Australasian Psychiatry http://apy.sagepub.com/ Use of restrictive interventions in a child and adolescent inpatient unit − predictors of use and effect on patient outcomes Suzanne G Duke, James Scott and Angela J Dean Australas Psychiatry 2014 22: 360 originally published online 1 May 2014 DOI: 10.1177/1039856214532298 The online version of this article can be found at: http://apy.sagepub.com/content/22/4/360
Published by: http://www.sagepublications.com
On behalf of:
The Royal Australian and New Zealand College of Psychiatrists
Additional services and information for Australasian Psychiatry can be found at: Email Alerts: http://apy.sagepub.com/cgi/alerts Subscriptions: http://apy.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Jul 21, 2014 OnlineFirst Version of Record - May 1, 2014 What is This?
Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
532298 research-article2014
APY0010.1177/1039856214532298Australasian PsychiatryDuke et al.
Australasian
Psychiatry
In-patient restrictive treatment
Use of restrictive interventions in a child and adolescent inpatient unit – predictors of use and effect on patient outcomes
Australasian Psychiatry 2014, Vol 22(4) 360–365 © The Royal Australian and New Zealand College of Psychiatrists 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1039856214532298 apy.sagepub.com
Suzanne G Duke Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD, Australia James Scott Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD and; Metro North Mental Health, Royal Brisbane and Women’s Hospital, Herston QLD and; Discipline of Psychiatry, The University of Queensland Centre for Clinical Research, Herston QLD, Australia
Angela J Dean Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD 4101 Australia, School of Medicine, The University of Queensland, Herston, QLD, Australia
Abstract Objectives: Restrictive interventions (seclusion, physical restraint, and use of acute/p.r.n. sedation) may have negative effects on patients. Identifying factors associated with use of restrictive interventions and examining their effect on admission outcomes is important for optimising inpatient psychiatric care. Methods: This study documented use of restrictive interventions within a child and adolescent psychiatric inpatient unit for 15 months. Two models examined predictors of use of restrictive interventions: (i) incident characteristics; and (ii) patient characteristics. The relationship between use of restrictive interventions and global clinical outcomes was also examined. Results: Of 134 patients admitted during the study period (61.9% female, mean age=13.8±2.9 years), 26.9% received at least one restrictive intervention. Incident factors associated with restrictive interventions were: physical aggression, early admission stage, and occurrence in private space. Patient factors that predicted use of restrictive interventions were developmental disorder and younger age. Use of restrictive interventions was not associated with increased length of stay or diminished improvement in global symptom ratings. Conclusions: Further research is needed to identify best practice in children at high risk for receiving restrictive interventions. Keywords: seclusion, restraint, aggression, developmental disorders, mental health inpatient units
A
ggressive behaviours are a common reason for referral to mental health services in children and adolescents.1,2 Aggression arising during inpatient admission may threaten the safety of staff and patients, and compromise the therapeutic milieu.2 As such, effective aggression management is an essential component of inpatient care.3 Most guidelines recommend preventive approaches, and use of least restrictive interventions such as de-escalation when aggression does arise.2 In situations where these interventions have been ineffective and the aggression poses a danger to others, restrictive interventions (RIs) may be indicated.2 These include seclusion, physical and mechanical restraint, and use of acute sedative medications (p.r.n. sedation).2
There has been widespread concern about the use of RIs, particularly seclusion and restraint; reducing, or eliminating, use of restraint and seclusion is a key principle of the Australian Government’s ‘National Safety Priorities in Mental Health’ report.4 Seclusion rates are one of the national mental health key performance indicators upon which the quality of mental health service delivery is measured.5 Furthermore, The Australian National
Corresponding author: Dr Angela J Dean, Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD 4101, Australia. Email: [email protected]
360 Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
Duke et al.
