The Behavioral Competence Inventory: A Measure of Functional Ability in Older Adults at Risk for Dementia

Paula Hartman-Stein, Ph.D.
Center for Healthy Aging
Kent, Ohio USA

Jeanette Reuter, Ph.D. Joneen Schuster, M.A
Kent Developmental Metrics Kent State University
Kent, Ohio USA Kent, Ohio USA

A poster presented at the 8th International Conference on Alzheimer’s Disease and Related Disorders

July 22, 2002, Stockholm, Sweden

Recent guidelines for the evaluation of dementia and age-related cognitive decline specify the need to obtain behavioral descriptions and estimations of performance from collateral sources (APA, 1998). In addition, in order to design appropriate behavioral and environmental interventions and monitor the efficacy of pharmacological treatment of symptoms of dementia, clinicians need more than neuropsychological measures of cognition and memory. The Behavioral Competence Inventory (BCI) (Hartman-Stein & Reuter, 1996), a measure of general behavioral functioning of an adult with memory concerns as observed by a knowledgeable caregiver, addresses these needs.

The purpose of this study is to further explore the utility of the BCI by examining the effects of mental status, rating of dementia, and depression on the breadth of the functional behavioral repertoire of older adults at risk for dementia from Alzheimer’s disease or related disorders. This study of an outpatient sample of adults referred to a psychologist because of their memory concerns examined the concurrent validity of the BCI by comparing it to other clinically useful scales that measure level of impairment or have value in the assignment of the diagnosis of depression. The BCI provides a structured and efficient method of gaining input from family and other caregivers that is an essential piece of a comprehensive geriatric assessment.



The subjects were referred to the out-patient private practice of the first author in Northeast Ohio for a cognitive assessment between 1996 and 2002. Family members or other informants completed the BCI for the 152 subjects included in this study. Thirty-two percent of the subjects were men, and 68 % were women. Their ages ranged from 52 to 90 years (mean=76.2 years, SD= 7.1). The sample, largely from a suburban area, was 97% Caucasian and 3% from minority groups, including African-American and Japanese-American.

The informants who completed the BCI included adult children or children–in-law (43%), spouses (40%), other relatives (10%) and non-relatives (7%).


Five clinical psychologists specializing in geropsychology, including the first author, conducted the evaluations that included semi-structured interviews with the patient and family member or caregiver, review of available medical records, and cognitive testing. A standard cognitive battery included the following: Mini-Mental State Exam (Folstein, Folstein, and McHugh, 1975); Dementia Rating Scale (Mattis, 1988) or Neurobehavioral Cognitive Status Exam (COGNISTAT) (Osato, Yang, & LaRue, 1993); Clock Face drawing (Goodglass & Kaplan, 1983); Logical Memory I and II from the Wechsler Memory Scale-Revised (Wechsler, 1987); Controlled Oral Word Associations (Benton, Hamsher, Rey & Sivan, 1994), Boston Naming Test (Kaplan, Goodglass, & Weintraub, 1983); Trailmaking A and B (Reitan, 1958); Comprehension, Vocabulary, and Block Design sub-tests from the Wechsler Adult Intelligence Scale-Revised (Wechsler, 1981); Geriatric Depression Scale (Yesavage et al., 1983); Cornell Scale for Depression in Dementia (Alexopolous, Abrams, Young, & Shamoian, 1988); and Health and Safety subtest from the Independent Living Scales (Loeb, 1996). Following the evaluation the clinician staged the level of impairment using the Clinical Dementia Rating Scale (Berg, 1988; Morris, 1998).

A subset of the measures from the complete evaluation including the Mini-Mental State Exam (MMSE), Geriatric Depression Scale (GDS), Clinical Dementia Rating (CDR) as well as the Behavioral Competence Inventory (BCI), were used in the analyses of this study.

The five clinical psychologists determined the diagnoses of the outpatient sample based upon all available data except the score from the patient’s BCI. The clinicians assigned diagnoses based upon criteria described in the Diagnostic and Statistical Manual of Mental Disorders-Fourth edition (American Psychiatric Association, 1994).

Seventy-five patients were assigned a primary diagnosis of Dementia of the Alzheimer’s type, 20 patients were diagnosed with Vascular dementia, and 10 patients received a diagnosis of "other dementia." Approximately 70% of the sample was diagnosed with dementia. Twenty patients were diagnosed with major depression as the primary diagnosis, and 27 patients were assigned other psychiatric diagnoses including personality disorders, substance abuse, or cognitive disorder, not otherwise specified. Of the patients diagnosed with dementia, 67 (64%) were diagnosed with dementia and depression.


