Peer support workers with lived experience of homelessness will provide advocacy, supportive counselling, assistance with navigating the system, and role-modeling for people experiencing homelessness in emergency departments in the Niagara region of Ontario, Canada. Previous research has demonstrated that people experiencing homelessness experience barriers to care and poor health outcomes, including increased risk of dying, and are frequent users of the emergency department for physical and mental health needs. The benefits of peer support have been studied in hospital psychiatric units and with people experiencing homelessness, but no studies have yet explored the potential impact of peer support workers on homelessness in the emergency department. The investigators plan to do a 12-month randomized study of peer support workers compared to usual care in the emergency department, and evaluate outcomes such as physical health, mental health, addictions, and behaviours, as well as cost-effectiveness and changes in how people experiencing homelessness seek health care. The investigators hope that the findings of this study will provide additional knowledge and evidence for future efforts to improve care for people experiencing homelessness.
Homelessness is strongly associated with an increased prevalence of chronic physical and mental health morbidity, leading to lower health-related quality of life and increased hospital readmissions compared to the general population. People experiencing homelessness have reported experience of discrimination and unwelcome treatment, neglect and rejection, lack of compassion from their health care providers and feeling as though their perceptions are disregarded. Alongside many daily barriers such as substance use, lack of housing, and food insecurity, these issues also negatively influence the desire to seek health care among people experiencing homelessness. A national strategy to determine the extent and nature of homelessness, 'Point-in-Time Counts' were conducted in over 60 communities across Canada in 2018 to provide a snapshot of the experiences and needs of people experiencing homelessness. The Niagara 'Point-in-Time' Count collected 408 surveys from people experiencing homelessness, with 57 percent of participants reporting a mental health issue, 36.2 percent reporting an addiction, and 33.9 percent reporting a medical condition. Respondents also reported an ongoing need for mental health services (54.3 percent), addiction services (29.2 percent), and medical services (26.9 percent). Of note, 219 participants reporting a collective 663 visits to the emergency department in the preceding 12 months. People experiencing homelessness are much more likely than housed people to access emergency departments for concerns with mental health, although this may not be an ideal environment to provide recovery-based mental health care. The placement of peer support workers in the emergency department may help overcome barriers to health care for people experiencing homelessness. Peer support is defined as a supportive relationship between a worker and a peer who share the commonality of a lived experience, with the peer support worker able to model recovery and use a recovery-oriented, person-centered approach with their peer. Increasing positive interactions in a clinical care setting could address the lack of trust perceived by people experiencing homelessness towards health care providers, helping to counteract negative perceptions of health care and increasing the likelihood that people experiencing homelessness will seek care in the future. The benefits of a peer support worker include feelings of empowerment and hope, increased sense of acceptance and empathy, improved satisfaction, and reduced feelings of stigma. Peer support workers in mental health services lead to reduced admissions, as well as earlier discharge when utilized following admission to hospital. Two recent reviews investigated the effectiveness of peer support services in individuals with severe mental illness. Consistently, they found that peers are at least as effective in providing services as non-peers in the same roles. Additionally, there are certain unique elements that can only be provided by peer support workers. Three randomized controlled trials reported better treatment engagement, fewer hospitalizations and hospital days, and lower rates of non-attendance compared to treatment as usual when peer support workers were added to a care team. When peer support workers served as patient advocates and community connectors for people with multiple psychiatric hospitalizations, the number of hospitalizations and days spent in hospital decreased. Furthermore, participants reported a decrease in depressive symptoms and increase in hope, self-care, and sense of well-being. The researchers noted that the active ingredient in these interventions seemed to be the instillation of hope through self-disclosure, role modeling, and empathy paired with conditional regard. Upon comparing and contrasting the core competencies of case managers and peer support workers, it was noted that the unique duties of peer support workers seemed to center around empowering clients via normalization, participating in personal development as a role model, and encouraging clients' educational growth. However, despite promising findings in a diversity of clinical care settings, there has been relatively limited research and few high quality studies of the effectiveness of peer support workers and their outcomes with people experiencing homelessness. People experiencing homelessness typically have complex comorbid mental illnesses, and they should theoretically benefit from peer support services via the sharing of lived experiences. An exploratory review of the literature on peer support and homelessness found 10 studies totaling 1341 participants that collectively reported improved overall quality of life, significant increases in social support, reduced rates of substance use, decreased homelessness, and better physical and mental health for the recipients of peer support. Although promising, six of the included studies were longitudinal, two were cross-sectional, and two were quasi-experimental. There was also significant heterogeneity in the delivery of peer services across studies, including peer support from mentors, peer support as part of a larger intervention, and peer support groups. Within the emergency department, there has been minimal research on peer support or similar interventions. Of 11 studies examined in a meta-analysis of the effectiveness of emergency department interventions for frequent users, case management was found to be the most frequently used. For people experiencing homelessness, case management demonstrated increased contact with health