Mental Health Commission is leading a project to examine best practice to reduce and eliminate seclusion and restraint in those with mental health problems.6 With such policy emphasis, many services are now introducing policies to reduce use of RIs.7–9 An important component of reducing use of RIs involves identifying which patient groups are most at risk of receiving these interventions. Aggression is considered an important predictor of all types of RI use.10 However, data is conflicting about the role of child factors.11–13 There is increasing recognition that these interventions may exert harm and may be traumatic for staff and patients.14–16 Despite this, no studies have examined the effect of exposure to RIs on admission outcomes in children. This project aimed to: (i) document the nature and frequency of RIs used within a child and adolescent psychiatric inpatient unit; (ii) identify situational predictors of RI use; (iii) identify patient characteristics predicting RI use; and (iv) examine the relationship between RIs and admission outcomes. In particular, we hypothesised that exposure to RIs would impair improvement in admission outcomes.
would provide brief ‘holding’ – a clinical team holding a child still until aggression subsided. This was only used for children younger than 14 years. No mechanical restraints were used in the unit. •• Use of sedative medication for aggression management (‘p.r.n. sedation’): The majority of p.r.n. sedation administered in the unit is oral. In situations where intramuscular p.r.n. sedation is administered, this would also typically involve brief physical restraint by staff. All episodes of seclusion and restraint were included within this evaluation. Seclusion and restraint were rarely used for indications other than aggression. P.r.n. sedation is typically used for a variety of indications. Only use of p.r.n. sedation for management of aggressive behaviour was classified as an RI; use of p.r.n. sedation for indications such as insomnia or anxiety was not classified as an RI. Other incident information Aggressive incidents were documented using a wardbased register. For all incidents, the following information was collected:
Methods Study site
The site was a 10-bed child and adolescent inpatient psychiatric unit based in a metropolitan children’s hospital. The unit provides short to medium duration admissions. Care is provided by a multidisciplinary team of medical, nursing and allied health staff specialising in child and adolescent mental health. Ward policies for management of aggression utilised a comprehensive behavioural management program.17 Based on national and institutional guidelines, this evaluation was conducted as a quality assurance activity. Data from this evaluation was collected after implementation and evaluation of a behavioural management policy.17 This evaluation forms part of a larger initiative within this unit to better understand determinants and impacts of challenging behaviours in children and adolescents, and improve approaches to aggression prevention and management.9,18–20 Restrictive interventions Information on the use of RIs was collected from a wardbased register of aggressive incidents and ward-based registers of RIs, maintained as required by state-based legislation. Interventions recorded were: •• Seclusion: episodes and duration of all locked interventions were recorded. The majority of patients would require a formal escort to enter seclusion. This escort alone was not classified as physical restraint. •• Physical restraint: episodes were recorded. This was defined as a specific technique where clinical teams
•• Time of day: coded as a dichotomous variable of before or after 3:00pm (the time that patients returned to the ward from school). •• Stage of admission: coded as a dichotomous variable reflecting whether the incident occurred during the first week of admission or later. •• Type of aggression: presence of physical aggression towards others (yes/no), verbal abuse (yes/ no), oppositional behaviour (yes/no), or selfharming behaviour (yes/no). It was possible for incidents to receive multiple classifications: for example, many incidents involving physical aggression also involved oppositional behaviour and verbal abuse. •• Incident location: coded as private space (bedroom, bathroom) or public space (hallways, recreation rooms, school, time-out room). Patient information Patient medical charts were reviewed to collect additional information: demographic details, duration of admission, and clinical details (diagnoses, history of abuse, suicidality (ideation or attempt), deliberate selfharm, and use of medications). Diagnostic assessments were undertaken prospectively during admission by child psychiatrists. For analysis, key diagnoses were grouped as developmental disorders (pervasive developmental disorder or mental retardation), externalising disorders (attention deficit or disruptive behaviour
361 Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
Australasian Psychiatry 22(4)
disorders (ADHD/DBD)) and internalising disorders (mood or anxiety disorders). Clinicians completed the Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA) at admission and discharge, as part of routine outcome monitoring. The HoNOSCA is a brief, clinician-rated measure of global mental health severity.21 If a patient experienced more than one admission during the study period, the first admission with completed outcome data was selected for inclusion in the analysis. Statistical analysis RI was defined as at least one episode of seclusion, restraint, or acute sedative medication for aggression. Two models identifying factors associated with use of RIs were developed: 1. Incident characteristics: time of day, stage of admission, types of behaviour and location. Aggressive incidents for which RIs were used were compared with aggressive incidents not managed with RIs. 2. Patient characteristics: demographics, diagnostic and clinical characteristics. Patients who received RIs during their admission were compared with patients not receiving RIs. Each model utilised initial bivariate comparisons: independent samples t-tests for continuous variables (e.g. age) and chi square tests for categorical variables (e.g. diagnoses, stage of admission). Variables significantly associated with use of RIs were then modelled with logistic regression. The relationship between exposure to RIs and HoNOSCA scores over time was analysed using repeated measures analysis of variance, with the between-groups factor being group (RI during admission, yes/no) and the within-groups factor being time (admission, discharge). Analysis examined overall change in scores between admission and discharge, and change over time between groups. For all analyses, alpha was set at 0.05.
