• E-mail
• Print
• Comment
• Font Size
• Digg
• del.icio.us
• Discuss article

Effects of a topical anesthetic on non-socially maintained self- injurious behavior

Posted on: Thursday, 27 November 2003, 06:00 CST

In the present study we hypothesized that the self-injurious behavior (SIB) of a 13-year-old male diagnosed as having autism and severe mental retardation* was maintained by automatic reinforcement. Further, we hypothesized SIB served a sensory reinforcement function related to sensation at the skin surface. The participant engaged in high rates of SIB in the form of self- slapping. A functional analysis resulted in undifferentiated findings. Additional observations indicated that SIB occurred only to exposed skin surfaces. To test for a sensory function, we applied a topical anesthetic to the participant's skin. SIB was reduced by 20 to 60% after application of the dermal anesthetic. These results support an automatic reinforcement hypothesis. Implications for treatment are discussed.

A small but growing literature has demonstrated that some behaviors produce direct stimulation independently of environmental influences. Such behaviors are classified as automatic reinforcement (Vaughn and Michael 1982). Experimentally confirming the automaticity of reinforcement may be difficult because the source of reinforcement typically is not subject to direct manipulation. However, assessment strategies have been used to infer an automatic function by the following: (1) ruling out social sources of reinforcement, (2) demonstrating reductions in behavior when hypothesized sources of automatic reinforcement are blocked, (3) offering alternative sensory activities that compete with reinforcement provided by the response, or (4) demonstrating persistence in the absence of all social stimulation.

For example, a functional analysis (Iwata et al. 1994) can be used to rule out social sources of reinforcement. Specifically, an undifferentiated functional analysis suggests that target behavior may be maintained by consequences produced by the response itself. That is, when elevated rates of a target behavior occur across all environmental situations, including conditions under which social reinforcement is absent (namely alone), a parsimonious conclusion is automatic reinforcement.

In addition, research has demonstrated reductions in a response by blocking reinforcement provided by that response, further suggesting automatic reinforcement. For example, Kennedy and Souza (1995) conducted a functional analysis of a young male's eye- poking. The analysis indicated that eye-poking was not socially maintained. Further analyses indicated that eye-poking was reduced to nearly zero rates when the participant wore safety goggles. This analysis implicated finger-eye contact or its sensory consequence as a source of reinforcement.

Research has also demonstrated that offering competing sensory activities can result in reductions in responses presumed to be automatically reinforced. For example, Kennedy and Sousa (1995) conducted a third analysis in the study described above. Eye-poking was reduced substantially when the participant was provided with a hand-held video game. This sensory stimulating activity effectively competed with eye-poking, further supporting the hypothesis that eye- poking was maintained by sensory consequences.

In the present study, we hypothesized that SIB was maintained by automatic reinforcement. Anecdotal information and direct observation suggested that the response-produced sensory stimulation related to access to the skin surface. For example, SIB directed to the face was eliminated by using a helmet with a face shield. We tested an automatic reinforcement hypothesis by blocking skin sensation through use of a topical anesthetic applied to the skin.

Method

PARTICIPANT AND SETTING

The participant was a 12-year-old male diagnosed as having autism and severe mental retardation. This diagnosis was made by multiple examiners in different clinical settings and was based on historical information, observation, and administration of the Childhood Autism Rating Scale (Shopler et al. 1980). Extensive chromosome, genetic, and metabolic workups did not identify any specific syndrome.

The participant was unable to obtain baseline measures on standard IQ testing. Developmental testing was consistent with a functional developmental level of less than 2 years of age. He was nonverbal, but communicated using approximately 50 keys on a touch talker. He engaged in high rates of self-slapping. All sessions were conducted in an outpatient therapy room located in a university- affiliated hospital. The participant took 1.1mg/kg/day of naltrexone throughout the study. Parental informed consent was provided before beginning the study.

MEASUREMENT, DESIGN, AND INTEROBSERVER AGREEMENT

SIB, in the form of face slapping, served as the dependent measure. SIB was defined as sharp and forceful contact between the hand and face. SIB was measured as a frequency count by using laptop computers. A quasi-experimental design was used to evaluate the effects of the intervention on SIB. An A-B design was used across two evaluations and was counterbalanced (B-A) during the last evaluation, for a total of three evaluations in each treatment condition.

