the FrSBe. As has been found in studies of AD, Parkinson disease, and Huntington disease, Chiarvalloti and DeLuca33 reported that preillness FrSBe scores of an MS group were in the normal range on both self and family forms. In contrast, after illness, clinically and statistically significant increases in abnormal behavior were noted, especially on the FrSBe Apathy and Executive scales. These clinical abnormalities correlated with neuropsychological tests assessing information processing, working memory, and executive control.
Moderate correlations between the FrSBe and neuropsychological tests that measure working memory and executive control have been reported in several studies.18,23 Whereas a relationship between frontal behavioral syndromes and cognition supports the construct validity of the measure, it is important that relationships are moderate, thus demonstrating the unique contribution of the measurement of noncognitive behavior change.
The FrSBe Professional Manual16 provides norms for self-ratings and family ratings. The normative sample included 436 men and women ranging in age from 18 to 95 years and in education from 10 years to doctoral level. Normative tables stratified for gender, age, and education provide T scores for the self-rating form and the family rating form.
Iowa Rating Scales of Personality Change (IRSPC)
The IRSPC was developed by Barrash and colleagues34 after their research indicated that standard personality instruments such as the Minnesota Multiphasic Personality Inventory (MMPI) were insensitive to changes occurring after frontal injury. They were also concerned that their instrument measure change from premorbid personality characteristics and that information from informants be used, given that frontal patients may lack insight. The IRSPC assesses 30 characteristics. For each characteristic, the informant rates both level (the extent to which the behavior is present) and degree of change from premorbid baseline. Informants are provided with descriptive exemplars of various levels on each scale. Twenty-seven items assess emotional functioning, behavioral control, social and interpersonal behavior, and higher-order cognitive abilities. Three additional items assess areas not thought to be associated with brain impairment (frugality, manipulativeness, and type A behavior), to allow detection of response bias.
Interrater agreement for the scales was high in a sample of focal lesion patients, ranging from 0.80 to 0.96 for both level and change scores.
Barrash et al34 studied personality changes measured by the IRSPC in 7 participants with bilateral ventromedial prefrontal lesions (PF-BVM), 14 participants with prefrontal lesions but not bilateral ventromedial involvement (PF-NBVM), and 36 with nonprefrontal lesions (NPF). PF-BVM participants showed a higher rate of acquired disturbances than NPF participants in blunted emotional experience, apathy, low emotional expressiveness, inappropriate affect, poor frustration tolerance, irritability, lability, indecisiveness, poor judgment, social inappropriateness, lack of planning, lack of initiation and persistence, and lack of insight. Differences between the PF-BVM and PF-NBVM groups were significant for several of these characteristics. What is especially notable is the scale's ability to capture a highly specific relationship between the acquired sociopathy behavioral syndrome and ventromedial prefrontal damage. Barrash et al report a factor analysis of the IRSPC based on a sample of 115 brain-damaged patients of mixed etiologies. They interpreted the analysis as supporting the conceptualization of the scale in assessing emotional functioning, behavioral control, social and interpersonal behavior, and higher-order cognitive functions.
There have been no published studies to date using the IRSPC in normal control subjects.
Neuropsychiatric Inventory (NPI)
The NPI35 was designed to measure a wide range of behavioral disturbance in dementia patients. It includes 10 items assessing delusions, hallucinations, dysphoria, anxiety, agitation/aggression, euphoria, disinhibition, irritability/lability, apathy, and aberrant motor activity. Information for the NPI is obtained from a caregiver familiar with the patient's behavior. The NPI uses a screening strategy to minimize administration time, examining and scoring only those behavioral domains with positive responses to screening questions. Both the frequency and the severity of each behavior are determined. The frequency and severity scores can be summed for two total scores on the NPI. Caregivers are also asked to estimate the amount of distress or burden that each behavior causes the family (not the patient).