Behavioral Competence Inventory (BCI) (Hartman-Stein & Reuter,1996). The BCI is a caregiver report measure that consists of 106 items and purports to measure separate domains of functional behaviors. The items are divided into seven scales; no item contributes to more than one scale. The Self-Care (SC) Scale consists of 12 items such as independence in bathing, grooming, and toileting. The Instrumental Activities of Daily Living (IDL) Scale contains 23 items describing such behaviors as the ability to use the telephone, drive safely, manage medications, and shop independently for food. The Memory/Orientation (MEM) scale includes 11 items such as remembering short lists, recognizing people, and knowing the date and day of the week. The Social Interaction (SOC) Scale includes 12 items, such as showing interest in past activities, trusting family members, engaging in polite social conversation, and having control of emotions most of the time. The Compensates for Incapacities (COM) Scale consists of 19 items, such as asking for and accepting help from others and using auditory or visual aids if necessary. The Behavioral Excesses (BEX) Scales includes 23 items, such as wandering from home, paranoid ideation, and displaying disinhibited sexual or aggressive behaviors. The Behavioral Deficits (BDE) scale includes 6 items, such as discontinuing hobbies, stopping reading, and having word-finding difficulties.

The BCI takes approximately 15 to 20 minutes to complete. The response choices include Yes, No, Don’t Know, or Not Applicable based upon whether the skills and behaviors have been present during the past month. Each item response of Yes is given 1 point toward the BCI total score on 5 of the scales, with a No response yielding 1 point on the Behavioral Excesses and Behavioral Deficits scales. Because the BCI measures the breadth of the functional behavioral repertoire of older adults, higher scores signify broader behavioral competencies credited to the patient.

A previous study of 149 out-patients attests to the reliability of the BCI scales, demonstrating that they showed adequate internal consistencies and represented seven overlapping but distinct constructs (Jarjoura, Hartman-Stein, Speight, & Reuter, 1999).


Clinical Dementia Rating Scale (CDR) (Berg, 1988). The CDR is a widely used method of identifying the presence and severity of impairment of dementia. Based upon a semi-structured interview of the patient and appropriate informant, the clinician rates the patient’s impairment in 6 categories including orientation, memory, judgment and problem-solving, home and hobbies, personal care, and community affairs. The CDR uses a five-point scale including no dementia = 0, questionable =0.5, mild = 1, moderate =2, and severe =3. A recent new version has improved clinical scoring rules for the global CDR (Morris, 1998).

Geriatric Depression Scale (GDS)(Yesavage, et al.1983). The GDS is a self-report measure that examines depressive symptoms in an elderly population. The long form of the questionnaire contains 30 yes/no self-referent statements. The GDS yields reliable and valid information on depressive symptoms in cognitively intact patients (Norris, Gallagher, Wilson, & Winigrad, 1987) and moderate sensitivity and specificity with dementia patients (Lichtenberg, Steiner, Marcopulos, & Tabscott (1992).

Mini-Mental-State Exam (MMSE) (Folstein, Folstein & McHugh, 1975). The MMSE contains eleven items yielding a maximum score of 30 points. It measures orientation, registration and short-term recall of three words, attention and calculation, language, and visual construction. This measure was designed for use with psychiatric inpatients but has been used widely with medically ill patients and community outpatients. It has documented reliability and validity for use in a variety of settings, but it was not designed as a diagnostic tool alone. The original author had cautioned that impairment on the instrument should trigger further evaluation (MacNeill & Lichtenberg, 1999). A score of 24 or lower is generally considered to indicate impairment. The mean MMSE score for this sample was 23.8 (SD =4.4, and the range was 8 to30).


I. Explanatory Multiple Regression

A stepwise multiple regression analysis was performed using the BCI Total score as the criterion variable, with CDR, MMSE, GDS, and age as predictor variables. The CDR score entered on the first step of the analyses, and it accounted for 29% of the BCI Total score variance (see Table 1). The GDS variable entered on the second step of the multiple regression analyses, and it accounted for an additional 12% of the BCI variance. The combination of these two predictors explained 41% of the variance in the BCI score. The MMSE and age variables did not meet entry criteria as established by the stepwise multiple regression and were dropped from the model because they added no predictive value. As shown in Table 2, the MMSE correlates highly (r=-.67, p< .001) with the CDR, and the strong relationship between these two variables accounts for the fact that the MMSE did not enter into the model. Age was also correlated with the BCI Total score (r=-.24, p<.01) as well as with the CDR (r=.24, p<.01) and the MMSE (r=-.18, p<.05). The BCI correlates significantly with the CDR, MMSE, and GDS, demonstrating significant concurrent validity between the BCI and each of these measures.