Results
Patient characteristics During the study period, 134 patients were admitted (mean age 13.8 years; SD 2.9: range 5.8–18). The majority were female (83/134: 61.9%). Average admission duration was 21.2 days (SD 25.6: median 12.0: range 1–175), and 15.9% of patients had experienced a prior admission to the unit (21/134). Most common diagnoses were mood disorders (39/134: 29.1%), anxiety disorders (29/134: 21.6%), adjustment disorders (23/134; 17.2%), ADHD/ DBD (22/134; 16.4%), pervasive developmental disorders (22/134; 16.4%), and eating disorders (12/134; 9.0%).
Only five young people were diagnosed with a psychotic disorder (3.7%). More than half of patients (72/134; 53.7%) received more than one diagnosis: 51 patients received two diagnoses (38.1%), 15 received three diagnoses (11.2%) and six patients received four diagnoses (4.5%). Frequency of RIs Documentation was available for 342 aggressive incidents. RIs were used in 180 incidents (52.6%). The most commonly used RI was seclusion (157/180; 87.2%), followed by p.r.n, sedation (18/180; 10.0%) and physical restraint (5/180; 2.8%). Mean duration of seclusion was 18.8 minutes (SD 14.0; Range 2.0–75.0). In situations where p.r.n. sedation was administered, only one incident involved parenteral (intramuscular) administration; the remaining episodes involved oral administration of p.r.n. sedation. Predictors of RIs – incident characteristics Aggressive incidents where a RI was used were compared with aggressive incidents where RIs were not used. Bivariate comparisons indicated that RIs were more likely to be used in the first week of admission (χ2=14.86; p
Published by: http://www.sagepublications.com
On behalf of:
The Royal Australian and New Zealand College of Psychiatrists
Additional services and information for Australasian Psychiatry can be found at: Email Alerts: http://apy.sagepub.com/cgi/alerts Subscriptions: http://apy.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav
>> Version of Record - Jul 21, 2014 OnlineFirst Version of Record - May 1, 2014 What is This?
Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
532298 research-article2014
APY0010.1177/1039856214532298Australasian PsychiatryDuke et al.
Australasian
Psychiatry
In-patient restrictive treatment
Use of restrictive interventions in a child and adolescent inpatient unit – predictors of use and effect on patient outcomes
Australasian Psychiatry 2014, Vol 22(4) 360–365 © The Royal Australian and New Zealand College of Psychiatrists 2014 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/1039856214532298 apy.sagepub.com
Suzanne G Duke Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD, Australia James Scott Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD and; Metro North Mental Health, Royal Brisbane and Women’s Hospital, Herston QLD and; Discipline of Psychiatry, The University of Queensland Centre for Clinical Research, Herston QLD, Australia
Angela J Dean Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD 4101 Australia, School of Medicine, The University of Queensland, Herston, QLD, Australia
Abstract Objectives: Restrictive interventions (seclusion, physical restraint, and use of acute/p.r.n. sedation) may have negative effects on patients. Identifying factors associated with use of restrictive interventions and examining their effect on admission outcomes is important for optimising inpatient psychiatric care. Methods: This study documented use of restrictive interventions within a child and adolescent psychiatric inpatient unit for 15 months. Two models examined predictors of use of restrictive interventions: (i) incident characteristics; and (ii) patient characteristics. The relationship between use of restrictive interventions and global clinical outcomes was also examined. Results: Of 134 patients admitted during the study period (61.9% female, mean age=13.8±2.9 years), 26.9% received at least one restrictive intervention. Incident factors associated with restrictive interventions were: physical aggression, early admission stage, and occurrence in private space. Patient factors that predicted use of restrictive interventions were developmental disorder and younger age. Use of restrictive interventions was not associated with increased length of stay or diminished improvement in global symptom ratings. Conclusions: Further research is needed to identify best practice in children at high risk for receiving restrictive interventions. Keywords: seclusion, restraint, aggression, developmental disorders, mental health inpatient units
A
