Evaluation of a Brief Intervention to Improve Engagement in Early Intervention Services for Psychosis

Background and rationale

For young adults experiencing early phase psychosis, interactions with health services during initial help seeking are believed to be crucial in framing future attitudes towards healthcare and engagement in subsequent treatment. Young adults experiencing psychosis often encounter significant adversity during initial help-seeking. Delays to accessing suitable early intervention treatment are common, and most young adults are referred to early intervention services for psychosis (EIS) via emergency and inpatient services while experiencing acute psychosis. These referral experiences can involve law enforcement and involuntary hospitalization which are highly distressing for young adults, especially in the context of first contact with services.

In general psychiatric samples (i.e., across diagnostic groups, ages and phases of illness), prior involuntary hospitalization is associated with negative views towards subsequent treatment with views of treatment as punitive and coercive rather than therapeutic, and feelings of betrayal towards healthcare. In addition, in young adults experiencing early phase psychosis with prior hospitalization, the prevalence of PTSD diagnosis is twice that of individuals without prior hospitalization with research attributing this increased prevalence to adversity specific to individuals' prior experiences of healthcare. Individuals experiencing early phase psychosis with comorbid PTSD, often display diminished recovery and engagement in treatment. While these outcomes alone are undesirable, less is known about the relationship to between adverse experiences with healthcare and future engagement in EIS.

Recent research at the Nova Scotia Early Psychosis Program (NSEPP), has shown a relationship between referral source to EIS and subsequent health care utilization: when controlling for symptom severity and functioning, participants referred to NSEPP from urgent services (emergency, inpatient and urgent care), compared to those referred from GPs, had a greater than 3-fold use of emergency and inpatient services 2 years following their acceptance to the NSEPP. A related qualitative study further linked service users' help-seeking and referral experiences to their perceptions towards EIS and engagement in treatment; participants with previous experiences of inpatient hospitalization especially those experiencing involuntary hospitalization or interactions with law enforcement, described increased adversity during healthcare interactions, as well as diminished trust towards treatment providers, confidence in the benefits of treatment, and engagement in treatment. Furthermore, participants often cited past negative experiences with healthcare as explanations for these outcomes. These results taken together suggest that many young adults referred to EIS, especially those referred from urgent services, are at risk for negative clinical outcomes including relapse which may be influenced by intermediate outcomes such as negative perceptions of EIS and decreased engagement in treatment. As engagement in treatment is associated with improved symptomatic and functional outcomes, this available research highlights the need for additional support to be provided to new patients in EIS in order to address potential negative perceptions towards healthcare and strengthen engagement in treatment which may predict long term clinical outcomes.

A recent trial of a brief engagement intervention called Just Do You, showed promise in addressing barriers to engagement and recovery for young adults with serious mental illness (SMI) and may be useful at the NSEPP. This manualized intervention aimed to strengthen engagement in treatment by facilitating the processing of prior negative experiences with healthcare, empowering participants to take leadership in recovery, thus increasing participants "buy-in" towards treatment. Additionally, the Just Do You intervention was initially designed to be 2 sessions in length. While more intensive psychotherapeutic interventions may provide additional benefit, these are often more resource intensive and associated with increased difficulty with recruitment and maintaining participant engagement. Considering these challenges, a briefer intervention focused on bolstering engagement may be more feasible and cost-effective for a wider population.

Despite the promising results of this initial trial and potential utility of this intervention, key differences between the NSEPP and the services where Just Do You was implemented may detract from the suitability of this intervention in its current form. Firstly, this project was conducted in a different treatment setting (New York) and in a general young adult psychiatric sample (which included but was not exclusive to those experiencing early phase psychosis), who were likely to have less severe illness on average and differing treatment needs compared to an early phase psychosis population. In addition, as a facilitator of engagement during sessions, the Just Do You intervention incorporated elements of music and art therapy modalities. While these appeared efficacious in the initial study, art and music therapy are specialized modalities that most settings (including the NSEPP) do not have resources or expertise to properly implement. As a result, the suitability/feasibility of this intervention for early psychosis populations (young adults) in Canada/Nova Scotia is unclear. This project aims to implement an adapted version of the Just Do You intervention at the NSEPP, to strengthen engagement in treatment for recent referrals.

