NADD Bulletin Volume VI Number 3 Article 2

Complete listing

Treatment Effects of Paroxetine on Disruptive and Self-Harming Behaviors Associated With Polyuria in an Adult With Mental Retardation, Autism, and Obsessive Compulsive Disorder

James M. Sperry & James K. Luiselli (The May Institute Inc., and The May Center for Applied Research)

Mark J. Hauser (Harvard Medical School)

Christine Magee & Michelle Magnifico (The May Center for Adult Services)


We describe a data-based case study in which the antidepressant medication, paroxetine, was effective in reducing severe behavior outbursts exhibited by a 27-year old woman who had mental retardation, autism, and obsessive compulsive disorder. The outbursts consisted of self-injury, disrobing, screaming, and urinating, and had been refractory to non-medication interventions. Administration of paroxetine was associated with near-elimination of behavior outbursts and the results were maintained at a 4-month follow-up assessment. Factors influencing therapeutic outcome are discussed.


Recent advances in psychopharmacology with persons who have developmental disabilities suggest that impulsive-aggressive behavior that is both self- and other-directed may be the result of serotonin (5-HT) dysregulation (Ashberg, Schalling, & Traksmann-Bendz, 1987; Zubieta & Alessi, 1993). Antidepressant medications such as the tricyclics and SSRIs enhance serotonergic functioning and as such, have been evaluated as intervention for challenging behaviors displayed by children and adults with mental retardation, autism, and pervasive developmental disorder (PDD). This research includes single-case and group-comparison trials of clomipramine (Garber, McGonigle, Slomka, & Monteverde, 1992; Brodkin, McDougle, Naylor, Cohen, & Price, 1997), fluoxetine (Bass & Beltis, 1991; King, 1991; Markowitz, 1992; Ricketts, Goza, Ellis, Singh, & Singh, 1993), and sertraline (Hellings, Kelley, Gabrielli, Kilgore, & Shah, 1996; Luiselli, Blew, & Thibadeau, 2001).

Paroxetine (Paxil ) is an SSRI used effectively to reduce self-injury in a 15-year old boy (Snead, Boon, & Presberg, 1994) and behavior outbursts in a 7-year old boy (Posey, Litwiller, Koburn, & McDougle, 1999), both diagnosed with autism. In a prospective study of paroxetine with 15 adults who had severe-to-profound mental retardation, Davanzo, Belin, Widawsky, and King (1998) found decreased aggression one month following medication administration but this outcome was not maintained four months later.

In contrast to other antidepressant medications used with persons who have developmental disabilities, paroxetine has been studied less frequently. Also, the diminished long-term effects reported by Davanzo et al. (1998) emphasizes the need for follow-up assessment when judging the potential benefit from this medication. Finally, as with any evaluation of pharmacologic efficacy, determining its clinical utility would be aided by objective measurement of presenting problems.

The following case report describes a data-based evaluation of paroxetine as intervention for multiple challenging behaviors in an adult female with mental retardation, autism, and obsessive compulsive disorder. In this analysis, we measured directly the frequency and duration of challenging behaviors, implemented procedures under naturalistic conditions, and assessed therapeutic outcome several months post-intervention.


Participant and Case History

Jane was a 27-year old woman diagnosed with severe-to-profound mental retardation, autism, and a seizure disorder. She did not speak but was able to communicate through gestures, facial expressions, and pointing to picture icons. Jane could perform most self-care skills, simple domestic tasks (e.g., laundry, making bed, emptying trash), and some leisure activities semi-independently. She preferred spending time alone and rarely initiated interaction with other adults.

When she was 10 years old Jane was enrolled in a residential school for children with developmental disabilities. Her history was significant for serious challenging behaviors that included self-injury (striking head and body with hands), disrobing, high-intensity screaming, and excessive urinating on her person. At times the self-injury was so severe that she was considered at risk for retinal detachment. Jane experienced multiple residential placements over the years that focused on behavior management and intervention. In several of these settings she wore different types of protective equipment to prevent self-inflicted tissue damage.


The setting was a day-habilitation center operated by a private behavioral healthcare organization serving children and adults with developmental disabilities. Jane attended the center six hours each weekday. The center was comprised of 18 women and men who had mental retardation, with a ratio of 1 staff person to 5 adults. Training activities with Jane and her peers concentrated on piece-work, clerical support, building maintenance, community travel, and social recreation.