Interrater reliability was assessed by having two raters score the NPI responses of 45 caregivers. Agreements between raters were all high, ranging from 89% to 100% for frequency and severity of each domain. Correlations ranged from 0.51 to 1.0. Test-retest reliability was determined by having 20 caregivers assessed twice within 3 weeks. Correlations were all quite high, with the overall correlation being 0.79 for frequency and 0.86 for severity.35
Content validity was tested by having 10 experts rate how well each question captured the intended behavioral domain. All but one question was rated highly. “Troublesome behavior” was thereafter reformulated as “aberrant motor behavior.” Concurrent validity was initially determined by comparing some of the subscale scores on the NPI with the corresponding scales or items of the Behavioral Pathology in Alzheimer's Disease (BEHAVE-AD)36 Rating Scale and Hamilton Depression Rating Scale. Moderate to high correlations were found in most cases. Several of the scores also had low to moderate correlations with Mini-Mental State Examination scores.35 Norton et al20 also demonstrated that the NPI Apathy and Disinhibition items correlated significantly with the FrSBe Disinhibition (r = 0.62) and Apathy (r = 0.37) scales.
Although it was not designed for this purpose, a number of studies have suggested that the NPI is sensitive to behavioral changes due to frontal systems impairment. For example, Geroldi et al37 investigated the relationship between delusions and regional brain atrophy in 41 AD patients with mild dementia severity. Presence or absence of delusions was assessed with the pertinent subscale of the NPI. AD patients without delusions had symmetrical enlargement of both temporal and frontal horns on computed tomography scan. AD with delusions showed relatively larger right temporal horns and left frontal horns.
Tekin et al38 demonstrated relationships between NPI scores and frontal pathology at autopsy. They identified 31 autopsy patients with a diagnosis of definite AD and examined their prior NPI scores. Brain sections were collected from bilateral orbitofrontal and left anterior cingulate, superior temporal, inferior parietal, occipital, and hippocampal cortices for quantification of neurofibrillary tangles (NFTs) and diffuse and neuritic plaques. Correlations were found between agitation and aberrant motor behavior on the NPI and greater NFT pathology in the orbitofrontal cortex in AD, whereas increasing apathy on the NPI was related to greater NFT burden in the anterior cingulate.
Mega et al39 retrospectively explored the behavioral and functional imaging profile of AD patients who responded to cholinesterase inhibitor therapy by using the NPI and baseline hexamethylpropyleneamine oxime single-photon emission computed tomography (SPECT). Thirty AD patients were divided into three groups (responders, nonresponders, and unchanged) based on their behavioral response to donepezil. Responders had significantly more pretreatment irritability, disinhibition, and euphoria on the NPI than nonresponders and significantly lower lateral orbital frontal and dorsolateral frontal perfusion bilaterally. The authors suggested that pretreatment orbitofrontal syndrome may predict behavioral response to cholinesterase inhibitor therapy in AD.
Frisoni and colleagues40 studied 162 consecutive patients with probable AD admitted to a dementia unit. Factor analysis was carried out on NPI subscales, leading to three syndromes: mood, psychotic, and frontal. Patients with the psychotic syndrome were older, had older age at dementia onset, had poorer cognition, were more often males, and had faster rate of dementia progression. Patients with the frontal syndrome had higher education, longer disease duration, and slower rate of progression.
Levy et al41 compared 22 patients with FTD and 30 patients with AD defined by research diagnostic criteria and SPECT. Groups were matched for dementia severity. Patients with FTD had significantly greater total NPI scores than patients with AD and exhibited more apathy, disinhibition, euphoria, and aberrant motor behavior. The NPI accurately assigned 77% of patients with FTD and 77% of patients with AD to the correct diagnostic group using disinhibition, apathy, and depression. Patients with FTD had higher levels of disinhibition and apathy with relatively lower levels of depression than patients with AD.
There is one study that presents limited normative date on the NPI. The instrument was administered to 40 spouses of normal, community-dwelling elderly persons.42 These normal participants scored 0 on all subscales except Dysphoria, Disinhibition, and Irritability. Their mean scores on these subscales were 0.25, 0.13, and 0.05, and their top scores were 6, 4, and 2, respectively. Discriminant validity between demented and normal or frontal and normal subjects has not been assessed in large samples to date. Effects of age and education on NPI scores have not yet been reported.