Study Objectives

In fulfillment of these aims, this project has 6 associated objectives described below:

Primary Objectives Objective 1: Engage patient partners in adapting the Just Do You intervention to ensure its suitability and successful delivery at the NSEPP.

Objective 2: Deliver tailored engagement intervention to recently accepted patients at the NSEPP.

Objective 3: Evaluate the effectiveness of this intervention at improving patient reported measures of treatment engagement, recovery, and insight.

Secondary Objectives

Objective 4: Explore trends in participants' administrative health outcomes as a consequence of intervention, including number of missed appointments, emergency room visits, and hospitalization for psychosis relapse.

Objective 5: Compare trends in administrative outcomes following intervention for patients referred to EIS from urgent services to those referred to EIS from all other sources.

Hypotheses

Based on past research conducted by at the Nova Scotia Early Psychosis Program and results from evaluations of the Just Do You intervention, The investigators hypothesize that for participants in the treatment group, this intervention will lead to i) significant improvement in treatment engagement and recovery over the 3-month follow-up period. Furthermore, the investigators hypothesize that participation in the brief intervention will lead to significantly greater decreases in internalized stigma compared to non-participation over follow-up. Additionally, over 6 months from admission to EIS, it is hypothesized that participants who receive the intervention will display: iii) reduction in missed appointments iv) reduction in number of emergency room visits, and inpatient hospitalizations and v) greater improvements in engagement, recovery, and health outcomes, as measured by effect size, for those referred to EIS from urgent services compared to referrals from other sources following a sub-group analysis.

Participant selection

All participants will be individuals aged 19 to 35 years old who are receiving services at NSEPP located at the Abbie J. Lane hospital, QEII Health Sciences Centre. All participants will have a diagnosis of a psychotic disorder using the DSM-5 criteria for schizophrenia spectrum and other psychotic disorders (e.g., schizophrenia, schizoaffective disorder, schizophreniform disorder, unspecified schizophrenia spectrum and other psychotic disorder, etc. Psychotic disorders must have been diagnosed in the last 5 years to be eligible to receive services at NSEPP; only those who have been accepted to NSEPP will be recruited.

For the current study, participants will be recruited within the first 3-12 months of treatment at the NSEPP. The Just Do You engagement intervention was initially designed to be delivered within the first 4-weeks of participants' treatment in order to maximize the potential long-term benefits of participants' strengthened engagement in treatment. However, concerns have been raised that recruitment within the first four weeks of treatment at NSEPP may not be feasible with an early phase psychosis population who are likely to be experiencing more severe illness compared to those in the original Just Do You trial. As a result, the proposed recruitment timeline was determined in order to ensure recruitment procedures of the current study are as similar to those of the initial Just Do You intervention as is feasible. While this recruitment timeline differs slightly from that of the original Just Do You trial, results from this study support potential benefit of the engagement intervention when delivered after the initial 4-weeks of treatment. Indeed, authors reported that participants who received this intervention after the initial period (50% of overall sample) showed no differences in improvement on outcomes compared to those who received the intervention prior. As a result, it was argued that this intervention may provide benefit even when delivered at various times in participants' treatment.

The investigators will be limiting study participation to those who are aged 19-35. Most early psychosis programs across the country have an age cap of 35 years old, so this restriction is in line with current clinical practices. The investigators aim to recruit 40 participants (20 controls and 20 intervention participants) into the study. The investigators do not anticipate any difficulties with this recruitment strategy as there are currently over 250 active patients at the NSEPP and the program accepts 50-70 new patients to the program each year. The principal investigator, a PhD student in clinical psychology, or a research assistant, will contact the potential participants via telephone, or in person at the NSEPP to provide information about the study at that time, and determine whether participants meet criteria for participation.

Research Plan

Objective 1: Engaging patient partners in adapting the Just Do You intervention.

The initial phase of this project involves active collaboration with patient partners in adapting the Just Do You intervention for suitable use in an early phase psychosis sample. Adaptation of the existing intervention (e.g., choosing suitable resources/exercises to facilitate discussion around past treatment experiences) will involve round table collaborative discussions with study personnel, including the peer support worker, and two members of the EPINS First Voice Advisory Group. Discussions will first be centered around highlighting specific components of the initial Just Do You intervention and discussing intervention acceptability and feasibility of delivery. As concerns may arise, patient partners and study personnel will work collaboratively to modify intervention components to support intervention acceptability while ensuring modifications are consistent with theoretical and empirical underpinnings of the intervention. During initial adaptation stages, emphasis will be placed on actively engaging members of the EPINS First Voice Advisory Group who have had negative experiences with healthcare prior to engaging in EIS to ensure participants' concerns are considered when adapting the Just Do You intervention. It is anticipated that initial intervention adaptation will require a 9-12-hour commitment of each representative over 3, 3- 4-hour initial meetings.