The two variable model (CDR + GDS) accounts for a significant amount of the BCI Total score variance (F=51.55, p<.0000). However, approximately 59% of the BCI Total score variance is independent of the CDR and GDS, indicating that more than half of the variance in the BCI is independent of the CDR and the GDS, measures that are often included as part of a geriatric screening battery (see Figure 1).


1. Grouping subjects according to diagnosis. A one–way ANOVA using three diagnostic groups was conducted to determine whether behavioral competence, as measured by the BCI differs among patients diagnosed with Dementia only (n=38), Dementia combined with Depression (n=67), and depression without Dementia (n=34). Of those patients diagnosed with Depression, 20 received a primary diagnosis of Depression while 14 received a secondary diagnosis of Depression. Thirteen patients from the larger sample of 152 were omitted from this analysis because they did not have a diagnosis of either depression or dementia. Using the three diagnostic groups as independent variables with BCI Total score as the dependent measure yielded an overall effect for group (F= 3.95, p<.0214). Using a Duncan post hoc test with a significance level of .05 revealed that the Depression only group had a BCI Total score (mean=78.29, SD=14.33) that was significantly higher compared to both the Dementia group (mean= 69.87, SD=15.65) as well as the Dementia plus Depression group (mean =71.21, SD=12.45). The BCI mean scores for the Dementia group compared to the Dementia plus Depression groups did not differ significantly (see Figure 2).


2. Grouping subjects based upon level of impairment. An ANOVA was conducted to determine whether behavioral competence, as measured by the BCI, differs among patients with different levels of dementia. The sample was divided into groups of No Dementia (n=49), Mild Dementia (n=68), and Moderate to Severe Dementia (n=35). The CDR was used to classify subjects by level of impairment. However, the 5 levels (0, .5, 1, 2, and 3) of the CDR were reduced to 3 levels for the purpose of this analysis. The No Dementia group had CDR ratings of 0 and .5, the Mild Dementia group had CDR ratings of 1.0 and the Moderate to Severe Dementia group had CDR ratings of 2.0 and 3.0.


A one-way ANOVA conducted using the 3 levels of impairment as independent variables with BCI total score serving as the dependent measure, yielded a significant effect for group (F= 25.325, p<.0000). Post hoc analyses employing the Duncan test of means with a significance level of .05 revealed that the No Dementia group had a significantly higher BCI score (mean=81.51, SD=12.71) than the Mild Dementia group (mean= 74.29, SD=11.42) and the Moderate to Severe Dementia group (mean=61.86, SD=14.15). The Mild Dementia group received a significantly higher BCI score than the Moderate to Severe Dementia group. Thus, each of the groups differed significantly from each other in the hypothesized order (see Figure 3).


The results of this study show that the Behavioral Competence Inventory provides a caregiver report measure of the functional behavioral repertoire of older adults, demonstrating concurrent validity with the Clinical Dementia Rating Scale and the Geriatric Depression Scale. The Behavioral Competence Inventory also yields scores that differ based upon both the level of impairment and the presence of clinical depression. Patients with depression display a broader repertoire of functional behaviors than patients with dementia. In our sample the presence of concomitant depression with dementia does not produce additional reductions of overall functional behaviors. Older adults with no apparent cognitive impairment demonstrate the broadest repertoire of functional performance.

The comprehensive treatment and care of patients with Alzheimer’s disease and other dementias require that our interventions take into account the patient’s remaining adaptive skills and compensatory abilities. Functional assessment instruments based upon caregiver information such as the Concordant Informant Dementia Scale (Waite, et al., 1998), Alzheimer’s Disease Functional Assessment and Change Scale (Mohs, et al., 2001), and Gottfries-Brane-Steen (Winblad, et al., 2001) are used to assess the degree of functional decline from the patient’s previous performance. Such measures are useful in monitoring the effectiveness of medications for dementia in clinical trials. However, such instruments emphasize only pathology and are limited in their utility to help design behavioral interventions

The Behavioral Competence Inventory can serve as a broad- spectrum measure of behavioral competence that provides additional information beyond that offered by ratings of level of cognitive impairment and decline in functioning. Our results suggest that it is useful as an aid to diagnosis, provides an efficient method for obtaining information from caregivers, has promise as a longitudinal measure of the rate of functional decline in patients at risk for dementia, and can serve as a guide for clinical interventions with older adults. Because of its ease in scoring and administration, it can be an efficient addition to a geriatric screening battery.

Note: In a preliminary set of analyses included in the conference abstract submission, the above analyses were run on a subsample of 101 older adults. We later expanded the sample to 152 and our final results differ slightly due to the addition of these subjects. However, these changes did not impact our final conclusions.


