Implementation of Modified Positive Behavioral Interventions and Supports in a youth psychiatric partial hospital program

Introduction

Seclusion and restraint are procedures utilized in child psychiatry services for the management of aggressive and self-injurious behaviors when no other measures are effective. Seclusion is defined as the involuntary confinement of a patient to an area from which he or she is physically prevented from leaving, while restraint is defined as the restriction of the freedom of movement or normal access of a patient to his/her own body, either manually or with physical or mechanical device. ¹ For centuries, the use of seclusion and restraint has been debated. On the one hand, some have argued that their use may be necessary to maintain safety on a treatment unit yet these procedures have raised concern for serious risk. These procedures can compromise safety if performed incorrectly or monitored inadequately. ² They can also heighten aggressive behaviors, cause additional trauma, reduce trust in clinical staff, place staff at greater risk of assault and injury, and even lead to fatalities.^3–5

Due to these negative effects and potential for harm, the use of seclusion/restraint (SR) in pediatric settings has been a source of congressional inquiries in the United States ⁶ and has significant policy implications at the federal level. ⁷ The Joint Commission has implemented guidelines that restrict the use of SR to emergency situations in which there is imminent risk that the individual may physically harm self or others. ⁸ Even then, clinical practice dictates that it is only to be used as a last resort. The Substance Abuse and Mental Health Services Administration, American Psychiatric Nurses Association, American Academy of Child and Adolescent Psychiatry, and other agencies and organizations have called for a reduction and ultimate elimination of the use of these practices in institutional and community settings.^9–12 Current policy dictates that SR use is carefully regulated and monitored. ¹³

While there has been an effort to reduce and/or eliminate the use of SR, ¹⁴ there has been limited research on specific behavioral programs to reduce SR in youth psychiatric settings. ¹⁵ This is particularly true for the partial hospital level of acute psychiatric care for youth. As a result, there is a need for evidence-based behavioral interventions to reduce SR and for this type of intervention to be evaluated in a partial hospital level of care. Of note, partial hospital is a nonresidential treatment program that is hospital based; the program provides diagnostic and treatment services on a level of intensity similar to an inpatient program, but on less than a 24 h basis.

Common core programming elements that cut across a variety of programmatic attempts to lower the rates of SR have been identified.^16,17 These elements include leadership that sets the agenda for change, data use, therapeutic environment, tools for reacting to patients that display dangerous behaviors, consumer involvement, and debriefing. For example, the state of Massachusetts included these essential elements before implementing a state-wide initiative that required the clinical staff to be provided a clear set of expectations, alternatives to SR, and technical support for programming and strength-based interventions. ² Following year 1 of the program, the overall number and duration of SR episodes were reduced statewide on child, adolescent, and mixed age units.

While such key common factors are important for initiating any attempt to lower SR, it is critical to study the specific behavioral management programs that flow from such systemic initiatives. One such promising approach is Modified Positive Behavioral Interventions and Supports (M-PBIS), which has been found to reduce the use of restrictive interventions on a youth psychiatric inpatient unit. ¹⁸ Specifically, in the inpatient setting, M-PBIS reduced the number of SR events, mean duration of events, and percentage of patients placed in seclusion or restraint, as was the overall seclusion rate for the unit. Furthermore, there was a significant reduction in the use of pro re nata (PRN) medications for agitation/aggression. ¹⁸ M-PBIS was adapted from PBIS which is a universal, school-wide prevention strategy that is currently implemented in schools across the nation. ¹⁹ PBIS has core components which provide a framework that emphasizes socially valued and measurable outcomes, empirically validated and practical interventions, systems that support implementation of these practices, and continuous collection and use of data for decision making. Within this framework, there are a number of guiding principles including investing in prevention, teaching and acknowledging appropriate behaviors, and using progress monitoring to assess fidelity and the impact on outcomes.

The current study sought to examine the effectiveness of M-PBIS implemented on a youth psychiatric partial hospital service. It was hypothesized that the implementation of the program would lead to a significant reduction in the use of SR. Given the concern that the reduction in SR can lead to increase in PRN medication use, ²⁰ unlike similar studies that have attempted to lower SR, the current study also monitored the use of PRNs.