Following adaptation of the intervention, eligible clinicians and the peer support worker will be trained on delivering the intervention by a clinical psychology Ph.D. student, and a licensed clinical psychologist, both of whom have experience working with this population and will be involved in initial adaptation of the Just Do You intervention. Training will involve a full-day workshop lead by investigators which will include presentation of the intervention, treatment targets, underlying theory, and role-play in delivering the intervention with clinicians and PSW. Following training, the investigators will begin delivery of the modified version of the Just Do You intervention to patients at the NSEPP.

During implementation and delivery phase of the project, study personnel will continue to meet with patient representatives once every two months at project steering committee meetings. In these meetings patient representatives will be informed of any study updates and will collaborate with study investigators in finding solutions to encountered challenges including difficulties with intervention delivery or study recruitment. Additionally, when applicable, in steering committee meetings, patient partners and co-investigators will collaborate on reviewing and preparing relevant study material including ethics amendments, consent forms, and materials summarizing study findings to be shared with a broader audience. As this project is anticipated to take 2-years to complete, patient representatives will be required to attend 12, 1-hour meetings with investigators in addition to the initial intervention adaptation meetings.

Throughout patient engagement activities in this project, The Public and Patient Engagement Evaluation Tool (PPEET) will be administered regularly (once after each meeting) to allow patient partners to provide regular feedback to research staff. The PPEET is an evidence-based brief survey measure designed to gain patient partners' assessments of engagement activities throughout the project. Results from this measure will allow investigators to assess degree to which patient partners believe meaningful engagement in project processes has been achieved and results will be used on an ongoing basis to improve engagement processes during the study.

Objective's 2 and 3: Delivering and evaluating the engagement intervention.

Following adaptation of Just Do You, delivery and evaluation of the intervention will begin. A repeated measures quasi-experimental design will be used to evaluate the engagement intervention's effectiveness. This design is most suitable as randomization of participants to treatment or control may be unethical given existing evidence supporting the intervention's effectiveness. Consequently, the investigators aim to approach all eligible patients within the first 3-12 months of treatment who have previously consented to being contacted for research purposes (roughly 80% of overall patient population). Eligible patients will be informed of the study protocol/rationale, as well as potential risks and benefits of participation in the intervention. Following this, eligible participants will choose between i) full participation in the treatment group, which involves engaging in the 2-part treatment intervention and completing study outcome assessments ii) participation in the control group which involves completing outcome assessments during regular clinical visits without engaging in the intervention and iii) refraining from participating in the study. Following this, those expressing interest in participating in either the treatment or control conditions will engage in a more in-depth consent discussion where additional specifics of participation will be outlined including risks/benefits, confidentiality, and the study protocol. Individuals providing informed consent to participate in the study will then complete a baseline assessment. This assessment includes completion of a demographic form which will include questions about participants' age, ethnicity, migrant status, gender identity, and sexual orientation , as well as questions about participants' past experiences with mental healthcare and referral source. In addition, participants postal code will be collected and used to extrapolate an estimate of participants' socioeconomic status (SES) using The Canadian Index of Multiple Deprivation (CIMD). This index developed by Statistics Canada, uses postal code data to estimate individuals' SES using participants' geographical location on four dimensions: Residential instability, Economic dependency, Ethno-cultural composition, and Situational vulnerability. These dimensions based on available evidence and up-to date census data, and are widely used in Canadian health research as proxy multi-dimensional estimates of SES. Collection of demographic data will allow investigators to better describe the participant sample and potentially allow for exploration of potential differential outcomes in minority communities.

