Variability in staff reports of the frequency of challenging behavior

https://doi.org/10.1016/S0891-4222(01)00069-5Get rights and content

Abstract

Information was collected from 88 staff about their observations of the occurrence and frequency of challenging behaviors in 22 individuals with developmental disabilities with whom they worked. Staff reports suggested considerable variability in challenging behavior in the presence of different staff and, from day to day, in the presence of the same staff. Variability was greater for stereotyped than for aggressive/destructive behavior, and for more frequent behavior. Managers reported generally less challenging behavior than their staff. The validity of the findings was discussed and their implications for research and practice considered.

Introduction

Natural variability in the occurrence of challenging behavior is more common than not. Such variability has been explained as a result of variations in the occurrence of setting events Carr et al 1996, Gardner et al 1986, antecedent events that evoke challenging behavior McGill 1999, Smith and Iwata 1997, the application of reinforcement contingencies (Iwata et al., 1994b), and biological factors (Carr, 1994). Such variability has, therefore, been of particular interest to those responsible for assessing challenging behavior, and this process of assessment can essentially be seen as one of hypothesizing the factors responsible for such variability either indirectly through questionnaires (e.g., Durand and Crimmins 1988, Matson et al 1999) or interviews (e.g., O’Neill, 1997), or more directly through observation (e.g., Emerson, Thompson, Reeves, Henderson, & Robertson, 1995). The most complex approaches to such assessment have relied not on naturally occurring variability but on the imposed variability (Sidman, 1960) that arises when an individual is exposed to specially designed settings intended as analogs to the natural environment (Iwata, Dorsey, Slifer, Bauman, & Richman, 1994a).

This pragmatic approach to variability may have diverted attention from the phenomenon itself. Thus, in general, we can conclude very little about the range of variability that is likely to be found in an individual’s behavior, about the extent to which variability can be uniformly expected across individuals, or about its relationship with the topography of challenging behavior. Such knowledge (which, in part, might be derived from a re-analysis of previously published studies containing single case data) is important for a number of reasons. First, an interest in variability naturally leads to an interest in why challenging behavior sometimes occurs at high rates/intensities and sometimes at low or zero rates/intensities, and to the investigation of factors correlated with this variation. Of course, a good functional assessment might be expected to provide this information in respect of an individual, but cannot provide this information across individuals. Knowledge about the variability of the challenging behavior of many individuals is required before we can identify correlated factors of potentially general relevance (cf., Iwata et al., 1994b). Second, information about variability may help to promote a parametric analysis of its correlated factors, drawing attention to those factors that should receive most effort both in intervening with individuals and larger-scale prevention. Third, information about variability may direct attention to the development of interventions that seek to increase the occurrence of the factors associated with low rates of challenging behavior. Such approaches have been described only rarely in the literature (e.g., Touchette, MacDonald, & Langer, 1985) but may avoid some of the implementation (Hastings & Remington, 1993), maintenance (Ager, 1991) and generalization (Stokes & Baer, 1977) difficulties frequently reported with more traditional behavioral interventions. Fourth, information about variability may promote a view of challenging behavior as partly under the control of the services provided to the person (Emerson, McGill, & Mansell, 1994). That is, the demonstration of variability may constitute a demonstration of both the service’s success (at times of low or zero occurrence) and the service’s failure (at times of high occurrence) that directs attention to the scope for, and the nature of, improvements in the service’s performance (cf., Gilbert, 1978).

One particular location of variability may be the people supporting or caring for the individual. Wing and Gould (1978) reported that parents described their children with developmental disabilities/autism as displaying more stereotyped and aggressive or difficult behavior than the children’s teachers or child care workers. They concluded that there were real differences in the children’s behavior in the presence of different informants. Holmes, Shah and Wing (1982), in a study of residents in a developmental disability hospital, found similar differences in the reports of nurses and day care workers and reached a similar conclusion, also noting that genuine differences in individual behavior could be found within the same setting. Differences within the same setting were, indeed, reported by the multiple raters used by Rojahn (1984) in his study of self-injurious behavior in an institution and by Touchette, MacDonald and Langer (1985), whose data included an individual recorded as displaying self-injury at high rates only when one member of staff was on duty. Touchette et al’s study included a reversal in which staff changed their shifts, with the rate of self-injury changing in parallel. None of these studies have set out to specifically investigate variability in challenging behavior so that the generality of conclusions drawn from them may be questioned. The phenomenon they have identified, however, is not specific to developmental disability (see, for example, Graham & Rutter, 1970) and is consistent with the established notion of situation specificity. This literature provides grounds, therefore, for supposing that challenging behavior may commonly vary across the presence of different staff or caregivers.