services, reduced lengths of admission, decreased emergency department visits for mental health problems, and reduced substance use, but only equivocal findings for physical and mental health. An example of an emergency department program that is similar to peer support can be found in a randomized trial of compassionate care for people experiencing homelessness. This study from 1995 found that compassionate contact from trained volunteers led to a 33% relative reduction in the overall frequency of visits per month and a significant reduction in the time until a first repeat visit; however, health outcomes were not evaluated. Although there are several types of emergency department interventions, there is overall a lack of evidence about program effectiveness. Very few studies are randomized controlled trials, most studies with high-risk individuals are of low quality, and stronger evidence on cost-effectiveness is necessary for all program types. Furthermore, the quality of care and health outcomes for people experiencing homelessness may be improved through peer support, but evidence in this area is underdeveloped and in need of studies with robust experimental measures. To fill this gap in knowledge, this study will evaluate a program of peer support for people experiencing homelessness who seek care in the emergency departments of a mid-sized Canadian regional municipality. The research questions for the study are as follows: 1. What, if any, are the changes in health service use (e.g., emergency department visits, hospitalizations) following receipt of peer support in the emergency department? 2. What, if any, are the changes to health status, health-related quality of life, substance use, or behaviours, following engagement with a peer support worker? 3. Is peer support in the emergency department a cost-effective intervention? 4. Does receiving peer support in the emergency department lengthen the interval of time before a clinically significant outcome such as a repeat visit to the emergency department or increased morbidity? 5. What are the mechanisms of peer support in the emergency department that may contribute to differences observed in the outcomes between the intervention and usual care groups?
Study Type
INTERVENTIONAL
Allocation
RANDOMIZED
Purpose
HEALTH_SERVICES_RESEARCH
Masking
DOUBLE
Enrollment
19
Participants randomized to the intervention group will be seen by the peer support worker during their stay in the emergency department. The goal of the peer support worker's practice is to address the participant's wellness from a holistic point of view that includes their relationships, connections to the community, general well-being, and sense of empowerment, with an overall focus on quality-of-life goals (Sunderland et al., 2013). The peer support worker will provide a multi-faceted intervention, which aims to foster a peer-to-peer relationship encompassing elements of empathic communication, brief supportive counseling, advocacy for non-medical needs in hospital, liaising and collaborating with medical and nursing staff, motivational interviewing, and connection with programs in the community.
McMaster University
Hamilton, Ontario, Canada
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in the number of hospitalizations, measured using administrative health data
The primary outcome of the study includes the number of hospitalizations. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in the number of days spent in hospital, measured using administrative health data
The primary outcome of the study includes the number of days spent in hospital. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in the number of emergency department visits, measured using administrative health data
The primary outcome of the study includes the number of emergency department visits. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in the number of hours spent in the emergency department at each visit, measured using administrative health data
The primary outcome of the study includes the number of hours spent in the emergency department at each visit. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in the number of procedures performed in the emergency department, measured using administrative health data
The primary outcome of the study includes the number of procedures performed in the emergency department. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in mean from the pre-intervention 12-month baseline to the post-intervention 12-month period in procedures performed during hospitalizations, measured using administrative health data
The primary outcome of the study includes the number of procedures performed during hospitalizations. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in proportions from the pre-intervention 12-month baseline to the post-intervention 12-month period in diagnoses at the time of discharge from the emergency department, measured using administrative health data
The primary outcome of the study includes the diagnosis at time of discharge from the emergency department. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in proportion from the pre-intervention 12-month baseline to the post-intervention 12-month period in mortality, measured using administrative health data
The primary outcome of the study includes mortality. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in proportion from the pre-intervention 12-month baseline to the post-intervention 12-month period in overdose presentations, measured using administrative health data
The primary outcome of the study includes overdose presentations. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in proportion from the pre-intervention 12-month baseline to the post-intervention 12-month period in psychiatric presentations, measured using administrative health data
The primary outcome of the study includes psychiatric presentations. This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Change in proportion from pre-intervention 12-month baseline to post-intervention 12-month period in composition of presenting complaints to the emergency department, measured using administrative health dat
The primary outcome of the study includes the composition of presenting complaints to the emergency department and potential differences over time in the proportions of different presentations (e.g., physical health vs. mental health, Canadian Triage and Acuity Scale (CTAS) levels). This outcome will be evaluated through linkage and evaluation of administrative data hosted at Niagara Health through Decision Support Services. Consent will be obtained from participants to perform data linkage with Niagara Health databases and collect information on health services use for the 12-month period preceding and 12-month period following study enrollment.