A number of other measures have been developed to measure aspects of behavior change in frontal disorders. Most of the scales reviewed in this section have received relatively little research attention to date and are not as well validated as the previously reviewed scales.
Shigenobu et al43 argued that stereotypies are among the best discriminators of FTD from other dementias. To assess the wide range of stereotypic behaviors encountered in FTD, they developed the Stereotypy Rating Inventory (SRI). The SRI assesses five distinctive stereotypic behavioral disturbances often seen in patients with FTD: eating and cooking behaviors, roaming, speaking, movements, and daily rhythm. The SRI uses the same technique as the NPI in that both the frequency and the severity of each behavior are determined. The authors presented data supporting the interrater and test-retest reliability and the content and concurrent validity of the instrument. Scores of FTD patients on the SRI were significantly higher than those of patients with AD, patients with VaD, and normal control subjects.
Bozeat and colleagues44 also designed a questionnaire to assess a wide range of neuropsychiatric changes in FTD. It incorporated features reported in previous studies of FTD and components of the NPI. The scale was completed by 37 caregivers of patients with AD and 33 patients with FTD (20 with temporal variant and 13 with frontal variant). Factor analysis showed four symptom clusters: stereotypic and eating behavior, executive dysfunction and self-care, mood changes, and loss of social awareness. Only stereotypic/eating behavior and loss of social awareness reliably differentiated AD from FTD. By contrast, executive dysfunction, poor self-care, and restlessness showed a significant effect of disease severity only, with the more impaired patients scoring more highly. Discriminant function analysis correctly classified 71.4% overall and 86.5% of the patients with AD. The patients with temporal and frontal FTD were behaviorally very similar on this scale, however.
Grafman et al45 examined the relationship between frontal lobe lesions and violent behavior in veterans participating in the Vietnam Head Injury Study. Fifty-seven normal control subjects and 279 veterans, matched for age, education, and time in Vietnam, who had experienced penetrating head injuries during their service in Vietnam, were studied. Two Aggression/Violence Scale scores, based on family observer ratings, were constructed. The results indicated that patients with frontal ventromedial lesions consistently demonstrated Aggression/Violence Scale scores significantly higher than control subjects and patients with lesions in other brain areas. Higher Aggression/Violence Scale scores were generally associated with verbal confrontations rather than physical assaults, which were less frequently reported.
In addition to the FrSBe Apathy subscale and the Apathy subscale of the NPI, a number of measures have been designed to specifically assess the apathy syndrome: the Dementia Apathy Interview and Rating,46 the Apathy Inventory,47 the Irritability-Apathy Scale,48 Starkstein's 14-item scale,49 and the Apathy Evaluation Scale (AES).50 The Apathy Inventory is a new measure of particular interest as it is specifically designed to provide separate assessments of emotional, cognitive, and behavioral aspects of apathy. The AES is perhaps the best known of these apathy measures and has been quite widely used in research. It is an 18-item scale utilizing 4-point Likert-type items. It was developed with the same core items for multiple raters including clinician (AES-C), informant (AES-I), and self-rated (AES-S) versions. Each AES form yields a total score, with higher scores indicating presence of greater degree of apathy. As reported in the original article,50 internal consistency reliability (coefficient α) was 0.86-0.94 across the three forms. Test-retest reliability ranged from 0.76 to 0.94 for informant and clinician ratings and 0.44 for AD self-ratings. The core 18 items were selected from a pool of 56 items based upon item-total correlations, correlation with the Hamilton Rating Scale for Depression (HRSD), and a preliminary factor analysis on a very small sample (n = 40). The original article then reports a validation study with a total of 123 subjects from five diagnostic groups: well elderly, left hemisphere stroke, right hemisphere stroke, probable AD, and major depression (MD). The AES discriminated between right hemisphere stroke, AD, and MD groups compared with well elderly, showed discriminability of apathy ratings from standard measures of depression and anxiety, and demonstrated predictive validity (ie, apathy ratings related to initiation and termination of engagement with video games and novelty toys). Further studies have explored the relationship between the AES and HRSD, finding convergence due to a subset of HRSD items consistent with the apathy syndrome. There is a consistent association of apathy with depression, but the relationship between apathy and depression on these measures differs across diagnostic groups.51,52 Recent studies have used the AES in TBI with mixed results. Andersson and Bergedalen53 found the AES correlated significantly with impaired memory, psychomotor speed, and tests of executive functions. However, Glenn et al54 found that the AES did not have reasonable sensitivity and specificity in predicting clinician impression of the presence of apathy. No studies with the AES have specifically focused on behavior change related to frontal systems damage. Marin et al50 presented mean scores for the three forms in a small, well elderly sample (n = 31). However, there have been no published studies to date examining the AES in a large sample of normal control subjects stratified for age or educational levels.
