Manualized Assessment and Treatment Model of Challenging Behavior

The investigators will stratify participants according to the severity of their challenging behavior (harmful or milder forms of challenging behavior).

Criteria for Harmful Challenging Behavior Participants will be classified as exhibiting harmful challenging behavior if they meet all the following criteria and as exhibiting milder challenging behavior if they do not meet all these criteria.

1. The child displays one or more of the following challenging behaviors (operationally defined on the Destructive Behavior Severity Scale, see attached) daily: (a) aggression, (b) elopement, (c) injury risk behavior, (d) pica, (e) property destruction, and/or (f) SIB.
2. The child's challenging behaviors are at a severity level that poses a clear and serious risk to oneself, others, or the environment (Severity Level 2 or higher on the Destructive Behavior Severity Scale, see attached).
3. The child's challenging behavior causes clear and significant stress on the child's family members (e.g., siblings are deprived of parental attention; family members are restricted from maintaining employment, going on family outings, attending church, eating at a restaurant, or taking a vacation due to the child's challenging behavior).
4. The child has had at least 3 months of outpatient behavior therapy with insufficient improvement in the challenging behavior or the child's challenging behavior poses an imminent danger to oneself or others such that less intensive services would be contraindicated.
5. The child has had at least 3 months of outpatient psychopharmacological intervention by a psychiatrist or developmental pediatrician with insufficient improvement in the challenging behavior, or the psychiatrist or developmental pediatrician has recommended adding intensive behavioral intervention.

The manualized protocol has three general components that will be used with participants in both groups including: (a) assessment, (b) treatment, and (c) caregiver training and generalization.

Assessment

Trial-Based Functional Analysis (FA) The investigators will begin by conducting a trial-based FA with each participant. During the trial-based FA, a multielement design will be used to evaluate conditions that maintain challenging behavior. During the first series of conditions, researchers will measure the latency to engage in challenging behavior from the start of each trial to determine an appropriate duration subsequent for trials. Conditions will be evaluated in the following order: monitored ignore/alone, attention, toy-play, escape, and tangible. Other conditions (e.g., social avoidance) will be evaluated as appropriate for each participant if little or no challenging behavior is observed in above conditions. Researchers will wear colored shirts to signal each condition. After the first series of trials (i.e., one trial per condition), the duration of each subsequent trial will be determined by multiplying the latency for each condition by 1.5. The minimum trial duration will be 30 s and the maximum trial duration will be 5 min. Each condition will arrange specific antecedent and consequent conditions for challenging behavior. Specifically, evocative events will be arranged (e.g., restricting attention in the attention condition) and contingent on the first instance of challenging behavior, reinforcement will be delivered for 20 s (e.g., attention will be delivered in the attention condition). Rates of challenging behavior will be analyzed across conditions to determine the variable(s) that maintained challenging behavior.

Multielement FA If definitive conclusions cannot be drawn about the conditions that maintain challenging behavior from the trial-based FA, the investigators will conduct a more-extended and traditional multielement FA. The multielement FA will begin with a 15 min monitored ignore screener to rule in or rule out automatic maintenance of challenging behavior (maintained by conditions other than those that researchers have control over). During this session, researchers will not provide any consequences for challenging behavior. Following the initial monitored ignore session, researchers will evaluate social conditions that maintain challenging behavior in a multielement design. All social test sessions will last 5 min. Conditions will be evaluated in the following order: monitored ignore/alone, attention, toy-play, escape, and tangible. Other conditions will be evaluated as appropriate for each participant (e.g., social avoidance). Researchers will wear colored shirts to signal each condition. Monitored ignore conditions will be conducted like the initial screener. During the social test conditions (attention, tangible, and escape), researchers will systematically arrange evocative conditions and deliver consequences contingent on challenging behavior. For example, during the escape condition, researchers will begin the session by delivering instructions to complete tasks. Researchers will deliver brief praise for task completion. Contingent on challenging behavior, researchers will provide a break from tasks for 20 s to determine whether escape from nonpreferred tasks maintains challenging behavior. Following 20 s, researchers will begin delivering instructions again until the next instance of challenging behavior. This sequence of events will continue until 5 min have elapsed. The toy-play condition will serve as the control condition, in which participants will have access to highly preferred tangible items and attention, and no task instructions will be delivered. The investigators will conduct a minimum of five sessions per conditions (e.g., five attention sessions). Rates of challenging behavior will be analyzed across conditions to determine the variables maintaining challenging behavior.

Treatment

Treatment decisions will be driven by informed caregiver preferences and choices. That is, the investigators will use the results of our assessments to inform caregivers regarding all the treatment options and choices available to them. For example, if two or more behavioral treatments (e.g., functional communication training or noncontingent reinforcement) are reasonable options for the patient, the supervising behavior analyst will explain the potential benefits and risks associated with each option and ask the caregiver to choose between them.