Time frame: 12 months pre-study enrollment and 12-months post-study enrollment
Mean change from baseline in psychiatric symptoms on the Colorado Symptom Index (CSI) at 6 months
The CSI is a 14-item self-report measure of psychiatric symptoms (Boothroyd \& Chen, 2008). Respondents report the frequency of various psychiatric symptoms as experienced over the past month on a 5-point Likert scale. Scores range from 0 to 56, with higher scores indicating report of more frequent psychiatric symptoms. The CSI has been psychometrically validated with a sample of people who were experiencing homelessness, living precariously, at risk for homelessness, or had prior history of homelessness (Conrad et al., 2001). The CSI was found to have excellent internal consistency with a Cronbach's alpha of 0.90 and a test/retest intraclass correlation coefficient of 0.79 with this population. The CSI is intended to evaluate change in persons with moderate and high severity psychological symptoms or emotional distress, and is sensitive to change in symptomatology over time. Administration of the questionnaire requires approximately 5 minutes.
Time frame: 6 months
Mean change from baseline in psychiatric symptoms on the Colorado Symptom Index (CSI) at 12 months
The CSI is a 14-item self-report measure of psychiatric symptoms (Boothroyd \& Chen, 2008). Respondents report the frequency of various psychiatric symptoms as experienced over the past month on a 5-point Likert scale. Scores range from 0 to 56, with higher scores indicating report of more frequent psychiatric symptoms. The CSI has been psychometrically validated with a sample of people who were experiencing homelessness, living precariously, at risk for homelessness, or had prior history of homelessness (Conrad et al., 2001). The CSI was found to have excellent internal consistency with a Cronbach's alpha of 0.90 and a test/retest intraclass correlation coefficient of 0.79 with this population. The CSI is intended to evaluate change in persons with moderate and high severity psychological symptoms or emotional distress, and is sensitive to change in symptomatology over time. Administration of the questionnaire requires approximately 5 minutes.
Time frame: 12 months
Mean change from baseline in psychopathology and crime/violence on the Global Appraisal of Individual Needs - Short Screener (GAIN-SS) at 6 months
The GAIN-SS is a 24-item self-report measure which assesses psychopathology and crime/violence across various dimensions and screens for people who are likely to have diagnoses or require clinical services (Dennis, Chan, \& Funk, 2006). The total score contributes to a Total Disorder Screener (TDScr), with four, 5- to 7-item subset screeners including the Internalizing Disorder Screener (IDScr), Externalizing Disorder Screener (EDScr), Substance Disorder Screener (SDScr), and Crime/Violence Screener (CVScr). The GAIN-SS is internally consistent with a Cronbach's alpha of 0.89 and good discriminant validity for each of the sub-screeners. The Total Disorder Screener has over 90% sensitivity and specificity with a ROC of 0.97 with a cut point of 3 or more. For the sub-screeners, scores can be divided into three groupings of 0 (low), 1-2 (moderate), and 3-5 (high), with the lower cut point of 1+ featuring 90% sensitivity and the upper cut point of 3+ featuring 90% specificity.
This platform is for informational purposes only and does not constitute medical advice. Always consult a qualified healthcare professional.
Time frame: 6 months
Mean change from baseline in psychopathology and crime/violence on the Global Appraisal of Individual Needs - Short Screener (GAIN-SS) at 12 months
The GAIN-SS is a 24-item self-report measure which assesses psychopathology and crime/violence across various dimensions and screens for people who are likely to have diagnoses or require clinical services (Dennis, Chan, \& Funk, 2006). The total score contributes to a Total Disorder Screener (TDScr), with four, 5- to 7-item subset screeners including the Internalizing Disorder Screener (IDScr), Externalizing Disorder Screener (EDScr), Substance Disorder Screener (SDScr), and Crime/Violence Screener (CVScr). The GAIN-SS is internally consistent with a Cronbach's alpha of 0.89 and good discriminant validity for each of the sub-screeners. The Total Disorder Screener has over 90% sensitivity and specificity with a ROC of 0.97 with a cut point of 3 or more. For the sub-screeners, scores can be divided into three groupings of 0 (low), 1-2 (moderate), and 3-5 (high), with the lower cut point of 1+ featuring 90% sensitivity and the upper cut point of 3+ featuring 90% specificity.
