SEVERE BEHAVIOR PROBLEMS Summary

713

JOURNAL OF APPLIED BEHAVIOR ANALYSIS 1997, 30, 713–716 NUMBER 4 (WINTER 1997)

ASSESSMENT-BASED INTERVENTION FOR SEVERE BEHAVIOR PROBLEMS IN

A NATURAL FAMILY CONTEXT

BOBBIE J. VAUGHN, SHELLEY CLARKE, AND GLEN DUNLAP UNIVERSITY OF SOUTH FLORIDA

Functional assessments and assessment-based interventions were conducted with a boy with disabilities and severe problem behavior in the context of two family routines: using the bathroom in the family home and dining in a fast-food restaurant. A multiple baseline design demonstrated the effectiveness of the intervention package as implemented by the boy’s mother in the two routines. The results provide a systematic replication and exten- sion of behavior-analytic interventions in natural family contexts.

DESCRIPTORS: family support, problem behaviors, functional assessment, assess- ment-based intervention

Many of the responsibilities of family life include conducting typical family routines in the home and community; however, these routines can be very difficult for families that include children with disabilities and severe problem behavior. Recently, behavior-analyt- ic research has produced encouraging find- ings through its development and evaluation of assessment-based and contextually appro- priate interventions in school and commu- nity environments (Carr & Carlson, 1993; Dunlap, Kern-Dunlap, Clarke, & Robbins, 1991). Such studies have used interview and observational data to develop multicompo- nent interventions for resolving serious be- havioral challenges. Still, despite its crucial

This research was supported by Field-Initiated Re- search Grant H133G60119 and Cooperative Agree- ment H133B2004 from the National Institute on Dis- ability and Rehabilitation Research (U.S. Department of Education) and by the Louis de la Parte Florida Mental Health Institute of the University of South Florida. However, no official endorsement by any sup- porting agency should be inferred. Appreciation is ex- tended to all members of the participating family, and to Lilliane Reyes and Arcadia Vera for assistance with data collection.

Reprint requests may be sent to Glen Dunlap, Di- vision of Applied Research and Educational Support, Department of Child and Family Studies, Louis de la Parte Florida Mental Health Institute, University of South Florida, 13301 Bruce B. Downs Blvd., Tampa, Florida 33612.

importance for the external validity of be- havior-analytic procedures, the application of assessment-based interventions by typical care providers (e.g., parents) in naturally oc- curring situations (e.g., family routines) has received little attention in the research lit- erature (Lucyshyn, Albin, & Nixon, 1997). Therefore, this controlled case study was conducted with the purpose of systematically replicating and extending the literature on assessment-based intervention in typical family contexts.

METHOD

Andrew was an 8-year-old boy with severe disabilities including agenesis of the corpus callosum. He and his mother participated in this study. Although Andrew could speak, his communication was supported by aug- mentative pictures and objects or was con- veyed through gestures or problem behav- iors. Andrew’s mother was a homemaker in her early 30s who attended a community college on a part-time basis.

The study was conducted in two contexts that were selected by Andrew’s mother be- cause they involved typical routines that were associated with significant behavior problems. The first context was the bathroom in the

714 BOBBIE J. VAUGHN et al.

home: Andrew was expected to follow his mother’s request and move from a living area to the bathroom, turn on the light, urinate in the toilet, flush, turn off the light, and exit the bathroom. The second context was a fast- food restaurant: Andrew was expected to exit the car with his mother and sister, enter the restaurant, sit and wait with his sister while his mother ordered the food at the counter, eat the food, and then leave the restaurant.

Sessions were held two times per week and were defined by predetermined initia- tion and termination points that were held constant throughout the study. All sessions were videotaped using a hand-held camcor- der and scored in continuous 10-s intervals. Dependent variables included disruptive be- havior and engagement. Disruptive behavior consisted of aggression, property destruc- tion, whining, collapsing on the floor, and attempting to run away, and was scored if any instance occurred within an interval. En- gagement was defined as following the nat- ural sequence of the routine or specific task instructions and was scored if Andrew was on task for at least 70% of the interval, as determined by observer judgment. It was possible for both dependent variables to be scored within a single interval. In addition, Andrew’s task performance on each step of the routines was scored from the videotapes on a 3-point scale that included nonperform- ance (1), performed with assistance (2), and performed independently (3). Interobserver agreement was determined for 48% of ses- sions. Agreement averaged 93% (range, 59% to 100%) for occurrences of disruptive be- havior and 95% (range, 50% to 100%) for intervals with engagement. Agreement on the task routines was 100% for the bath- room and 98% for the restaurant.