Harmful Challenging Behavior

Once the variable(s) that maintain challenging behavior are identified, the investigators will inform caregivers of the treatment options and develop a function-based treatment for each participant. If challenging behavior is found to be socially maintained, the investigators will typically recommend functional communication training (FCT) to teach a functional communication response (FCR) (e.g., touching a card with a picture of the participant consuming the reinforcer), but the investigators will also offer NCR as a potential treatment option to the caregivers. During FCT, reinforcement will be discontinued for challenging behavior and only the alternative communication response will be reinforced. For participants with challenging behavior determined to be maintained by social-positive reinforcement (attention and/or tangible), the investigators also will typically recommend using a multiple schedule FCT (mult-FCT) to signal when reinforcement is available and to thin the schedule of reinforcement to render the treatment more practical for caregivers to implement. For participants with challenging behavior determined to be maintained by social-negative reinforcement (escape), the investigators will recommend using a chained schedule FCT (chained-FCT). If the parents choose NCR over FCT for socially maintained challenging behavior, the investigators will deliver the functional and competing reinforcers on time-based schedules and use multiple and chained schedules to signal when noncontingent reinforcement is available (similar to multiple- and chained-FCT). If challenging behavior is found to be automatically maintained, the investigators will recommend using NCR with competing items and response blocking for treatment. The investigators will also use multiple and chained schedules to thin the reinforcement schedules and increase the practicality of the this treatment.

Functional Communication Training

Pretraining. The investigators will train participants to use the FCR during mult-FCT and chained-FCT using prompting and prompt fading.

Mult-FCT. The investigators will choose two colored signals and randomly assign two different colored cards as the SD (which signals the availability of reinforcement) and S-delta (which signals the unavailability of reinforcement). The session will start with an SD component followed by an S-delta component followed by a quasi-random rotation of the SD and S-delta components. SD and S-delta components will be 60 s and 2 s, respectively. If the participant uses the FCR without displaying challenging behavior for at least 3 s during the SD component, the therapist will deliver functional reinforcement for the rest of the SD component. If the participant emits a challenging behavior within 3 s of emitting the FCR, the therapist will withhold reinforcement until the participant emits FCR without also emitting challenging behavior. The therapist will deliver no differential consequence (extinction) for the FCR with the S-delta present and for challenging behavior during the SD and S-delta components (extinction). The duration of the S-delta component will be systematically increased as challenging behavior remains low and independent FCRs remain high to make the treatment more practical for caregivers.

Chained-FCT. The procedures for chained-FCT will be like mult-FCT, except participants will be required to complete work (e.g., academics, daily living skills) during the S-delta component. The SD component will be 20 s and the S-delta component will be as long as it takes participants to complete one task. Initially, the participant will be required to complete just one task (e.g., fold one towel) during the S-delta component. Over time, the number of tasks required during the S-delta component will be systematically increased as challenging behavior remains low, task completion remains high, and independent FCRs remain high.

Competing Items and Response Blocking. The investigators will conduct a competing stimulus assessment (CSA) to identify items that produce at least an 80% reduction in challenging behavior when the participant is engaged with them. The investigators will evaluate conditions in which participants have free access to the item(s), when engagement with the item(s) is prompted, and when challenging behavior is physically blocked. Treatment will include one or more of the following components: non-contingent reinforcement informed by the CSA, reinforcing adaptive behavior, and blocking challenging behavior. The ending duration of the S-delta component will be based on input from caregivers and the participant's school personnel.

Mild Challenging Behavior

Once the variable(s) that maintain challenging behavior are identified, the investigators will inform caregivers of the treatment options and develop a function-based treatment for each participant. Caregivers of patients with mild challenging behavior will receive training using the Research Units in Behavioral Intervention (RUBI) protocol informed by the functional analysis conducted as a part of the initial assessment (Bearss et al., 2018). The results of the functional analysis will inform the development of the training delivered as part of the RUBI protocol. An outline of the standard RUBI protocol is displayed below.

1. Week 1: Introduction to behavioral principles
2. Week 2: Prevention strategies
3. Week 3: Daily schedules
4. Week 4: Home visit and assessment
5. Weel 5: Reinforcement I
6. Week 6: Reinforcement II
7. Week 7: Planned ignoring
8. Week 8: Assessment
9. Week 9: Compliance training
10. Week 10: Functional communication training
11. Week 11: Teaching skills I
12. Week 12: Assessment
13. Week 13: Teaching skills II
14. Week 14: Generalization \& maintenance
15. Week 15: Optional lessons
16. Week 16: Assessment
17. Week 18: Telephone booster

Caregiver training and generalization

Using a combination of training strategies (e.g., behavioral skills training, video modeling) and caregiver preference, the investigators will train participants' caregivers to implement treatment techniques with accuracy. The investigators will generalize treatment effects for all patients to novel people and settings and maintain an 80% reduction in challenging behavior and high levels (85% or higher) of appropriate behavior (e.g., FCRs).