When Jane displayed challenging behaviors they typically occurred simultaneously in the form of an "outburst." For the purpose of data collection, a behavior outburst was defined as a combination of at least two of the following responses: (1) self-injury that included a slap to the body, audible contact of head against a stationary object, or scratching skin, (2) removing any article of clothing in a location outside of a bathroom, (3) vocalizing above a conversational tone longer than a 3-second duration, and (4) urinating in clothing or subsequent to disrobing in a location other than a bathroom. It should be noted that each behavior outburst was "ritualized" in appearance. That is, the outburst started with Jane screaming, engaging in self-injury, and running from the immediate environment to a "quiet area" (described below). When in this area she then removed her clothing and urinated on her body and the floor. Urinating occurred every time an outburst was displayed and it was always the terminal response.


Staff working with Jane recorded each behavior outburst on a standardized data sheet (frequency count). The time (to the nearest minute) that the outburst began and ceased (60 consecutive seconds of nonagitation) also was recorded to yield a duration measure. Staff were taught the measurement procedures by a clinical specialist at the center and data were reviewed with them during regularly scheduled supervision meetings.

Inter-observer agreement (IOA) for the frequency and duration measures was assessed one day per week by having two staff record data simultaneously but independently. To be scored as an agreement both staff had to record the same number of behavior outbursts during the day, and a corresponding duration plus or minus 60 seconds. Frequency and duration IOA were 100% on all occasions.


The case study included baseline (no medication), medication, and follow-up phases. Before medication was prescribed, a behavior support plan had been designed for Jane and it was in effect across all phases of the study. Briefly, functional behavioral assessment (Iwata, Vollmer, & Zarcone, 1990) suggested that behavior outbursts were reinforced by "escape" from training activities, access to new clothes, and sensory pleasurable consequences. The resulting behavior support plan consisted of several components:

(1) Jane received social praise from staff when she completed scheduled activities and choice of a preferred event at the end of each day in which behavior outbursts did not occur.

(2) Contingent on an outburst, one staff person directed Jane to walk to a designated "quiet area" in the center building. She was first given a verbal instruction, followed by the instruction with a gesture, and finally the instruction paired with light touch to her shoulder if she did not respond to the initial cue.

(3) Once in the "quiet area," staff stopped interacting with Jane and allowed her to compose herself. Any attempt at self-injury was blocked by staff according to approved physical management/crisis intervention guidelines. Jane remained in the "quiet area" until she demonstrated 60 consecutive seconds of nonagitation.

(4) If Jane urinated in the "quiet area" she was required to clean herself with an antiseptic wash cloth and dress in new clothing if her original clothes were wet. She then assisted staff in cleaning any section of the "quiet area" where urine was present. When this task was completed, Jane washed her hands in a nearby bathroom and returned to ongoing activities according to the daily schedule.

Baseline (8 weeks). As noted previously, staff implemented the behavior support plan during baseline. Jane continued to receive carbamazepine (1000 mg daily), which she had been taking several years preceding the study. No other psychotropic medications were prescribed.

Paroxetine (23 weeks). Through consultation with a board-certified psychiatrist, a medication trial of paroxetine was recommended. The psychiatric evaluation suggested that Jane had "bland affect" and that her behavior outbursts were "perseverative" and resembled "rituals." The elevated frequency of urinating (polyuria) which was exhibited every time Jane had an outburst also was highlighted as an unusual clinical presentation. So conceived, it was reasoned that Jane might have unrecognized obsessive compulsive disorder (OCD) as is sometimes the case with persons who have developmental disabilities (Geyde, 1996; Kostinas, Scandlen, & Luiselli, 2001; Vitiello, Spreat, & Behar, 1989).

Based on the preceding formulation the decision was to evaluate the potential therapeutic effect from an SSRI. Jane was prescribed paroxetine at 10 mg per day, and two weeks later it was increased to a 20 mg daily dose. Nursing staff at the center provided oversight and side effects monitoring. This information, combined with the frequency and duration data for the behavior outbursts, were reviewed by the psychiatrist at scheduled consultations. Jane continued to take 1000 mg of carbamazepine each day as she did in the baseline phase.

Follow-up. Data are reported during a 4-month follow-up phase while the 20 mg per day dose of paroxetine was still in effect.


