Trial of Social Work Services in Fracture Clinic Setting

Each year, approximately 15% of Canadians experience an orthopaedic injury serious enough to affect their ability to engage in everyday activities. These injuries include fractures and dislocations that are typically managed operatively by an orthopaedic trauma surgeon and frequently result in prolonged recovery periods. Despite a comprehensive and well-researched approach to medical aspects of recovery, this patient population seldom receives adequate social support for the non-physical consequences of their injuries. Following a serious orthopaedic injury, patients may face numerous life changes including negative impacts on relationships, independence, emotional well-being, financial stability, and ability to work or participate fully in daily activities. Patients may also experience new or worsening mental health problems. Studies have found orthopaedic trauma patients to have higher-than-average rates of many types of mental health disorders, including depression and anxiety. A 2017 systematic review found that, post-injury, the weighted pooled prevalence of depression and post-traumatic stress disorder in orthopaedic trauma patients was 33% and 27%, respectively. Post-injury, there is a correlation between symptoms of depression and reduced physical function and magnitude of disability. As they often necessitate prolonged periods away from work, traumatic orthopaedic injuries significantly affect employment and income. For many, financial repercussions are immediate, leading to income loss even with minor injuries. It is not uncommon for physical impairment from orthopaedic injuries to keep patients from returning to work for 6 months or more, with compounding detrimental effects on relationships, self-worth, and future job prospects. Research on the financial toll of orthopaedic injuries found that fractures were associated with substantial individual and household income loss up to 5 years after injury, and 1 in 5 patients sustained catastrophic income loss in the 2 years after their injury. Navigating the bureaucratic and logistical challenges of securing employment insurance or Workplace Safety and Insurance Board benefits post-injury can be overwhelming, adding to the burden faced by the injured person or their caregivers. The stress of a serious orthopaedic injury may also lead to or exacerbate substance abuse problems, which may include increased alcohol intake and drug use. In addition, opioids are a ubiquitous part of perioperative pain management following an orthopaedic injury. Despite increasing regulations, more than 80% of orthopaedic trauma patients are prescribed opioids in the post-surgical period. Patients may have low levels of information about safe opioid use, alternative methods of pain management, and what to expect during recovery. When coupled with other post-injury challenges, including mental health issues, job loss, and income reduction, these factors may lead to opioid addiction, potentially resulting in permanent job loss, being without a home, and even death. The combined stress of physical recovery, financial strain, risk of addiction, changes in relationship dynamics, and the intricacies of obtaining aid or services can be emotionally and mentally taxing for orthopaedic patients and their support networks. Orthopaedic trauma patients urgently need outpatient support for their recovery from injury. The orthopaedic fracture clinic is the only outpatient care that these patients receive. Supplementing medical treatment with a comprehensive support system and strategies to mitigate the non-physical impacts of orthopaedic injury has the potential to significantly reduce the negative impact of these traumatic injuries. To address this urgent social need, the investigators propose the integration of a dedicated social worker within the fracture clinic setting to provide support to patients with serious orthopaedic injuries. A clinical trial is needed to evaluate the efficacy of having a social worker in the fracture clinic before widespread hospital policy changes can be implemented. TRIAL DESIGN To efficiently and effectively address the above objectives, the investigators have designed this trial as a batched stepped-wedge cluster randomized trial. The batched stepped-wedge cluster randomized trial is a pragmatic design commonly used in the evaluation of service delivery interventions. In the batched stepped wedge trial, clusters (in this case, surgeon clinics) of patients can be randomized and move through the trial design in 'batches' (e.g. a group of four clusters), which allows for flexibility in the timing of initiation of different clusters. The stepped wedge design itself begins with each cluster in the batch being given a date for crossover into the treatment group, in a randomly determined order. All clusters then enter an initial period in which no patients are exposed to the intervention. Subsequently, at regular intervals ("steps"), clusters cross from the control to the intervention under evaluation. This process continues until all clusters have crossed over to the intervention. At the end of the batch period there is a period when the intervention is being implemented at all clusters. Data collection takes place throughout the batch period, so that all clusters contribute observations under both control and intervention observation periods. The investigators will define a cluster as an individual orthopaedic surgeon fracture clinic. The investigators will conduct this trial in two to four batches, with four clusters in each batch. The number of batches will depend on the enrollment rates during the previous batches. For each batch, before participant enrollment begins, the