Method

This naturalistic, prospective study was based on data collected in a child and adolescent psychiatric partial hospital service. The pediatric partial hospital serves patients 5–17 years of age, with the goal of psychiatric stabilization and generalization of therapeutic gains. The partial hospital service has a 12-bed capacity. Patients attended the program from 7:30 am to 3:30 pm. The program includes programming for group therapy, individual sessions, schoolwork with public school teachers, as well as contact with psychiatrist and social work providers on a daily basis. Approximately one-third of patients attending the program were step-downs from an acute inpatient service, while the rest were step-up admissions from community clinics and local private child psychiatry practices. Patients attend the program for approximately two weeks (i.e. 10 treatment days, Monday–Friday). The study comprised admissions from September 2011 to December 2013.

The setting is an academic hospital in an urban center, with staffing consisting of one nurse and one psychiatric assistant for every six patients. Approval to review and abstract data was obtained from the Johns Hopkins School of Medicine institutional review board (IRB00111159). Information was collected on SR, PRN medication use of agitation/aggression, and length of stay September 2011–December 2013; of note, the month of training (November 2011) was not included in analyses. The preintervention period included September–October 2011. The postintervention period included December 2011–December 2013. SR included open door seclusion, locked door seclusion, and any type of restraint. PRN use included medication given for the indication of acute agitation/aggression. Possible medications included diphenhydramine, haloperidol, lorazepam, risperidone, olanzapine, and aripiprazole. PRN medications were collapsed across medication type, dose, and route of administration. Demographic and clinical variables were abstracted from electronic medical records and included age, sex, race/ethnicity, insurance status (Medicaid versus other forms), and primary discharge diagnosis (which were collapsed into diagnostic categories).

M-PBIS is a three-tiered continuum of support as previously described in Reynolds et al. ¹⁸ As was done for the inpatient service, in the partial hospital service, the Tier 1 strategies included establishing commitment from the staff (of note, 100% of the unit staff voted to implement the program), defined set of positively worded expectations (i.e. be safe, be responsible, be respectful), a behavior matrix identifying how to meet expectations in different program locations (e.g. activity room, meals; Figure 1), strategies for teaching the patient and their family the behavioral expectations during an initial orientation to the unit and throughout the course of patient’s stay (e.g. posters of expectations specified in Figure 1 are hung throughout the unit), a reward system in which positive/adaptive behaviors receive positive reinforcement (i.e. a stamp from staff given with labeled praise when expected behavior is displayed, stamps exchanged for various rewards at select times during the day; Figure 1), 5:1 positive to negative adult-to-child interaction ratio, a data monitoring system (e.g. SR and PRN medication rates, fidelity to program, and patient characteristics) including feedback to staff regarding rates of SR, staff recognition through weekly peer nomination, regular involvement of administrators, and an action committee comprised of a variety of unit staff to monitor fidelity to program and problem solve unit behavior management issues. Tier 2 included targeted problem-solving conversations with only those patients who demonstrated problem behavior on the unit that was not responsive to Tier 1 strategies (e.g. ongoing disruptive behavior in groups). Tier 3 interventions included functional behavior assessments and individualized behavior plans for a small minority of patients who did not respond to Tier 1 or Tier 2 strategies (e.g. increasing food intake for a patient with OCD who had food rituals).

OPEN IN VIEWER

Figure 1.

SR rate over the study period.

Staff-wide training included an initial day-long course (8 h) conducted to train unit staff across disciplines (e.g. nurses, psychiatric assistants, social workers, and physiatrists), yearly booster sessions, and ongoing consultation with an on-site psychologist. As new staff members joined the team, the same initial training was conducted (condensed into 4 h session). The program cost to the unit included compensation for 75% of a full time equivalent of a licensed psychologist, who served as the primary consultant for the M-PBIS intervention. In turn, the consultant’s time was devoted to training staff, data management, and ongoing consultation. Additional yearly costs also included rewards for patients (e.g. personal grooming supplies, stuffed animals, socks, balls).