Figure 1 presents the average behavior outbursts per day displayed by Jane each week. During the baseline phase the average daily frequency was 2.2 (range = 1.3-3.4). Subsequent to the introduction of paroxetine, Jane had fewer behavior outbursts per day, demonstrating a steady reduction in frequency throughout the phase. During the 23 week course of medication administration the average was .49 behavior outbursts per day (range = 0-1.8). The results at follow-up revealed low-frequency behavior outbursts that averaged .15 per day.


Insert Figure 1 about here


The average duration per behavior outburst each week is shown in Figure 2. In baseline, outbursts lasted an average of 5.6 minutes (range = 4.4-6.7 minutes). The average duration, 5.2 minutes, did not change appreciably with the addition of paroxetine (range = 3.0-6.7 minutes). At follow-up there were two months where the average duration increased (7.5 minutes at 1-month and 8.0 minutes at 4-months) and two months where the average duration remained at previous levels (5.2 minutes).


Insert Figure 2 about here



The addition of paroxetine was associated with a dramatic reduction in the frequency of behavior outbursts demonstrated by an adult with mental retardation, autism, and obsessive compulsive disorder. Jane had a complex diagnostic profile and a protracted history of serious challenging behaviors. At the time this case study was initiated she was engaging in high-frequency screaming, self-injury, and disrobing. These behaviors interfered with her habilitation, disturbed other consumers, were socially stigmatizing, and posed a health risk due to potential tissue damage. The occurrence of urinating contemporaneously with these behaviors also was perplexing and had a detrimental impact on Jane and the physical surroundings. By the follow-up phase, behavior outbursts essentially had been eliminated. As a result, Jane was able to participate in training activities more proficiently, her rate of skill acquisition increased, and she gained greater exposure to the community.

We discussed earlier that paroxetine was selected because of the effectiveness of SSRIs in treating externalizing behavior disorders in persons with developmental disabilities (Stigler, Poser, & McDougle, 2002), including repetitive, compulsive, and "perseverative" features. Jane presented with rituals such as ordering her environment and completing tasks in a rigid fashion, and her outbursts also were exhibited as a sequence of behaviors in a fixed pattern. The apparent therapeutic benefit from paroxetine may have been the result of its anti-obsessional properties or perhaps a generalized effect that enhanced her affect and responsiveness during training activities. Subjective reports from staff, for example, confirmed that her mood improved positively with medication, as she appeared "happier," smiled more often, and had increased energy. These findings and observations suggest that Jane was highly responsive to antidepressant medication although the mechanism responsible for change is uncertain.

Side effects from antidepressant medications such as paroxetine include gastrointestinal distress, mania, motor restlessness, and sedation (Benefield & Tramonte, 1997; Riddle, King, & Hardin, 1990; Venkataraman, Naylor, Miwi, & King, 1992). Importantly, none of these or other untoward effects were observed with Jane. On this note it is significant that she was able to tolerate a relatively low dose of medication, with desirable clinical outcome maintained, over a near 40-week course of treatment.

The data from this case study revealed that the frequency of behavior outbursts decreased with medication but duration was unaffected. In fact, the average amount of time per outburst actually "spiked" during two of the follow-up months. Thus, when Jane had an outburst it could be expected to last a similar duration, independent of how often its occurred. This finding underscores the importance of objective measurement when evaluating intervention efficacy because medication can have differential effects on presenting problems.

As a case study our evaluation can only suggest and not confirm that paroxetine caused thae gains achieved with Jane. First, the medication trial was "open label" and not a double-blind and placebo-controlled evaluation. Second, we chose to not discontinue the medication with Jane in a reversal to baseline phase. Manipulations of this type simply were too difficult to coordinate in a "real world" setting, nor was it justified to stop treatment, even temporarily, with an individual who had such extreme challenging behaviors and responded so well to medication. On a clinical level, our findings add to the small body of literature concerning pharmacologic treatment with paroxetine for persons who have developmental disabilities. Furthermore, it appears that the positive behavior-change from this medication can be maintained long-term and without adverse effect.




Ashberg, M., Schalling, D., & Traksmann-Bendz, L. (1987). Psychobiology of suicide, impulsivity, and related phenomena. In H. Y. Meltzer (Ed.), Psychopharmacology: Third generation in progress (pp. 655-688). New York: Raven Press.

Bass, J. N., & Beltis, J. A. (1991). Therapeutic effects of fluoxetine on naltrexone-resistant self-injurious behavior in an adolescent with mental retardation. Journal of Child and Adolescent Psychopharmacology, 1, 331-340.