Methods Centre will randomize the clusters (i.e. clinics) to determine when they will begin the intervention (i.e. social worker support in the fracture clinic). All clinics in the batch will begin participant enrollment into the non-intervention arm (i.e. usual care with no social worker services). At 10 weeks, one clinic will begin the intervention while the others will continue enrolling in the non-intervention arm as per randomization. At 20 weeks the second clinic will cross over to begin treatment, and at 30 weeks the third clinic will crossover. The fourth clinic will cross over after 40 weeks. All clinics (clusters) will enroll for 12 months. To ensure feasibility of this design, the investigators will conduct a Vanguard phase that will include two clusters who will enroll for a period of 8 months. As per the above, both clusters will being participant enrollment into the non-intervention arm (i.e. usual care with no social worker services). At 10 weeks, one cluster will begin the intervention while the other cluster will continue enrolling in the non-intervention arm as per randomization. At 20 weeks the second cluster will cross over to begin treatment. Participant enrollment in both clusters will continue for 10 additional weeks. Data from the vanguard phase will be included in the trial, if there are no significant changes to the protocol. Batch timelines may also overlap each other, and the timing will depend upon funding and the readiness of different clusters to begin recruitment. The overall sample size for the trial will be approximately 2,000 participants. METHODOLOGY The Surgery Methods Centre in the Department of Surgery at McMaster University will coordinate this trial. The Principal Investigator will chair the trial's Steering Committee, which will provide oversight and guidance for the trial. Methods Centre personnel will obtain research ethics board approval for the trial. Research personnel at each selected cluster (the institution affiliated with each cluster) will obtain local Research Ethics Board approval prior to initiating local trial activities. Clusters will be selected at hospitals in the United States and Canada. Cluster Eligibility and Selection of Clusters Clusters will be defined by orthopaedic surgeon clinical practices and participating sites may contribute one or more clusters. Clusters will be selected in batches of four. Methods Centre personnel will carefully screen potential clusters for eligibility. Cluster inclusion criteria are: 1) adequate research personnel infrastructure to manage the trial; 2) adequate surgically managed fracture volume to complete enrollment within the timelines; and 3) ability to secure a qualified social worker to perform the trial intervention. The exclusion criteria are: 1) lack of interest in the trial; 2) anticipated challenges with complying with the protocol; 3) conflicting studies that would inhibit patient participation; and 4) budgeting or contract constraints. The screening process will begin with potential clusters (orthopaedic surgeons' clinics) completing a feasibility questionnaire that includes the cluster eligibility criteria. Clusters that meet the eligibility criteria at this stage will be invited to participate in a meeting to review local logistics and to confirm interest and eligibility. Study personnel will document reasons for cluster ineligibility. Of note, clusters (individual orthopaedic surgeons clinic days) within the same batch and across the batches may be at the same or different hospital and/or university. The investigators will document key demographic information for each cluster. Cluster Allocation and Concealment The Methods Centre will randomize each batch of clusters prior to the start of patient recruitment to determine when they will begin the intervention. At the time of randomization, Methods Centre personnel will notify each cluster of the timing of their crossover from usual care to the treatment phase of the trial. Patient Screening and Informed Consent Research personnel will assess eligibility for all patients who meet the definition of the screening population. After an initial screening process following local research ethic board guidelines, eligible patients will engage in the informed consent process. To obtain informed consent, personnel at each cluster will adhere to the following procedures: 1) Present trial information in a manner that is understandable to the patient; 2) Discuss the trial with the patient and answer any questions they have; 3) Confirm that the patient understands the risks and benefits of participating in the trial and that their participation is voluntary; 4) Complete the consent process and obtain signatures from the patient and person obtaining consent. Site research personnel will follow the process of obtaining and documenting informed consent forms in accordance with Good Clinical Practice. Participants may withdraw their consent at any time. If a potentially ineligible participant is enrolled, the Adjudication Committee will review the participant's medical record and determine eligibility. This decision will be based on the information available at the time of informed consent. Data Collection and Participant Follow-Up After obtaining informed consent from a participant, research personnel will complete the baseline case report forms (CRFs). Baseline data will be obtained from the participant, the participant's medical record, and/or the participant's treating physicians and entered by the Research Coordinator. Baseline data collection points include participant and injury characteristics such as age, sex, gender, socioeconomic status, co-morbidities, mechanism of injury, among others. At baseline, participants will