Time frame: 12 months
Mean change from baseline in health status on the EuroQol-5D-5L (EQ-5D-5L) at 6 months
The EQ-5D is a 5-item generic measure of health status that provides a descriptive profile for respondents to provide a self-rating of their health and produce a utility score for cost-effectiveness analysis. Health is defined across five dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, which encompass a broad range of different disease areas, specialties, and types of care. Each dimension offers the following five levels: No problems, slight problems, moderate problems, severe problems, and unable to/extreme problems. The EQ-5D-5L offers a total of 3125 health states with a value set for Canada demonstrating a range of health utilities from -0.148 for the worst health state (55555, that is, extreme problems with mobility, self-care, etc.) to 0.949 for the best health state (11111, that is, no problems with any of the dimensions).
Time frame: 6 months
Mean change from baseline in health status on the Visual Analogue Scale (VAS) at 6 months
The Visual Analogue Scale is a 20-cm scale for respondents to provide a self-rating of their health and produce a utility score for cost-effectiveness analysis. This instrument offers a total of 3125 health states with a value set for Canada demonstrating a range of health utilities from -0.148 for the worst health state (55555, that is, extreme problems with mobility, self-care, etc.) to 0.949 for the best health state (11111, that is, no problems with any of the dimensions).
Time frame: 6 months
Mean change from baseline in health status on the EuroQol-5D-5L (EQ-5D-5L) at 12 months
The EQ-5D is a 5-item generic measure of health status that provides a descriptive profile for respondents to provide a self-rating of their health and produce a utility score for cost-effectiveness analysis. Health is defined across five dimensions of mobility, self-care, usual activities, pain/discomfort, and anxiety/depression, which encompass a broad range of different disease areas, specialties, and types of care. Each dimension offers the following five levels: No problems, slight problems, moderate problems, severe problems, and unable to/extreme problems. The EQ-5D-5L offers a total of 3125 health states with a value set for Canada demonstrating a range of health utilities from -0.148 for the worst health state (55555, that is, extreme problems with mobility, self-care, etc.) to 0.949 for the best health state (11111, that is, no problems with any of the dimensions).
Time frame: 12 months
Mean change from baseline in health status on the Visual Analogue Scale (VAS) at 12 months
The Visual Analogue Scale is a 20-cm scale for respondents to provide a self-rating of their health and produce a utility score for cost-effectiveness analysis. This instrument offers a total of 3125 health states with a value set for Canada demonstrating a range of health utilities from -0.148 for the worst health state (55555, that is, extreme problems with mobility, self-care, etc.) to 0.949 for the best health state (11111, that is, no problems with any of the dimensions).
Time frame: 12 months
Mean change from baseline in mental health and physical health on the 12-item Short Form Survey (SF-12) at 6 months
The SF-12 is a 12-item questionnaire that has been investigated and utilized for measuring health status among people experiencing homelessness (Larson, 2002). It features categorical questions that assess limitations in role functioning as a result of physical and emotional health, as well as numerous Likert scale responses including a 3-point scale to assess limitations in physical activity and physical role functioning, a 5-point scale for pain and another for overall health, and a 6-point scale to assess mental health, vitality, and social functioning. The SF-12 produces a summary score each for mental health and physical health, which are able to differentiate between diagnostic groups and levels of severity of illness. Internal consistency in a sample of people experiencing homelessness demonstrated a Cronbach's alpha of 0.82 for physical health and 0.79 for mental health.
Time frame: 6 months
Mean change from baseline in mental health and physical health on the 12-item Short Form Survey (SF-12) at 12 months
The SF-12 is a 12-item questionnaire that has been investigated and utilized for measuring health status among people experiencing homelessness (Larson, 2002). It features categorical questions that assess limitations in role functioning as a result of physical and emotional health, as well as numerous Likert scale responses including a 3-point scale to assess limitations in physical activity and physical role functioning, a 5-point scale for pain and another for overall health, and a 6-point scale to assess mental health, vitality, and social functioning. The SF-12 produces a summary score each for mental health and physical health, which are able to differentiate between diagnostic groups and levels of severity of illness. Internal consistency in a sample of people experiencing homelessness demonstrated a Cronbach's alpha of 0.82 for physical health and 0.79 for mental health.
Time frame: 12 months