The independent variable in this study was the process of intervention development, which began with a functional assessment (O’Neill et al., 1997) that included a detailed interview with Andrew’s mother and several

observations by the first two authors in the two routines. The assessment data led to the development of hypotheses regarding the oc- currence of disruptions in the two routines. Disruptions during toileting appeared to oc- cur because the routine required transition to a relatively unreinforcing activity, and An- drew seemed to have difficulty following the sequence independently. In response, the in- tervention involved a visual portrayal (sched- ule) of the activity sequence that included a picture of a preferred toy that was delivered following successful completion of the toilet- ing routine. Similarly, disruptions during the restaurant routine were hypothesized to be a function of inadequate reinforcement during the periods of waiting (e.g., ordering) and an inability to participate. The intervention in- cluded an adapted menu to facilitate An- drew’s participation in ordering, active en- couragement of Andrew’s involvement in paying for food and securing soft drinks, and a photograph of a preferred toy (reward) that was provided during the car ride home. Thus, although the routines differed along many di- mensions, both hypothesis-based interven- tion packages were comprised of supplemen- tary rewards and visual schedules to facilitate participatory engagement, implemented in a multiple baseline across settings.

Andrew’s mother implemented all of the in- terventions during the sessions, with one of the authors providing suggestions and encour- agement during the first few intervention ses- sions. All subsequent sessions, including all of the weekly follow-up sessions, were conducted independently by Andrew’s mother. A probe session was conducted during follow-up for the restaurant routine. In this session, An- drew’s father was present and a restaurant was selected in which ordering was issued to a server (rather than ordering at a counter). To expedite the waiting period, Andrew was pro- vided with drawing materials (a common practice in many family restaurants). All other procedures were the same.

715FAMILY-CENTERED INTERVENTION

Figure 1. Percentage of intervals with disruptive behavior and engagement across the bathroom and restau- rant routines. Connected data points represent sessions that occurred approximately two times per week. Dis- connected points span a period of at least 1 week. The square data points during follow-up in the restaurant routine depict a probe session that was conducted in a different restaurant environment.

RESULTS AND DISCUSSION

Figure 1 depicts a clear demonstration of the effectiveness of assessment-based inter- vention packages. The interventions pro- duced rapid and durable reductions in dis- ruptive behavior and corresponding increases in engagement. The one-session probe (the square data points) in the family restaurant also yielded desirable behavior. In terms of Andrew’s task performances, scores for the bathroom routine averaged 1.77 during baseline and increased to scores of 2.8 in both intervention and follow-up; the restau- rant routine yielded average scores of 2.5 in baseline, 2.9 in intervention, and 3.0 in fol- low-up.

These results contribute to an emerging lit- erature on assessment-based behavioral inter- vention in typical settings and routines. In re- cent years, there have been a number of em- pirical demonstrations of a functional assess- ment process that leads to multicomponent interventions in schools and community en- vironments (e.g., Carr & Carlson, 1993; Dunlap et al., 1991); however, there have been very few such investigations in family contexts, despite the fact that family routines are arguably the most common and most in- fluential interactions for children with disabil- ities. The present study contributes a con- trolled case study with a mother implement- ing all of the procedures in circumstances that the family identified as the most problematic.

716 BOBBIE J. VAUGHN et al.

It is important to note that the contexts in- cluded a private routine in the home (the bathroom) as well as a routine in a highly public setting (the fast-food restaurant). In both settings, problem behavior decreased while engagement and independent task com- pletion increased.

It is important to acknowledge that the study is limited by having only one partici- pating family and only two routines. Further, despite the presence of several weeks of follow- up data, the scope of the investigation was inadequate to claim the important benefit of long-lasting lifestyle change. Nevertheless, when combined with the accumulating base of information on assessment-based family support (Lucyshyn et al., 1997), the present data can be viewed as adding encouraging tes- timony to the efficacy and feasibility of func- tional, assessment-based, and contextually ap- propriate interventions.

REFERENCES

Carr, E. G., & Carlson, J. I. (1993). Reduction of severe behavior problems in the community using a multicomponent treatment approach. Journal of Applied Behavior Analysis, 26, 157–172.

Dunlap, G., Kern-Dunlap, L., Clarke, S., & Robbins, F. R. (1991). Functional assessment, curricular revision, and severe behavior problems. Journal of Applied Behavior Analysis, 24, 387–397.

Lucyshyn, J. M., Albin, R. W., & Nixon, C. D. (1997). Embedding comprehensive behavioral support in family ecology: An experimental, sin- gle-case analysis. Journal of Consulting and Clinical Psycholog y, 65, 241–251.

O’Neill, R. E., Horner, R. H., Albin, R. W., Storey, K., Sprague, J. R., & Newton, J. S. (1997). Func- tional assessment and program development for prob- lem behavior: A practical handbook. Pacific Grove, CA: Brooks/Cole.

Received March 31, 1997 Initial editorial decision May 21, 1997 Final acceptance July 23, 1997 Action Editor, Craig H. Kennedy