Participants will then complete a baseline clinical assessment which includes measures of engagement in treatment (Yatchmenoff Client Engagement Scale), self-reported recovery (Recovery Assessment Scale), psychotic symptom severity (PANSS-6), global impression of illness severity Clinical (CGI-I and -S) Social and Occupational Functioning (SOFAS), illness insight (SUMD), current substance use (WHO-ASSIST FC), experiences of adverse and traumatic life events (TALE), and self/internalized stigma (ISMI). These measures have all been validated and previously used in early psychosis populations. The engagement and recovery measures were previously employed by investigators in the initial Just Do You trial and have been chosen to ensure comparability between studies.

Intervention Procedure and Follow-up

Following baseline assessment, participants receiving the intervention will engage in two intervention modules (between 90-120 minutes each) led by a clinician and PSW. This intervention is based on the "young adult framework" for mental health and is informed by theories of health behaviour change and formal decision theory. Targets of these sessions are common cognitive and affective barriers to engagement in mental health services for young adults, especially those with past negative experiences with healthcare (e.g., behavioural beliefs, image impressions, emotional reactions, personal efficacy, norms). Following outlined practices, and to reduce burden of participation, participants will engage in the first module immediately following baseline assessment. Module 1 will focus on discussions of the recovery process and principles of recovery while exploring stigma and participants' perceptions and attitudes towards clinical service. Module 2 , which will be delivered 4-weeks following the first session, will focus on processing prior difficult experiences with healthcare and psychosis, and resulting perceptions towards treatment that commonly act as barriers to engagement. Emphasis will be placed on instilling participants hope in recovery and building trust in current EIS.

As the NSEPP largely provides care in person, and study sessions are designed to coincide with participants regular treatment sessions at NSEPP, Participants will be encouraged to complete the intervention modules in person. However, in order to increase accessibility to the intervention for participants who may have increased difficulty travelling to NSEPP, participants will be given the option to attend sessions virtually through Zoom for Healthcare. This interface has been carefully vetted by NS health for use with patients and is deemed to be the most secure virtual method to ensure patient/participant privacy and security. In addition to a baseline assessment, treatment and control groups will complete follow up measures of participants' self reported recovery and engagement outcomes 2 and 4-weeks following initial study enrollment. Finally, participants in the treatment group will complete assessment measures 3-months post-delivery of second session, and control participants will complete measures 4-months following enrollment. At final assessment, participants' PANSS-6, CGI-I, SOFAS, SUMD, WHO-ASSIST and ISMI-9 scores will be measured in addition to participants' recovery (RAS), and engagement scores (Yatchmenoff client engagement scale). Based on preference, participants will be provided the option to complete self report measures at follow-up times either in person at the NSEPP, or virtually. Those electing to complete follow-up measures through REDCap, will be invited by email to complete the web-based self-report measures. De-identified data will be automatically entered into the secure web-based REDCap database capture platform with only participant ID approved by NS Health and managed in accordance with provincial privacy standards. For participants electing to complete measures virtually, clinician rated assessments will be completed in person at the NSEPP or completed virtually via Zoom for healthcare for NSHealth. Completion of the above protocol will translate to a total length of follow up of 4 months for all participants

Objective's 4 and 5: Monitoring administrative health outcomes:

In addition to evaluating effectiveness of the intervention by administering the above outcomes, administrative health data will be used to explore intervention impact on participants high level clinical outcomes. Firstly, to obtain an additional objective indicator of participants' treatment engagement, the investigators will measure participants' rate of missed appointments 3-months prior and 6-months following intervention. In addition, the investigators will measure rate of hospitalization and ER visits over the same time (cumulative # of visits/person-year). As the investigators have prior ethics approval to monitor high level demographic and outcome data for all NSEPP patients (including those accepted to treatment who never engaged in EIS), the investigators will use all past and present patients who did not receive the intervention as a comparison group and measure participants' administrative health outcomes over the same timeline. Finally, outcomes for participants referred to NSEPP from urgent services will be compared to those of participants referred from other services to assess differential effectiveness of the intervention in previously identified high-risk groups.