Several scales have been developed to specifically address behavioral functioning in TBI. Whereas frontal systems are much involved in TBI, review of these scales is beyond the scope of this article. The interested reader may wish to investigate the following scales: Key Behaviors Change Inventory,55 Neuropsychology Behavior and Affect Profile,56 Neurobehavioral Rating Scale,57 and the Neurobehavioral Inventory.58
The FBI and NPI are administered via interview by an expert rater, whereas the BRIEF, DEX, IRSPC, and FrSBe are completed by a caregiver familiar with the patient. There are forms of the DEX, FrSBe, and IRSPC that can be filled out by the patient for comparison with caregiver ratings. A form of the NPI designed to be filled out by caregivers (rather than expert raters) was developed by Kaufer et al.59 Both administration methods have advantages. The interview method permits the rater to ask detailed questions about a given problem but requires that the rater be knowledgeable about relatively rare syndromes (eg, alien hand) and capable of distinguishing similar syndromes (eg, depression and apathy). Caregiver-completed rating scales can be completed with minimal supervision by the examiner and hence are very cost efficient.
All the scales reviewed provide indexes of the frequency or severity of the behavior. However, measurement issues with rating scales persist. It is difficult to include in one scale the measurement of discrete, episodic behavior such as getting in trouble with the law and continuous behavior such as lack of motivation that is present daily. Some rare disruptive behaviors may be highly salient and greatly impact daily functioning (eg, Capgras delusions), but may not lend themselves to easy measurement with a Likert-type scale.
The DEX, FBI, and IRSPC do not have standardized norms, and the NPI has only a small sample of normative data, whereas the BRIEF and FrSBe are supported by large normative samples. One might argue that the abnormal behaviors assessed by the FBI and NPI, such as delusions and alien hand, occur so infrequently in normal persons that norms are superfluous. However, this is ultimately an empirical question, and it would be reassuring to clinicians to know that false positives do not occur with these scales. Although it may be apparent that there has been some change from premorbid functioning, it will not be clear whether this change is a significant difference or is pathologic in severity.
In the first of two validation studies of the FBI,11 the FTD and AD groups differed in age (as might be expected clinically) and in gender. However, there were no tests for age or gender effects, and these factors were not covaried in the group comparisons on the FBI. Given that substantial differences were found in younger people and in males in normative studies of the FrSBe,16 this is of some concern and underlines the need for normative interpretation standards. Some of the FBI items that loaded highly on the discriminate function14 separating frontal dementias from other disorders are relatively rare, even in frontal populations (eg, alien hand syndrome). It would be expected that such items would have high specificity but low sensitivity. It remains to be seen, therefore, whether these results will survive cross-validation.
