COMBI >> Scales >> ABS >> Syllabus


Jennifer Bogner, Ph.D.
Ohio Regional TBI Model System, at

Email address protected by JavaScript.
Please enable JavaScript to use email address.




Bogner, J. (2000). The Agitated Behavior Scale. The Center for Outcome Measurement in Brain Injury.
combi/coglog ( accessed ).





ABS Syllabus


Agitation during the acute phase of recovery from brain injury is often observed, usually treated, but rarely measured. Denny-Brown (1945) first observed that excited and restless behavior may be natural sequela of recovery from traumatic brain injury. The term "agitation" was used to define Level IV of the Rancho Los Amigos Levels of Cognitive Functioning Scale. This "confused-agitated" stage was described as a time when a patient is in a heightened state of activity and has diminished capability for processing new information and responding to events in the environment (Malkmus, Booth & Kodimer, 1980). Levin and Grossman (1978) found that approximately one-third of patients with closed head injuries on a neurosurgery service experienced agitation marked by disinhibited movement, restlessness, thrashing, and aggressiveness. Brooke, Questad, Patterson, and Bashak (1992) conducted a prospective study of 100 patients with closed head injuries admitted for acute hospitalization and found 11% showed marked aggression and 35% restlessness. Reyes, Bhattacharyya and Heller (1981) studied patients admitted to an acute rehabilitation unit and found that more than half exhibited restless and agitated behavior. At Ohio State University (Bogner & Corrigan, 1995), data compiled from a prospective sample of 100 consecutive admissions to the rehabilitation unit indicated that 42% of the sample exhibited agitated behavior.

Because of its disruption of therapeutic goals, significant agitation during rehabilitation has to be addressed, and various behavioral, environmental and pharmacological interventions have been used. A survey conducted by Herbel, Schermerhorn and Howard (1990) found that facilities treating agitated patients use a variety of techniques, with most having options for one-to-one supervision, behavior modification, physical restraints, environmental modifications and medication. With regard to the latter, the research literature is sprinkled with case studies of the effectiveness of various medications in improving agitation. While there is not clinical consensus about which medications are effective in what circumstances (Fugate, Spacey, & Kresty, 1997), there is an understanding that those pharmacologic interventions which reduce agitation through sedation can delay, if not prevent, patients' cognitive and functional improvement during the acute phase of recovery (Mysiw & Sandel, 1997). Empirical support for this contention was provided by Corrigan and Mysiw (1988) who found that improvement in cognition was a prerequisite to improved agitation for patients who demonstrated both.

Despite the documented frequency of agitation in patients with traumatic brain injuries, as well as the need to treat this condition, until recently there has not been an objective measure of agitation. The Agitated Behavior Scale (ABS; Corrigan, 1989) was developed to allow objective assessment of this behavior, particularly serial assessments for the evaluation of interventions to reduce agitation. To construct the ABS, a preliminary pool of 39 items was generated from review of the literature and use of Kelly's (1955) construct elicitation methodology with interdisciplinary staff experienced in brain injury rehabilitation. In a pilot study, 14 items were selected from this pool based on degree of inter-rater reliability, differentiation of agitation, frequency of occurrence, and representation of factors present in the original item pool. The final 14-item scale was then validated on an independent sample and was found to have appropriate levels of inter-rater reliability, internal consistency and concurrent validity (Corrigan, 1989).

Subsequent studies have provided support for the construct validity of the ABS (Corrigan & Mysiw, 1988; Corrigan, Mysiw, Gribble & Chock, 1992; Corrigan, Bogner, & Tabloski, 1996; Novack & Penrod, 1993; Tabloski, McKinnon-Howe, & Remington, 1995). Corrigan and Bogner (1994) reported additional evidence of construct validity via delineation of the factor structure of the ABS. In their sample of 212 patients with recently acquired brain injury who exhibited agitation during acute rehabilitation, confirmatory factor analyses revealed that agitation is represented by one general construct with three underlying, correlated factors: Aggression, Disinhibition, and Lability. They concluded that the Total Score remains the best measure of agitation, but subscale scores may provide important additional clinical and research data.

Based on studies investigating the construct validity of the Agitated Behavior Scale, as well findings in the research literature regarding traumatic brain injury and other conditions which have agitation as a potential sequela, the following definition of agitation has been proposed:

  • Agitation is an excess of one or more behaviors that occurs during an altered state of consciousness (Bogner & Corrigan, 1995).

This definition emphasizes the importance of "excessiveness" over the type of behavior manifested. "Excessiveness" has been operationalized as the degree to which the behavior interferes with functional activities and the extent to which the behavior can be inhibited. As suggested by the factor analysis, no one type of behavior defines agitation, though some component behaviors may be more dominant at times. The definition also requires that the behavior occur during an altered state of consciousness which, for traumatic brain injury, includes the diminished arousal present from time of injury through the clearing of post-traumatic amnesia. This definition is reflected in the criteria used to rate the behaviors which comprise the ABS.



A rating of "1" is ascribed when the behavior in the item is not present. Ratings of "2," "3," and "4" indicate the behavior is present and differentiate the degree or severity. Degree can be a function of either the frequency with which the behavior occurs or the intensity of individual occurrences.

Raters should be instructed that the basis for establishing the degree is the extent to which the occurrence of the behavior described in the item interferes with functional behavior that would be appropriate to the situation:

  • We suggest a rating of "2" or "slight" be ascribed when the behavior is present but does not prevent the conduct of other, contextually appropriate behavior. Patients may redirect themselves spontaneously or the continuation of the agitated behavior does not preclude the conduct of the appropriate behavior.
  • A rating of "3" or "moderate" indicates the individual may need to be redirected from an agitated to an appropriate behavior, but is able to benefit from such cueing.
  • A rating of "4" or "extreme" is ascribed when the individual is not able to engage in appropriate behavior due to the interference of the agitated behavior, even when external cueing or redirection is provided.



The Total Score is calculated by adding the ratings (from one to four) on each of the fourteen items. Raters are instructed to leave no blanks; but, if a blank is left, the average rating for the other fourteen items should be inserted such that the Total Score reflects the appropriate possible range of values. The Total Score is the best overall measure of the course of agitation (Corrigan, 1989; Corrigan & Bogner, 1994).

Subscale scores are calculated by adding ratings from the component items:

  • Disinhibition is the sum of items 1, 2, 3, 6, 7, 8, 9 and 10.
  • Aggression is the sum of items 3, 4, 5 and 14. (It is not an error that Item #3 is in both scores.)
  • Lability is the sum of items 11, 12 and 13.

In order to allow subscale scores to be compared to each other and to the Total Score, it is recommended that an average item score for each factor be calculated and multiplied by fourteen. This procedure provides subscale scores with the same range as each other and the Total Score, which is useful for graphic presentation.




Copyright © 1998-2012
Home | Background | Scales | Survey | Newsletter


NIDRR Logo A project funded by the National Institute on Disability and Rehabilitation Research.