Benefield, W. H., & Tramonte, S. M. (1997). Guidelines for selecting antidepressant therapy. The Habilitative Mental Healthcare Newsletter, 16, 109-112.

Brodkin, E. S., McDougle, C. J., Naylor, S. T., Cohen, D. J., & Price, L. H. (1997). Clomipramine in adults with pervasive developmental disorders: A prospective open-label investigation. Journal of Child and Adolescent Psychopharmacology, 7, 109-121.

Davanzo, P. A., Belin, T. R., Widawski, M. H., & King, B. H. (1998). Paroxetine treatment of aggression and self-injury in persons with mental retardation. American Journal on Mental Retardation, 5, 427-437.

Garber, H. J., McGonigle, J. J., Slomka, G. T., & Monterverde, E. (1992). Clomipramine treatment of stereotypic behaviors and self-injury in patients with developmental disabilities. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 1157-1160.

Geyde, A. (1996). Issues involved in recognizing obsessive compulsive behavior developmentally disabled clients. Seminars in Clinical Neuropsychiatry, 1, 142-147.

Hellings, J. A., Kelley, L. A., Gabrielli, W. F., Kilgore, E., & Shah, P. (1996). Sertraline response in adults with mental retardation and autistic disorder. Journal of Clinical Psychiatry, 57, 333-336.

Iwata, B. A., Vollmer, T. R., & Zarcone, J. R. (1990). The experimental (functional) analysis of behavior disorders: Methodology, applications, and limitations. In A. C. Repp & N. N. Signh (Eds.), Perspectives on the use of nonaversive and aversive interventions for persons with developmental disabilities (pp. 301-330). Sycamore, IL: Sycamore Publishing.

King, B. H. (1991). Fluoxetine reduced self-injurious behavior in an adolescent with mental retardation. Journal of Child and Adolescent Psychopharmacology, 1, 321-329.

Kostinas, G., Scandlen, A., & Luiselli, J. K. (2001). Effects of DRL and DRL combined with response cost on perseverative verbal behavior of an adult with mental retardation and obsessive compulsive disorder. Behavioral Interventions, 16, 27-37.

Luiselli, J. K., Blew, P., & Thibadeau, S. (2001). Therapeutic effects and long-term efficacy of antidepressant medication for persons with developmental disabilities. Behavior Modification, 25, 62-78.

Markowitz, P. I. (1992). Effect of fluoxetine on self-injurious behavior in the developmentally disabled: A preliminary study. Journal of Clinical Psychopharmacology, 12, 27-31.

Posey, D. J., Litwiller, M., Koburn, A., & McDougle, C. J. (1999). Paroxetine in autism. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 111-112.

Ricketts, R. W., Goza, A. B., Elis, C. R., Singh, Y. N., & Singh, N. N. (1993). Fluoxetine treatment of severe self-injury in young adults with mental retardation. Journal of the American Academy of Child and Adolescent Psychiatry, 32, 865-869.

Riddle, M. A., King, R. A., & Hardin, M. T. (1990). Behavioral side effects of fluoxetine in children and adolescents. Journal of Child and Adolescent Psychopharmacology, 3, 193-198.

Snead, R. W., Boon, F., & Presberg, J. (1994). Paroxetine for self-injurious behavior. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 909-910.

Stigler, K. A., Posey, D. J., & McDougle, C. J. (2002). Recent advances in the pharmacotherapy of autism. Expert Review of Neurotherapeutics, 2, 499-510.

Venkataraman, S., Naylor, M. W., Miwi, B., & King, C. A. (1992). Mania associated with fluoxetine treatment in adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 31, 276-281.

Vitiello, B., Spreat, S., & Behar, D. (1989). Obsessive compulsive disorder in mentally retarded patients. Journal of Nervous and Mental Disease, 177, 232-236.

Zubieta, J. K., & Alessi, N. E. (1993). Is there a role of serotonin in the disruptive behavior disorders? A literature review. Journal of Child and Adolescent Psychopharmacology, 3, 11-35.

Authors' Note: This study was conducted at The May Center for Adult Services, Revere, MA. Requests for reprints should be sent to James K. Luiselli, Ed.D., ABPP, BCBA, The May Center for Applied Research, The May Center Inc., One Commerce Way, Norwood, MA 02062 (e-mail:



Figure Captions

Figure 1. Average behavior outbursts per day.

Figure 2. Amerage duration (minutes) per behavior outburst