Interobserver agreement was assessed during 33% of sessions. One individual collecting interobserver agreement data was blind to the purpose of the study and the experimental conditions. Agreement was calculated with the Reliable programme (Repp et al. 1989), which uses an algorithm to assess agreement between two data sets on a point-by-point basis. A 5-second window was allotted for agreement between observers. Mean interobserver agreement was 93% (range 91 to 95%).

PROCEDURES

Before conducting the current study, the participant had been hospitalized for assessment and treatment of SIB. To evaluate whether SIB was maintained by social (namely external) reinforcement, a functional analysis (Iwata et al. 1994) was conducted during his in-patient stay. During the analysis, sessions were conducted in which reinforcement in the form of escape from task, access to tangible items, and attention was provided contingent on SIB. The frequency of SIB during these sessions was compared with a control session. The results of the analysis indicated that in spite of the presence or absence of various reinforcement contingencies, high and consistent frequencies of SIB were observed. Further, extended (45-minute) sessions alone, in which the participant was alone in a room and observed through a one- way window, indicated stable rates of SIB across each analysis. These analyses combined suggested that SIB was not maintained by social consequences.

Figure 1: Number of self-injurious responses per minute after no application of anesthetic and after application of anesthetic.

An additional assessment was conducted to determine the specific topography (namely location) of SIB. Parental report indicated that the highest rates of SIB occurred during showering, when all skin surfaces were exposed. This was systematically evaluated by covering and exposing various skin surfaces (helmet/no helmet, long-sleeved/ short-sleeved shirt, long/short pants). Direct observations confirmed that SIB occurred almost exclusively to exposed skin surfaces. When skin surfaces were covered, SIB to that location ceased. Further, when all skin surfaces were covered (helmet, long sleeves, long pants) SIB was virtually absent. This analysis provided support that reinforcement might be related to direct contact with skin.

Several medication trials, including clonidine, fluoxetine, haloperidol, and propranol, were not effective in reducing SIB. Because these medications address alternative hypothesized biologic functions for SIB (namely, synthetic opioid antagonists might reduce SIB maintained by endorphins) this added further support to the hypothesized function of sensory reinforcement (see, for example, Mace and Mauk 1995).

The current study was conducted across 3 days, all within one month. Prior to the beginning of the study, parental informed consent and approval from the hospital's internal review board were obtained. Sessions were scheduled on the basis of the availability of the family for trips to the hospital. All sessions were 10 minutes in duration and were conducted in a manner similar to play; that is, the participant interacted with his mother in preferred activities. No reinforcement was provided during the session. All exposed skin surfaces were covered with heavy clothing, with the exception of the participant's cheeks, which were his preferred site of SIB. Skin surface was normal and free from abrasions.

During the Anesthetic condition, 1g of eutectic marcaine lidocaine analgesic (EMLA), a dermal analgesic commonly used for minor skin procedures such as the insertion of intravenous catheters (Choy et al. 1999), was applied to the participant's cheeks 1 hour before the session. The active ingredients of EMLA are lidocaine (2.5%) and prilocaine (2.5%). The EMLA was manufactured in a patch form, with the cream analgesic located on a patch approximately 5cmx5cm, which was placed directly on the skin surface of the participant's cheeks. Just before the session, the occlusive dressing was removed. During the No Anesthetic condition, sessions were conducted without applying the EMLA. On day 3, when the No Anesthetic session followed the Anesthetic session (B-A), it was conducted 4 hours after the application of the anesthetic to allow the analges\ic effect to resolve.

Results

On day 1, self-slapping occurred at a rate of 46/min in the No Anesthetic condition. After application of the anesthetic, SIB decreased to 26/min (43% reduction). On day 2, SIB was observed at a rate of 29/min before application of the anesthetic, and 16/min after application of the anesthetic (45% reduction). The order of sessions was reversed on day 3 (Fig 1). Rates of SIB were 17/min during the Anesthetic condition, and 23/min during the No Anesthetic condition (26% reduction) .

The results indicate that SIB was 26 to 45% lower during sessions when the topical anesthetic was applied. The smallest differences occurred on day 3, when the Anesthetic condition was conducted before the No Anesthetic condition. There is a slight possibility that this might be attributable to residual effects of the medication. However, EMLA effects generally persist for up to 2 hours after removal of the patch, so this is unlikely. Extinction resulting from carry-over or natural variability across days might also account for these differences.