In the study described here, variability in the reported occurrence and frequency of challenging behavior in the presence of different staff was investigated directly, with a view to estimating the extent of variability, the extent to which variability can be uniformly expected across individuals, and its relationship with the topography of challenging behavior. Additionally, relationships with the status, role and gender of reporters were investigated.

Section snippets

Settings

Data were gathered in 18 residential service settings, 13 provided by private agencies, 3 by voluntary organizations, 1 by the NHS and 1 by a Local Education Authority (a residential school). Settings varied widely in size having on average 10 clients (range: 2–27) and 14 staff (range: 5–26). Some settings were parts of larger units (e.g., a bungalow in a village community) while some were completely independent of other services.

Participants

Data were gathered on 22 people with developmental

Results

On the 88 completed BPIs respondents reported an average of 10.3 (range: 1–22) problem behavior topographies including 4.0 self-injurious behaviors (range: 0–12), 4.4 aggressive/destructive behaviors (range: 0–9) and 1.8 stereotyped behaviors (range: 0–5). Frequency ratings were converted to numerical scores for each item (from 0 = “never” to 6 = “more than hourly”). The average total frequency score was 31.8 (range: 4–89), 11.6 for self-injury (range: 0–42), 12.6 for aggressive/destructive

Discussion

Conventionally, differences between staff in their reports of challenging behavior have been treated as a matter of reliability. The central proposition of this paper has been that such differences may reflect valid (rather than unreliable) differences in staff experience. The main findings of the study can be summarized as follows. Staff reports of the occurrence and frequency of challenging behaviors displayed by people with developmental disabilities were highly correlated but widely

References (34)

  • E. Emerson et al.

    Descriptive analysis of multiple response topographies of challenging behavior across two settings

    Research in Developmental Disabilities

    (1995)
  • J.L. Matson et al.

    A validity study on the Questions About Behavioral Function (QABF) scalePredicting treatment success for self-injury, aggression, and stereotypies

    Research in Developmental Disabilities

    (1999)
  • A. Ager

    Effecting sustainable change in client behaviourThe role of the behavioural analysis of service environments

  • L.D. Burgio et al.

    A descriptive analysis of nursing staff behaviors in a teaching nursing homeDifferences among NAs, LPNs, and RNs

    The Gerontologist

    (1990)
  • E.G. Carr

    Emerging themes in the functional analysis of problem behavior

    Journal of Applied Behavior Analysis

    (1994)
  • Carr, E. G., Reeve, C. E., & Magito-McLaughlin, D. (1996). Contextual influences on problem behavior in people with...
  • J. Cohen

    A coefficient of agreement for nominal scales

    Educational and Psychological Measurement

    (1960)
  • J. Cohen

    Weighted kappaNominal scale agreement with provision for scaled disagreement or partial credit

    Psychological Bulletin

    (1968)
  • J.R. Dura et al.

    Prevalence of stereotypy among institutionalized nonambulatory profoundly mentally retarded people

    American Journal of Mental Deficiency

    (1987)
  • V.M. Durand et al.

    Identifying the variables maintaining self-injurious behavior

    Journal of Autism and Developmental Disorders

    (1988)
  • W.I. Gardner et al.

    Reducing aggression in individuals with developmental disabilitiesAn expanded stimulus control, assessment, and intervention model

    Education and Training in Mental Retardation

    (1986)
  • T.F. Gilbert

    Human competenceEngineering worthy performance

    (1978)
  • P. Graham et al.

    The selection of children with psychiatric disorder

  • R. Hastings et al.

    “Is there anything on…why ‘good’ behavioural programmes fail?” A brief review

    Clinical Psychology Forum

    (1993)
  • S. Holburn

    A renaissance in residential behavior analysis? A historical perspective and a better way to help people with challenging behavior

    The Behavior Analyst

    (1997)
  • N. Holmes et al.

    The Disability Assessment ScheduleA brief screening device for use with the mentally retarded

    Psychological Medicine

    (1982)
  • Cited by (13)

    View all citing articles on Scopus

    This research was funded by a grant from the NHS National Research and Development Programme for People with Physical and Complex Disabilities. We are very grateful to the people with developmental disabilities, staff members and services that gave their consent and cooperation for the conducting of this study. David Hughes is now at Cumbria Care Services. Kerry Teer is now at Department of Psychological Medicine, University of Glasgow. Lynne Rye is now at East Kent Community NHS Trust and the Tizard Centre.

    View full text