ggressive behaviours are a common reason for referral to mental health services in children and adolescents.1,2 Aggression arising during inpatient admission may threaten the safety of staff and patients, and compromise the therapeutic milieu.2 As such, effective aggression management is an essential component of inpatient care.3 Most guidelines recommend preventive approaches, and use of least restrictive interventions such as de-escalation when aggression does arise.2 In situations where these interventions have been ineffective and the aggression poses a danger to others, restrictive interventions (RIs) may be indicated.2 These include seclusion, physical and mechanical restraint, and use of acute sedative medications (p.r.n. sedation).2
There has been widespread concern about the use of RIs, particularly seclusion and restraint; reducing, or eliminating, use of restraint and seclusion is a key principle of the Australian Government’s ‘National Safety Priorities in Mental Health’ report.4 Seclusion rates are one of the national mental health key performance indicators upon which the quality of mental health service delivery is measured.5 Furthermore, The Australian National
Corresponding author: Dr Angela J Dean, Kids in Mind Research, Mater Research, Mater Health Services, South Brisbane, QLD 4101, Australia. Email: [email protected]
360 Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
Duke et al.
Mental Health Commission is leading a project to examine best practice to reduce and eliminate seclusion and restraint in those with mental health problems.6 With such policy emphasis, many services are now introducing policies to reduce use of RIs.7–9 An important component of reducing use of RIs involves identifying which patient groups are most at risk of receiving these interventions. Aggression is considered an important predictor of all types of RI use.10 However, data is conflicting about the role of child factors.11–13 There is increasing recognition that these interventions may exert harm and may be traumatic for staff and patients.14–16 Despite this, no studies have examined the effect of exposure to RIs on admission outcomes in children. This project aimed to: (i) document the nature and frequency of RIs used within a child and adolescent psychiatric inpatient unit; (ii) identify situational predictors of RI use; (iii) identify patient characteristics predicting RI use; and (iv) examine the relationship between RIs and admission outcomes. In particular, we hypothesised that exposure to RIs would impair improvement in admission outcomes.
would provide brief ‘holding’ – a clinical team holding a child still until aggression subsided. This was only used for children younger than 14 years. No mechanical restraints were used in the unit. •• Use of sedative medication for aggression management (‘p.r.n. sedation’): The majority of p.r.n. sedation administered in the unit is oral. In situations where intramuscular p.r.n. sedation is administered, this would also typically involve brief physical restraint by staff. All episodes of seclusion and restraint were included within this evaluation. Seclusion and restraint were rarely used for indications other than aggression. P.r.n. sedation is typically used for a variety of indications. Only use of p.r.n. sedation for management of aggressive behaviour was classified as an RI; use of p.r.n. sedation for indications such as insomnia or anxiety was not classified as an RI. Other incident information Aggressive incidents were documented using a wardbased register. For all incidents, the following information was collected:
Methods Study site
The site was a 10-bed child and adolescent inpatient psychiatric unit based in a metropolitan children’s hospital. The unit provides short to medium duration admissions. Care is provided by a multidisciplinary team of medical, nursing and allied health staff specialising in child and adolescent mental health. Ward policies for management of aggression utilised a comprehensive behavioural management program.17 Based on national and institutional guidelines, this evaluation was conducted as a quality assurance activity. Data from this evaluation was collected after implementation and evaluation of a behavioural management policy.17 This evaluation forms part of a larger initiative within this unit to better understand determinants and impacts of challenging behaviours in children and adolescents, and improve approaches to aggression prevention and management.9,18–20 Restrictive interventions Information on the use of RIs was collected from a wardbased register of aggressive incidents and ward-based registers of RIs, maintained as required by state-based legislation. Interventions recorded were: •• Seclusion: episodes and duration of all locked interventions were recorded. The majority of patients would require a formal escort to enter seclusion. This escort alone was not classified as physical restraint. •• Physical restraint: episodes were recorded. This was defined as a specific technique where clinical teams