Data analysis plan

The goal of this project is to adapt the Just Do You intervention for use at NSEPP and deliver and evaluate the effectiveness of this intervention. In accordance with the stated objectives and consistent with the initial Just Do You trial, the investigators will measure changes in participants' engagement, recovery, insight, stigma and symptoms over follow-up and compare changes on these outcomes between control and active participants. To address this first hypothesis - that participants who engaged in the intervention will show greater improvements in engagement, self-reported recovery, and insight, compared to those in the control condition, the investigators will use linear mixed regression models which are most suitable for repeated measures designs. Potential differences in overall illness severity (CGI-S, PANSS-6, SOFAS, WHO-ASSIST) between control and treatment groups at baseline will be assessed using independent t-tests. If significant differences on these measures are observed between groups, baseline scores on these outcomes will be included as predictors in analyses, in addition to treatment group. The investigators aim to recruit 40 participants (20 controls and 20 intervention participants) into the study. This sample size was calculated to ensure 80% power of detecting the minimum effect size previously reported in the trial of the Just Do You intervention (Cohen's D ranged from 0.50 to 1.15). This sample is feasible given the clinic's large patient pool (250+ active patients, 50 -70 new admissions/year) and the nature of research at NSEPP being embedded into clinical practice. To measure impact of the intervention on participants' self reported engagement, recovery, and insight, the investigators will use linear mixed regression models which are suitable for repeated measures designs.

To address the second and third hypotheses (in fulfillment of study objectives 4 and 5) - that those receiving the study intervention will show a trend of fewer overall emergency department visits and hospitalizations for mental health reasons, as well as fewer missed treatment appointments compared to those who received treatment as usual, Poisson regression models will be used. To assess for trends in rate of missed appointments, hospitalization, and ER visits and to compare potential differential outcomes for those referred to EIS via urgent or non-urgent sources, both treatment group and referral source (i.e., urgent vs non-urgent referral will be included as predictors in analyses using time of follow up as an offset variable in analysis. If after reviewing the distribution of hospitalizations, Poisson regression does not fit the data well, (i.e., data are over dispersed or are characterized by considerable proportion of sample not having urgent care visits) modified Poisson regression can be used: If the sample variance is significantly greater than the mean number of outcomes, negative binomial regression will be used to account for this extra variance. If data are skewed towards zero hospitalizations, A zero inflated Poisson model will be used.

Due to the small sample size (20 active participants and 20 controls, with potential to include roughly 500 additional retrospective controls) the investigators may not be adequately powered to detect small differences in administrative health outcomes between participant groups or include additional important demographic and clinical variables in analyses. Nonetheless, as this is a pilot project, the investigators will assess and to report observed trends in health outcomes which are anticipated to act as preliminary "proof of concept" findings supporting potential subsequent follow-up studies. It is the investigators aim that these initial study findings will facilitate potential follow-up studies evaluating this intervention that are adequately powered to detect differences in administrative health outcomes while control for additional potential confounders.

To address the fourth hypothesis - That participant's with past urgent service referral will benefit more from the intervention than those referred from other sources (e.g., GP, family/self Community mental health), Subgroup comparisons will be conducted. The overall effect size, attributable to the intervention, measured with Cohen's D, will be compared between participants with urgent vs non urgent referral to EIS.

Confidentiality

The principal investigator will keep a file that links participant ID numbers to names that will be password-protected and stored on a computer that is housed at the Abbie J Lane building. The only location where identifying information will be kept will be on the secure computer at NSEPP; this information will be kept in a password-protected file and will be accessible only to the principal investigator, and supervising investigators, if necessary; Research assistants and the peer support worker will have access to the de-identified file that contains ID numbers and the remainder of participant information (e.g., outcomes). In addition to being in a locked room in the NSEPP. The computer on which files will be stored is also encrypted, to prevent a breach of confidentiality should the computer be lost or stolen. Potential participants' contact information and information regarding date of acceptance to the NSEPP will be taken from the NSEPP clinic database of those participants who gave prior consent to the clinic to release this information for research purposes, limiting the need to gain access to the potential participant's health record. Completed measures, either paper and pencil or virtually through REDCap will be stored only with the date and unique participant ID number; names or other identifying information will not be used. Paper questionnaires will be kept separately each in a separate locked filing cabinet in a locked room with limited access. Participant ID number and identifying information will be linked only through the password-protected file located on the principal investigator's computer housed at the NSEPP. Consent forms will be stored under lock and key in the principal investigator's office at Abbie J Lane, as will study data (stored in a separate locked filing cabinet). Seven years following the publication of the study results, all data will be destroyed/permanently erased, as required by the Nova Scotia Health Authority. The records on the principal investigator's computer will be permanently erased by the principal investigator with the assistance of IT, and the physical records at the Abbie J Lane hospital will also be destroyed by the PI.