Outcome Summaries of Latency-Based Functional Analyses Conducted in Inpatient Units of Hospital Settings

Background:  

Latency-based functional analysis (LBFA) may be a viable alternative to traditional Functional Analysis (FA) when evoking and reinforcing high rates of problem behavior is not advisable, or when client time, appropriate and safe assessment space, and/or staffing are limited. In practice, this means that the conditions under which families reported seeing problem behavior were systematically presented for the participant and that data collectors measured how soon problem behaviors occurred when different conditions and potential reinforcers were applied. Comparing this latency to problem behavior across different conditions is a valid way to confirm the functional reinforcer(s) for a behavior. Embedded within a randomized controlled trial (RCT) to assess for potential cost-benefit of behavior analytic services within typical inpatient hospital settings, we conducted LBFAs of the problem behavior of 18 children diagnosed with autism in inpatient hospital settings.  

Objectives:  

The purpose of this study was to provide preliminary descriptive evidence of the capability of LBFAs to identify functional reinforcers for problem behavior when conducted within typical inpatient hospital settings. 

Methods:  

Eighteen children (17 male, 1 female) with autism ages 6-16 years (mean 10.3) participated in this study.  Twelve had been admitted to a university-based child and adolescent psychiatric hospital and six had been admitted to a medical floor of a university-based children’s hospital. Additionally, all had a reported history of chronic problem behavior.  We conducted all assessments in inpatient units of the above-mentioned hospital settings; either in subjects’ rooms or in multi-purpose rooms found within these units. 

Definitions for each subject’s target response(s) were operationalized based on caregiver report during pre-FA interviews (definitions available upon request). A preference assessment was conducted with the patient, followed by initiation of a latency-based FA.  Assessment procedures were based on those described by Thomasson-Sassi et al. (2011). 

Therapists also tracked latency to occurrence of non-targeted problem behavior during FA trials when subjects engaged in multiple topographies of problem behavior.  Because latencies for these topographies were obtained within the context of highly controlled antecedent manipulations, we conceptualized data obtained from this secondary analysis as a structured descriptive assessment (SDA; Freeman, Anderson, & Scotti, 2000), with latency to unconsequated problem behavior as its dependent variable.  

Results:  

Eighteen latency-based FAs were conducted in inpatient hospital settings to identify the function of problem behavior of 15 subjects diagnosed with ASD.  Eight latency-based FAs successfully identified the functions of eight response topographies.  Conversely, eight latency-based FAs produced inconclusive results. Four latency-based SDAs were simultaneously conducted on the secondary responses of four subjects. Differentiated response patterns emerged for two patients. 

Conclusions:  

Despite some limitations, which were specific and unique to the study settings and therapeutic contexts, results indicate it is possible to identify behavior function through latency-based FAs conducted by novel therapists in inpatient hospital units after evoking and reinforcing a minimal number of target responses. Furthermore, we outlined a latency-based strategy for collecting descriptive data on non-targeted problem behavior during FA sessions.