After implementation, trained rater observations were conducted on the unit to examine the ratio of positive to negative adult-to-child interactions and use of the reinforcement system, that is providing positive reinforcement in the form of stamps within 2 seconds of adaptive behavior in tandem with labeled praise and contingent on a specified behavior. The 10 min observations were conducted (N = 190) at select times during the day (e.g. at meals and social worker leader group) by doctoral-level psychologists.

Data analysis

First, analyses were conducted to examine the demographic and clinical characteristics across the intervention period by using chi-square for categorical and independent-sample t-tests for continuous data. Next, the outcomes of interest were compared for the pre- and postintervention time periods: (1) SR, expressed per 1000 patient hours (independent-sample t-tests), (2) mean duration of SR episodes in minutes (independent-sample t-tests), (3) proportion of patients who were in SR (chi-square), and (4) proportion of patients who received a PRN (chi-square). In order to control for differences between groups on demographic and clinical characteristics, logistics regressions with relevant covariates were conducted. Finally, indicators of fidelity to the program were examined including the ratio of positive to negative adult-to-child comments and proper reinforcement provision. All analyses were conducted with IBM SPSS Statistics 23.

Results

There were 442 admissions during the study period from September 2011 to January 2013. Demographic and clinical characteristics of the children are summarized in Table 1. The demographic characteristics of the patients changed significantly during the course of the study; the average age increased, the proportion of black patients decreased, the proportion of those with Medicaid decreased, and the length of stay decreased (see Table 1). In addition, there was variability in the primary discharge diagnosis: the proportion of patients with ADHD/DBD decreased and the proportion of those with psychosis, depressive mood/bipolar, anxiety, and adjustment increased.

OPEN IN VIEWER

Table 1.

Characteristics of 442 admissions in the partial hospital program.

Variable	Preintervention (n = 21)	Postintervention (n = 421)	t/χ2	p
Age (M(SD))	11.48 (3.50)	13.27 (2.72)	−2.92	.004
Gender (% Male)	52.4	38.5	1.62	.203
Race (% Black)	85.7	47.1	11.92	.001
Insurance (% Medicaid)	100.0	54.3	16.99	.001
Primary diagnosis			13.40	.037
ADHD/DBD (%)	38.1	11.6
Psychotic (%)	0	1.4
Depressive (%)	33.3	41.3
Mood/bipolar (%)	14.3	20.0
Anxiety (%)	9.5	15.4
Adjustment (%)	0	5.5
Other (%)	4.8	4.8
Length of stay (M(SD))	14.81 (8.18)	10.72 (8.66)	2.12	.035

ADHD/DBD: attention-deficit/hyperactivity disorder and disruptive behavior disorders; PRN: pro re nata medication; SR: seclusion/restraint.

A total of 109 SR events occurred during the study period, comprising 102 locked door seclusions, four open door seclusions, and three restraints. After the complete implementation of M-PBIS, there was a marked reduction in the use of SR. The mean seclusion rate (seclusion hours per thousand patients hours) significantly decreased from 3.56 (SD = 1.94) to 0.48 (SD = 0.64) (see statistics in Table 2)—an 86.52% decrease. Figure 2 depicts the monthly seclusion rate across the study period. The percent of patients who were in SR significantly decreased from 47.6 to 6.7% (χ² = 42.60, p < .001). A total of 441 PRN medications were given during the study period. After the complete implementation of M-PBIS, there was a reduction in the use of PRNs. The percent of patient who received a PRN significantly decreased from 33.3 to 12.9% (χ² = 7.04, p = .008). Given the difference in demographic and clinical characteristics between the pre- and postintervention groups, two logistics regressions were conducted with (1) SR or (2) PRN use as the dependent variables, and age, race, insurance, length of stay, primary discharge diagnosis included as independent variables, and intervention. Controlling for demographic and clinical characteristics, the intervention group remained significant for seclusion (B = 2.50, SE = 0.68, OR = 12.22, 95% CIs (3.22, 46.34), p = .001) and approached significant for PRN use (B = 1.03, SE = 0.53, OR = 2.80, 95% CIs (0.99, 7.92), p = .052).

OPEN IN VIEWER

Figure 2.