Alexopoulos, G.S., Abrams, R.C., Young, R.C., & Shamoian, C.A. (1998). Cornell Scale for Depression in Dementia. Biological Psychiatry, 23, 271-84.

American Psychiatric Association (1994). Diagnostic and Statistical manual of Mental Disorders, Fourth Edition. Washington, D.C.

American Psychological Association, Presidential Task Force on the Assessment of Age-Consistent Memory Decline and Dementia (1998). Guidelines for the evaluation of dementia and age-related cognitive decline. Washington, DC: American Psychological Association.

Benton, A.L., Hamsher, K. de S., Rey, G.J., & Sivan, A.B. (1994). Multilingual Aphasia Examination (3rd Ed). Iowa City, IA: AJA Associates.

Berg, L. (1998). Clinical Dementia Rating (CDR). Psychopharmacolgy Bulletin, 24 (4), 637-639.

Folstein, M.F., Folstein, S.E., & McHugh, P.R. (1975). Mini-Mental State: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12, 189-198.

Hartman-Stein, P.E., & Reuter, J.M. (1996). The Behavioral Competence Inventory, Second edition. Kent, OH: Center for Healthy Aging.

Jarjoura, D., Hartman-Stein, P., Speight, J., & Reuter, J. (1999). Reliability and construct validity of scores on the Behavioral Competence Inventory: A measure of adaptive functioning. Educational and Psychological Measurement, 59, 855-865.

Kaplan, E.F., Goodglass, H., & Weintraub, S. (1983). The Boston Naming Test. (2nd ed.) Philadelphia: Lea & Febiger.

Lichtenberg, P.A., Steiner, D.A., Marcopulos, B.A., & Tabscott, J.A. (1992). Comparison of the Hamilton Depression Rating Scale and the Geriatric Depression Scale: Detection of depression in dementia patients. Psychological Reports, 70, 515-521.

Loeb, P.A. (1996). The Independent Living Scales, New York: Psychological Corporation.

Mattis, S. (1988). Dementia Rating Scale: Professional manual. Odessa, FL: Psychological Assessment Resources.

MacNeill, S.E. & Lichtenberg, P.A. (1999). Screening instruments and brief batteries for Assessment of Dementia. In P.A. Lichtenberg (Ed.) Handbook of Assessment in Clinical Gerontology (pp417-441) New York: John Wiley & Sons, Inc.

Marcotte, T.D., van Gorp, W., Hinkin, C.H., & Osato, S. (1997). Concurrent Validity of the Neurobehavioral Cognitive Status Exam subtests. Journal of Clinical and Experimental Neuropsychology, 19, 386-395.

Mohs R.C., Doody, R.S., Morris, J.C., Ieni, J.R., Rogers, S.L., Perdomo, C.A., & Pratt, R.D. (2001). A 1- year, placebo-controlled preservation of function survival study of donepezil in AD patients. Neurology 57, 481-488.

Norris, J.T., Gallagher, E., Wilson D., & Winigrad, C.H (1987) Assessment of depression in geriatric medical out-patients: the validity of two screening measures. Journal of American Geriatric Society, 35: 989-995.

Osato, S., Yang, J., & LaRue, A. (1993). The Neurobehavioral Cognitive Status Examination: An exploratory study with older psychiatric inpatients. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 1, 221-230.

Reitan, R. M. (1958). Validity of the Trail Making Test as an indicator of organic brain damage. Perceptual and Motor Skills, 8, 271-276.

Waite, L.M., Broe, G.A., Casey, B., Bennett, H.P., Jorm, A.F., Creasey, H., Cullen, J., & Grayson, D.A. (1998). Screening for dementia using an informant interview. Aging, Neuropsychology, and Cognition, 5, 194-202

Weschler, D. (1981). Adult Intelligence Scale-Revised. San Antonio: Psychological Corporation.

Weschler, D. (1987). Weschler Memory Scale-Revised. San Antonio: Psychological Corporation.

Winblad, B., Engedal, K, Soininen, H., Verhey, F., Waldemar, G., Wimo, A., Wetterholm, A.-L., Zhang, R., Haglund, A., Subbiah, P., & the Donepezil Nordic Study Group (2001). A 1-year, randomized, placebo-controlled study of donepezil in patients with mild to moderate AD. Neurology 57, 489-495.

Yesavage, J.A., Brink, T. L., & Rose, T. L., Lum, O., Huang, O., Adey, V., & Leirer, V. (1983). Development and validation of a geriatric depression scale: A preliminary report. Journal of Psychiatric Research, 7, 7-49.