In addition to the research demonstrating its reliability and validity in discriminating patients with frontal from nonfrontal focal lesions, the FrSBe has been found to adequately discriminate dementias with more frontal system neuropathology from dementias with more posterior neuropathology and to have good reliability in psychiatric patients. Moderate relationships have been found with traditional neuropsychological measures typically associated with frontal systems. In terms of ecologic validity, scores on the FrSBe have also been shown to correlate with important everyday outcomes in dementia and psychiatric patients, including failure on instrumental ADLs and caregiver burden. In the future, the factor analysis of the FrSBe may be used as a basis for further refinement of the scale and potentially allow the scale to be shortened. Further research with the FrSBe is needed in additional pertinent populations, especially head injury and adults with ADHD.60
The FrSBe and the IRSPC are the only measures that have been shown to discriminate frontal from nonfrontal focal lesion groups.19,34 There is no study to date demonstrating that the BRIEF can discriminate frontal lesions from other lesion locations or other disorders. In fact, the authors note that “while the BRIEF captures executive profiles characteristic of specific disorders in the clinical setting, it is not diagnostic in its own right and is best used within the context of a broad based evaluation.” The FBI has been shown to discriminate FTD from AD, but it failed to discriminate leukotomized schizophrenic patients from control subjects in the single study of focal lesion patients using that scale. The NPI was designed to be a wide-ranging inventory of neuropsychiatric and behavioral changes in dementia and is widely used for this purpose. However, it contains several items rating behaviors thought to be subserved by frontal systems, such as disinhibition and apathy. The NPI has not yet been applied to focal lesion groups, but research reviewed above indicates that it is sensitive to frontal systems change measured via functional neuroimaging.
To date, the NPI is the only rating scale in widespread use in clinical trials. In fact, the NPI has become the de facto standard for measuring neuropsychiatric symptoms in industry-sponsored clinical trials for dementia. Further research is needed on the ability of all the scales reviewed here to measure behavior change longitudinally.
In summary, substantial progress has been made in the measurement of the complex behavioral syndromes that result from damage to frontal systems. The early confusing and contradictory descriptions of frontal behavior change have been replaced by more systematic behavioral description in cortical and subcortical lesion cases, in different age ranges, in multiple dementia diagnostic groups and multiple stages of dementia severity, and in psychiatric groups and syndromes (apathy versus depression). Scales are available that compare premorbid and postinsult behavior (FrSBe, IRSPC), frontal versus nonfrontal lesions (FrSBe, IRSPC), dementia and other neurologic disorders (DEX, FBI, FrSBe, NPI), childhood disorders (BRIEF), and neuropsychiatric disorders (FrSBe, NPI). Also, the investigator may choose interview formats or family/self-rating forms. Although further research is clearly necessary for all scales, clinicians and investigators can now choose among several measures that have research supporting the reliability and validity of the instrument.
1. Eslinger PJ, Damasio AR. Severe disturbance of higher cognition after bilateral frontal lobe ablation: patient EVR. Neurology
2. Alexander GE, DeLong MR, Strick PL. Parallel organization of functionally segregated circuits linking basal ganglia and cortex. Annu Rev Neurosci
3. Joel D. Open interconnected model of basal ganglia-thalamocortical circuitry and its relevance to the clinical syndrome of Huntington's disease. Mov Disord
4. Cummings JL. Frontal subcortical circuits and human behavior. Arch Neurol
5. Mega MS, Cummings JL. Frontal subcortical circuits and neuropsychiatric disorders. J Neuropsychiatry Clin Neurosci
6. Gioia GA, Isquith PK, Guy SC, et al. Behavior rating inventory of executive function. Neuropsychol Dev Cogn
7. Gioia G, Isquith P, Kenworthy L, et al. Profiles of everyday executive function in acquired and developmental disorders. Neuropsychol Dev Cogn
8. Bogod NM, Mateer CA, MacDonald SW. Self-awareness after traumatic brain injury: a comparison of measures and their relationship to executive functions. J Int Neuropsychol Soc
9. Evans JJ, Chua SE, McKenna PJ, et al. Assessment of the dysexecutive syndrome in schizophrenia. Psychol Med
10. Chan RC. Dysexecutive symptoms among a non-clinical sample: a study with the use of the Dysexecutive Questionnaire. Br J Psychol
11. Kertesz A, Davidson W, Fox H. Frontal behavioral inventory: diagnostic criteria for frontal lobe dementia. Can J Neurol Sci
12. The Lund and Manchester Groups. Clinical and neuropathological criteria for frontotemporal dementia. J Neurol Neurosurg Psychiatry
13. Neary D, Snowden JS, Gustafson L, et al. Frontotemporal lobar degeneration: a consensus on clinical diagnostic criteria. Neurology
14. Kertesz A, Nadkarni N, Davidson W, et al. The Frontal Behavioral Inventory in the differential diagnosis of frontotemporal dementia. J Int Neuropsychol Soc
15. Black DN, Stip E, Bedard M, et al. Leukotomy revisited: late cognitive and behavioral effects in chronic institutionalized schizophrenics. Schizophr Res
16. Grace J, Malloy PF. Frontal Systems Behavior Scale (FrSBe): Professional Manual
. Lutz, FL: Psychological Assessment Resources; 2001.