M-PBIS passport.

OPEN IN VIEWER

Table 2.

Seclusion/restraint and use of PRNs pre- and postintervention in 442 partial hospital admissions.

Variable	Preintervention (n = 21)	Postintervention (n = 421)	t/χ2	p
Any SR (%)	47.6	6.7	42.60	.001
SR rate (M(SD))	3.56 (1.94)	0.48 (0.64)	5.69	.001
Any PRN (%)	33.3	12.9	7.04	.008

PRN: pro re nata medication; SR: seclusion/restraint.

Postintervention, staff members approximated the goal of at least five positive interactions to one negative interaction with patients. During observations, staff provided on average 4.22 ± 5.11 praising comments, 0.41 ± 0.80 corrective comments, and 0.24 ± 0.59 critical comments. Furthermore, staff members generally awarded stamps appropriately (on average 4.19 ± 6.79 stamps given during the observation), that is contingent on behavior (2.37 ± 3.82), with labeled praise (2.26 ± 3.66), and within 2 s of identified behaviors (2.12 ± 3.65).

Discussion

This naturalistic study of M-PBIS in a pediatric partial hospital program shows a marked reduction in the use of SR and PRN medications for agitation/aggression in a youth psychiatric partial hospital service setting following the implementation of M-PBIS. There were significant reductions in the percent of patients receiving this level of restrictive intervention as well as the overall seclusion rate. Findings remained after controlling for demographic and clinical differences pre- and postintervention. A critical point to highlight is the accompanying reduction in the use of PRN medications for agitation/aggression. Prior work has highlighted the concern that the reduction in SR will lead to an increase in PRN medication use. ²¹ Not only did PRN use not increase following M-PBIS implementation but there was a significant reduction in its use.

These findings speak to the utility of M-PBIS in reducing multiple forms of restrictive interventions which accordingly serve to reduce potential for emotional and physical harm to patients and possibly improve the atmosphere on the unit and experience for staff. This study builds upon prior published work ¹⁸ demonstrating the utility of M-PBIS on a youth psychiatric inpatient setting by demonstrating the positive impact of this approach in a partial hospital setting.

M-PBIS has a number of strengths to highlight. First, it is relatively inexpensive and does not require extensive staff training. It does require a designated consultant who possesses the prerequisite skills associated with the basic tenets of PBIS to provide ongoing training, consultation, and data analysis. These M-PBIS characteristics support the potential for portability to other psychiatric inpatient and partial hospital program units. Next, M-PBIS is not a one-size-fits-all approach. Instead, it provides a core components framework and strong theoretical grounding from which services can operate. Core components are a defined set of positive behavioral expectations, a system to teach behavioral expectations, reinforcement of appropriate behaviors, data collection/evidence-based decision making, individual support for those not responding to the unit-wide system, active support by all stakeholders (including all staff members agreeing to the program), and positive recognition among staff.

The strengths of the approach need to be considered in light of the study’s limitations. First, the study lacks much data on patient characteristics and SR practices prior to implementation of M-PBIS. In addition, there are a number of shortcomings related to the unmasked, pre–post design, including additional confounders, Hawthorne effect, and secular trends. Furthermore, the study would have been strengthened by the inclusion of patient outcomes, perceptions of staff, children, and patients’ families concerning the implementation of the new model of care, and staff stress level and morale. Our impression is that reduced use of SR and PRN medications generally improved staff morale and the work environment, but this was not quantified. Finally, it will be critical to assess the utility of M-PBIS on reducing SR in other hospital settings, allowing for differences in clinical care practices, staffing challenges, and population served. The data call for a future randomized clinical trial to more formally test the efficacy and effectiveness of M-PBIS in other youth psychiatric inpatient settings.

Despite these limitations, the current study provides preliminary evidence that M-PBIS is a promising intervention to use in youth psychiatric partial hospital service to reduce SR and PRN medication use. Reduction and ultimate elimination of SR is a critical objective in improving the delivery of acute care mental health services for youth. The attainment of this objective will follow federal and state health department mandates to reduce and eliminate SR in psychiatric settings, a challenge which is yet unmet. ²⁰