Discussion

The most likely explanation for the observed difference in the frequency of SIB across experimental conditions is that the topical anesthetic reduced sensory input to the surface of the skin. Thus, the pharmacologic action functioned in a manner similar to physical blocking of the response, resulting in a decrease in SIB. However, it is also possible that the effectiveness is attributable to a masking effect, where the EMLA alleviated an irritating condition.

Although the anesthetic resulted in decreases in SIB, it did not eliminate it. This is probably because it was not possible to entirely anesthetize the participant's cheek area. Because the EMLA was manufactured in the form of a small patch, the analgesic effects were limited to the area covered by the patch.

Several limitations in the current research should be noted. First, the participant's parents were not blind to the experimental conditions. Although unlikely, there is a slight possibility that parental interaction differed across experimental conditions. Future research might blind parents to the experimental conditions. Alternately, the use of a placebo would further control for this and other potential confounds.

An additional limitation pertains to session length. Observations were limited to 10 minutes because the purpose of this evaluation was assessment only. This brief session length is conventional for assessment purposes (e.g. Iwata et al. 1994), but treatment implications are limited. Further research and additional replications are necessary before considering this intervention for clinical use.

In spite of the limitations, this research contributes to the existing literature in several ways. First, our study diverged from previous assessments that have used apparatus (such as goggles, helmet, or wrist weights) to examine whether SIB has an automatic function. The disadvantage of apparatus is that it can physically interfere with SIB or can reduce SIB through punishment. In the present study we demonstrated an alteration of the direct consequences of SIB without the use of such devices. This might provide a clearer demonstration of reinforcement in the form of sensory consequences.

In addition, this is the first study that has used a topical anesthetic for SIB. Research on the use of EMLA in children has been limited to its analgesic effects when used for minor skin procedures (Koh et al. 1999). However, it should be noted that this is a preliminary investigation conducted with only one participant. In addition, the effect of the EMLA was not large and it did not offer a practical or feasible intervention strategy. More research is needed before clinical use to determine whether it offers a potential treatment option for specific subtypes of SIB. Further, with regard to the applied significance of the findings, protective equipment is often stigmatizing for individuals with disabilities. As an alternative, the EMLA resulted in reductions in SIB with no visible changes in the participant's appearance. Thus, with further research, interventions of this nature might represent a non- stigmatizing strategy for reducing SIB. Additional research might identify alternative anesthetic and nerve-block agents that have similar but more precise mitigating effects on SIB.

DOI: 10.1017/S0012162203001427

Accepted for publication 30th July 2003.

* UK usage: learning disabilities.

References

Choy L, Collier J, Watson AR. (1999) Comparison of lignocaine- prilocaine cream and amethocaine gel for local analgesia before venepuncture in children. Acta Pediatr 88: 961-964.

Iwata BA, Dorsey MF, Slifer KJ, Bauman KE, Richman GS. (1994) Toward a functional analysis of self-injury. J Appl Behav Anal 27:197-209.

Kazdin AE. (1982) Single-Case Research Designs: Methods for Clinical and Applied Settings. New York: Oxford University Press.

Kennedy CH, Souza G. (1995) Functional analysis and treatment of eye poking. J Appl Behav Anal 28: 27-37.

Koh JL, Fanurik D, Stoner PD, SChmitz ML, VonLanthen M. (1999) Efficacy of parental application of eutectic mixture of local anesthetics for intravenous insertion. Pediatrics 103: 79.

Mace FC, Mauk JE. (1995) Bio-behavioral diagnosis and treatment of self-injury. Ment Retard Dev Disabil Res Rev 1: 104-110.

Repp AC, Harman ML, Felce D, VanAcker R, Karsh KL. (1989) Conducting behavioral assessments on computer collected data. Behav Assess 2: 249-268.

Schopler E, Reichler RJ, DeVelluius RF, Daly K. (1980) Toward objective classification of childhood autism: Childhood Autism Rating Scale (CARS). J Autism Dev Disord 10: 91-102.

Vaughn ME, Michael JL. (1982) Automatic reinforcement: an important but ignored concept. Behaviorism 10: 217-227.

Lee Kern *PhD, Lehigh University, Bethlehem, PA;

Deborah Bailin MA, Friends School;

Joyce E Mauk MD, Child Study Center of Fort Worth and the University of North Texas, USA.

* Correspondence to first author at College of Education, Lehigh University, 111 Research Drive, Bethlehem, PA 18015, USA.

E-mail: lek6@lehigh.edu

Copyright Mac Keith Press Nov 2003

More News in this Category