17. Stout JC, Ready RE, Grace J, et al. Factor analysis of the Frontal Systems Behavior Scale (FrSBe). Assessment
18. Velligan DI, Ritch JL, Sui D, et al. Frontal Systems Behavior Scale in schizophrenia: relationships with psychiatric symptomatology, cognition and adaptive function. Psychiatry Res
19. Grace J, Stout JC, Malloy PF. Assessing frontal lobe behavioral syndromes with the Frontal Lobe Personality Scale. Assessment
20. Norton LE, Malloy PF, Salloway S. The impact of behavioral symptoms on activities of daily living in patients with dementia. Am J Geriatr Psychiatry
21. Cahn-Weiner DA, Grace J, Ott BR, et al. Cognitive and behavioral features discriminate between Alzheimer's and Parkinson's disease. Neuropsychiatry Neuropsychol Behav Neurol
22. Ready RE, Ott BR, Grace J, et al. Apathy and executive dysfunction in mild cognitive impairment and Alzheimer's disease. Am J Geriatr Psychiatry
23. Paulsen JS, Stout JC, DelaPena J, et al. Frontal behavioral syndromes in cortical and subcortical dementia. Assessment
24. Stout JC, Wyman MF, Johnson SA, et al. Frontal behavioral syndromes and functional status in probable Alzheimer's disease. Am J Geriatr Psychiatry
25. Paulsen JS, Ready RE, Stout JC, et al. Neurobehaviors and psychotic symptoms in Alzheimer's disease. J Int Neuropsychol Soc
26. Spinella M. Relationship between drug use and prefrontal-associated traits. Addict Biol
27. Boyle PA, Malloy PF, Salloway S, et al. Executive dysfunction and apathy predict functional impairment in Alzheimer disease. Am J Geriatr Psychiatry
28. Rymer S, Salloway S, Norton L, et al. Impaired awareness, behavior disturbance, and caregiver burden in Alzheimer disease. Alzheimer Dis
29. Saint-Cyr JA, Trepanier LL, Kumar R, et al. Neuropsychological consequences of chronic bilateral stimulation of the subthalamic nucleus in Parkinson's disease. Brain
30. Saint-Cyr JA, Trepanier LL. Neuropsychologic assessment of patients for movement disorder surgery. Mov Disord
31. Trepanier LL, Saint-Cyr JA, Lozano AM, et al. Neuropsychological consequences of posteroventral pallidotomy for the treatment of Parkinson's disease. Neurology
32. Trepanier LL, Kumar R, Lozano AM, et al. Neuropsychological outcome of GPi pallidotomy and GPi or STN deep brain stimulation in Parkinson's disease. Brain Cogn
33. Chiaravalloti N, DeLuca J. Assessing the behavioral consequences of multiple sclerosis: an application of the Frontal Systems Behavior Scale. Cogn Behav Neurol
34. Barrash J, Tranel D, Anderson SW. Acquired personality disturbances associated with bilateral damage to the ventromedial prefrontal region. Dev Neuropsychol
35. Cummings JL, Mega M, Gray K, et al. The Neuropsychiatric Inventory: comprehensive assessment of psychopathology in dementia. Neurology
36. Reisberg B, Auer SR, Monteiro IM. Behavioral pathology in Alzheimer's Disease (BEHAVE-AD) Rating Scale. Int Psychogeriatr
. 1996;8(suppl 3):301-308, 351-354.