Harmful Challenging Behavior. The investigators will train at least one of the caregivers to 90% or greater accuracy when implementing the techniques below. To maximize safety, the investigators will assess caregiver behavior with therapists role-playing as the patient before the investigators have the caregiver implement the skills with the patient directly. Caregiver training will include the steps listed and described below, and the investigators anticipate progressing to a new step each week of the patient's admission in the Severe Behavior Program:

1. The investigators will collaborate with the caregiver to identify critical safety concerns within the home or community and teach strategies to minimize risk (e.g., clearing rooms during challenging behavior, using buckle guards to reduce out-of-seat behavior during transportation).
2. The investigators will train the caregiver in the following behavior-management techniques until the caregiver responds with 90% or greater accuracy for each of the following: Responses to aggression role played by behavior analyst, Responses to self-injurious behavior role played by behavior analyst, Procedures for managing challenging behavior that poses imminent threat with a behavior analyst role-playing as the patient
3. The investigators will teach the caregiver to identify the common functions of challenging behavior. The caregiver will respond with 90% accuracy to comprehension questions.
4. The investigators will teach the caregiver how to collect data at home relevant to the patient's challenging and adaptive behavior. The caregiver will display 90% or greater data accuracy when the behavior analysts role plays as the patient with challenging behavior.
5. The investigators will teach the caregiver about extinction and train the caregiver to implement extinction procedures relevant to the patient's initial treatment arrangement. The caregiver will display 90% or greater accuracy when the behavior analyst role plays as the patient with challenging behavior.
6. The investigators will teach the caregiver about reinforcement-based interventions and train the caregiver to deliver differential reinforcement of alternative or other behavior and noncontingent reinforcement relevant to the patient's initial treatment arrangement. The caregiver will display 90% or greater accuracy when the behavior analyst role plays as the patient with challenging behavior.
7. The investigators will train the caregiver to implement three-step guided compliance procedures. The caregiver will display 90% or greater accuracy when the behavior analyst role plays as the patient with challenging behavior.
8. The investigators will teach the caregiver about the importance of conducting periodic preference assessments with the participant and how to implement one practically. The caregiver will display 90% or greater accuracy when the behavior analyst role plays as a patient during a preference assessment.
9. The investigators will train the caregiver to implement the reinforcement component of the patient's terminal treatment package. The caregiver will display 90% or greater accuracy (a) when the behavior analyst role plays as the patient with challenging behavior and (b) during two sessions with the patient.
10. The investigators will train the caregiver to implement the extinction component of the patient's terminal treatment package. The caregiver will display 90% or greater accuracy (a) when the behavior analyst role plays as the patient with challenging behavior and (b) during two sessions with the patient.
11. The investigators will train the caregiver to implement both the reinforcement and extinction components of the patient's terminal treatment package while also managing challenging behavior. The caregiver will display 90% or greater accuracy (a) when the behavior analyst role plays as the patient with challenging behavior and (b) during two sessions with the patient.
12. The investigators will coach the caregiver to conduct the entirety of the patient's treatment at the terminal schedule with minimal trainer involvement. The caregiver will display 90% or greater accuracy during two sessions with the patient without requiring therapist feedback.

Milder Challenging Behavior. Caregivers of patients with mild challenging behavior will receive training on a variety of topics throughout implementation of the RUBI protocol (outlined above).

Data Points

For participants with harmful challenging behavior who are being treated in our intensive outpatient clinic, trained observers will use laptop computers with BDataPro® software developed in our lab to score frequency of participant challenging behavior and therapist behavior. The investigators will define challenging behavior as pushing, pinching, scratching, kicking, or biting others; hitting others with body parts or objects; banging, throwing, overturning, or tearing objects; climbing on objects not made for that purpose; striking, scratching, rubbing, poking, or biting self. The investigators will convert challenging-behavior frequency to a rate by dividing the number of responses in a session by the session duration in minutes. The investigators will assess procedural integrity by scoring whether therapists correctly implemented the planned antecedents, prompts, and consequences for target responses. The investigators will convert correct antecedents, prompts, and consequences to a percentage after dividing the number of correct responses by the number of correct-response opportunities.

Observation, reliability, and validity of dependent and procedural-integrity measures. A second observer will score at least 1/3 of sessions independently to assess data accuracy (reliability). Observers will score procedural integrity for at least 1/3 of sessions. The investigators will retrain therapists with less than 90% procedural integrity for two consecutive sessions.

For patients with mild challenging behavior who are receiving the RUBI program, the investigators will collect indirect measures of challenging behavior via caregiver report during each week of training. Additionally, the investigators will use the Destructive Behavior Severity Scale, the Clinical Global Impressions Severity Subscale, and the Aberrant Behavior Checklist-Irritability Subscale to collect data on challenging behavior. These assessments will be administered at each week labeled as "Assessment" in the above outline for the RUBI program