•• Time of day: coded as a dichotomous variable of before or after 3:00pm (the time that patients returned to the ward from school). •• Stage of admission: coded as a dichotomous variable reflecting whether the incident occurred during the first week of admission or later. •• Type of aggression: presence of physical aggression towards others (yes/no), verbal abuse (yes/ no), oppositional behaviour (yes/no), or selfharming behaviour (yes/no). It was possible for incidents to receive multiple classifications: for example, many incidents involving physical aggression also involved oppositional behaviour and verbal abuse. •• Incident location: coded as private space (bedroom, bathroom) or public space (hallways, recreation rooms, school, time-out room). Patient information Patient medical charts were reviewed to collect additional information: demographic details, duration of admission, and clinical details (diagnoses, history of abuse, suicidality (ideation or attempt), deliberate selfharm, and use of medications). Diagnostic assessments were undertaken prospectively during admission by child psychiatrists. For analysis, key diagnoses were grouped as developmental disorders (pervasive developmental disorder or mental retardation), externalising disorders (attention deficit or disruptive behaviour
361 Downloaded from apy.sagepub.com at GEORGIAN COURT UNIV on November 18, 2014
Australasian Psychiatry 22(4)
disorders (ADHD/DBD)) and internalising disorders (mood or anxiety disorders). Clinicians completed the Health of the Nation Outcome Scales for Children and Adolescents (HoNOSCA) at admission and discharge, as part of routine outcome monitoring. The HoNOSCA is a brief, clinician-rated measure of global mental health severity.21 If a patient experienced more than one admission during the study period, the first admission with completed outcome data was selected for inclusion in the analysis. Statistical analysis RI was defined as at least one episode of seclusion, restraint, or acute sedative medication for aggression. Two models identifying factors associated with use of RIs were developed: 1. Incident characteristics: time of day, stage of admission, types of behaviour and location. Aggressive incidents for which RIs were used were compared with aggressive incidents not managed with RIs. 2. Patient characteristics: demographics, diagnostic and clinical characteristics. Patients who received RIs during their admission were compared with patients not receiving RIs. Each model utilised initial bivariate comparisons: independent samples t-tests for continuous variables (e.g. age) and chi square tests for categorical variables (e.g. diagnoses, stage of admission). Variables significantly associated with use of RIs were then modelled with logistic regression. The relationship between exposure to RIs and HoNOSCA scores over time was analysed using repeated measures analysis of variance, with the between-groups factor being group (RI during admission, yes/no) and the within-groups factor being time (admission, discharge). Analysis examined overall change in scores between admission and discharge, and change over time between groups. For all analyses, alpha was set at 0.05.
Results
Patient characteristics During the study period, 134 patients were admitted (mean age 13.8 years; SD 2.9: range 5.8–18). The majority were female (83/134: 61.9%). Average admission duration was 21.2 days (SD 25.6: median 12.0: range 1–175), and 15.9% of patients had experienced a prior admission to the unit (21/134). Most common diagnoses were mood disorders (39/134: 29.1%), anxiety disorders (29/134: 21.6%), adjustment disorders (23/134; 17.2%), ADHD/ DBD (22/134; 16.4%), pervasive developmental disorders (22/134; 16.4%), and eating disorders (12/134; 9.0%).
Only five young people were diagnosed with a psychotic disorder (3.7%). More than half of patients (72/134; 53.7%) received more than one diagnosis: 51 patients received two diagnoses (38.1%), 15 received three diagnoses (11.2%) and six patients received four diagnoses (4.5%). Frequency of RIs Documentation was available for 342 aggressive incidents. RIs were used in 180 incidents (52.6%). The most commonly used RI was seclusion (157/180; 87.2%), followed by p.r.n, sedation (18/180; 10.0%) and physical restraint (5/180; 2.8%). Mean duration of seclusion was 18.8 minutes (SD 14.0; Range 2.0–75.0). In situations where p.r.n. sedation was administered, only one incident involved parenteral (intramuscular) administration; the remaining episodes involved oral administration of p.r.n. sedation. Predictors of RIs – incident characteristics Aggressive incidents where a RI was used were compared with aggressive incidents where RIs were not used. Bivariate comparisons indicated that RIs were more likely to be used in the first week of admission (χ2=14.86; p