37. Geroldi C, Bresciani L, Zanetti O, et al. Regional brain atrophy in patients with mild Alzheimer's disease and delusions. Int Psychogeriatr
38. Tekin S, Mega MS, Masterman DM, et al. Orbitofrontal and anterior cingulate cortex neurofibrillary tangle burden is associated with agitation in Alzheimer disease. Ann Neurol
39. Mega MS, Dinov ID, Lee L, et al. Orbital and dorsolateral frontal perfusion defect associated with behavioral response to cholinesterase inhibitor therapy in Alzheimer's disease. J Neuropsychiatry Clin Neurosci
40. Frisoni GB, Rozzini L, Gozzetti A, et al. Behavioral syndromes in Alzheimer's disease: description and correlates. Dement Geriatr Cogn Disord
41. Levy ML, Miller BL, Cummings JL, et al. Alzheimer disease and frontotemporal dementias. Behavioral distinctions. Arch Neurol
42. Cummings JL. The Neuropsychiatric Inventory: assessing psychopathology in dementia patients. Neurology
. 1997;48 (suppl 6):S10-S16.
43. Shigenobu K, Ikeda M, Fukuhara R, et al. The Stereotypy Rating Inventory for frontotemporal lobar degeneration. Psychiatry Res
44. Bozeat S, Gregory CA, Ralph MA, et al. Which neuropsychiatric and behavioural features distinguish frontal and temporal variants of frontotemporal dementia from Alzheimer's disease? J Neurol Neurosurg Psychiatry
45. Grafman J, Schwab K, Warden D, et al. Frontal lobe injuries, violence, and aggression: a report of the Vietnam Head Injury Study. Neurology
46. Strauss ME, Sperry SD. An informant-based assessment of apathy in Alzheimer disease. Neuropsychiatry Neuropsychol Behav Neurol
47. Robert PH, Clairet S, Benoit M, et al. The Apathy Inventory: assessment of apathy and awareness in Alzheimer's disease, Parkinson's disease and mild cognitive impairment. Int J Geriatr Psychiatry
48. Burns A, Folstein S, Brandt J, et al. Clinical assessment of irritability, aggression, and apathy in Huntington and Alzheimer disease. J Nerv Ment Dis
49. Starkstein SE, Mayberg HS, Preziosi TJ, et al. Reliability, validity, and clinical correlates of apathy in Parkinson's disease. J Neuropsychiatry Clin Neurosci
50. Marin RS, Biedrzycki RC, Firinciogullari S. Reliability and validity of the Apathy Evaluation Scale. Psychiatry Res
51. Marin RS, Firinciogullari S, Biedrzycki RC. The sources of convergence between measures of apathy and depression. J Affect Disord
52. Marin RS, Firinciogullari S, Biedrzycki RC. Group differences in the relationship between apathy and depression. J Nerv Ment Dis
53. Andersson S, Bergedalen AM. Cognitive correlates of apathy in traumatic brain injury. Neuropsychiatry Neuropsychol Behav Neurol
54. Glenn MB, Burke DT, O'Neil-Pirozzi T, et al. Cutoff score on the Apathy Evaluation Scale in subjects with traumatic brain injury. Brain Inj
55. Kolitz BP, Vanderploeg RD, Curtiss G. Development of the Key Behaviors Change Inventory: a traumatic brain injury behavioral outcome assessment instrument. Arch Phys Med Rehabil
56. Nelson L, Satz P, Mitrushina M, et al. Development and validation of the Neuropsychology Behavior and Affect Profile. Psychol Assess
57. Levin HS, High WM, Goethe KE, et al. The Neurobehavioural Rating Scale: assessment of the behavioural sequelae of head injury by the clinician. J Neurol Neurosurg Psychiatry
58. Kreutzer JS, Marwitz JH, Seel R, et al. Validation of a neurobehavioral functioning inventory for adults with traumatic brain injury. Arch Phys Med Rehabil
59. Kaufer DI, Cummings JL, Ketchel P, et al. Validation of the NPI-Q, a brief clinical form of the Neuropsychiatric Inventory. J Neuropsychiatry Clin Neurosci
60. Dinn WM, Robbins NC, Harris CL. Adult attention-deficit/hyperactivity disorder: neuropsychological correlates and clinical presentation. Brain Cogn