also complete the LIMB-Q scales, the satisfaction with care question, opioid use question, and the EQ-5D. The investigators will follow participants for 6 months after enrollment, as this aligns with the usual clinical follow-up timeline for orthopaedic injuries. Follow-up may be completed in-person at the fracture clinic, by telephone, by mail, or via secure electronic methods. Research personnel may also review the participant's medical records for emergency room and urgent care visits. The social worker at each clinic will document the social work interactions by maintaining a log that documents each point of contact they have with each participant. This log will include the medium of contact, themes discussed, and number of referrals made. At each trial follow-up visit, investigators will ask participants if they saw a social worker outside of the fracture clinic and document any care referrals that they received. SAMPLE SIZE The investigators chose the sample size to detect differences between the social worker intervention and usual care groups for all-cause visits to emergency room or urgent care centers (primary outcome). The investigators estimated our sample size in accordance with the methodology of Kasza et al. using the R swdpwr package. For a binary outcome, the required cluster period size in a batched stepped wedge cluster randomized trial for a desired level of power depends on the number of time periods, number of clusters assigned to each sequence, number of sequences, number of batches, number of periods of overlap between successive batches, effect size, and ICC parameters. Investigators will assume 80% study power with a two-sided Type I error rate of 5%, either two, three, or four batches of four sequences over five time periods, one cluster allocated to each sequence, no period overlaps, a 10% prevalence rate in the treatment group, a 14.8% prevalence rate in the control group19, and a within-period intra-cluster correlation coefficient of 0.05. Based on these assumptions, the trial will require a total sample size of 1880 patients (cluster size of 47 patients per period) for two batches, 1800 patients (cluster size of 30 patients per period) for three batches, and 1760 patients (cluster size of 22 patients per period) for four batches. To account for a 10% loss to follow-up rate, the total sample size, regardless of the batch size, will be rounded to 2000. DATA ANALYSIS The investigators will prepare a detailed statistical analysis plan (SAP) prior to completion of the trial which will provide a detailed description of all planned statistical analyses. Outcomes Analysis The primary and secondary analyses will be conducted following intention-to-treat (ITT) principles, ensuring that participants are analyzed in their originally assigned treatment groups, regardless of adherence or protocol deviations. Reporting of the trial results will follow the 2010 CONSORT statement and the extension statements for a batched, stepped-wedge cluster randomized trial. For the primary analysis of time to any all-cause return visits to emergency room or urgent care centers, the investigators will perform a Cox proportional hazards regression model, with treatment included as the independent variable, and cluster pre-specified variables prognostic of the outcome, such as pre-injury work status, included as covariates. Results will be presented as a hazard ratio and corresponding 95% confidence interval (CI). For secondary analyses, investigators will follow mixed effects linear regression modelling for the LIMB-Q and satisfaction with care to account for repeated measures, with treatment, time of assessment, and pre-specified variables prognostic of each outcome included as independent variables in fixed effects. Each model will also include a cluster indicator as the random intercept. These secondary outcomes will be summarized as adjusted mean differences between groups with corresponding 95% CIs. The investigators will use logistic regression modeling for self-reported opioid use at 3 and 6 months using. Self-reported opioid use will be included as the dependent variable and each model will include the treatment variable, cluster, and pre-specified variables prognostic of the outcome as independent variables. Investigators will present results as adjusted odds ratios with corresponding 95% CIs. The investigators will conduct a cost-effectiveness analysis from the payer's (hospital) perspective incorporating both participant utilities and direct healthcare costs and intervention costs. They will obtain health utilities from the EQ-5D. Costs will include the cost of employing a social worker, as well as healthcare use costs, such as fracture clinic visits, emergency room visits, urgent care facilities, and hospitalizations. The investigators will compare these costs between the social worker intervention versus usual care groups using a cost-effectiveness analysis framework. The SAP will provide additional details on economic analysis plans. The investigators will also descriptively report the interactions made between the social worker and participants regarding the medium of contact, themes discussed, and the number and types of referrals made. Sensitivity Analyses The investigators will conduct a sensitivity analysis to include only those participants who had at least one visit with a social worker. This analysis will follow the same analyses methods as described for the primary outcome, including appropriate regression models and adjustments for stratification variables. Investigators will conduct a sensitivity analysis that varies definition of return to emergency room and urgent care centre to include only visits that occurred at the same hospital/health system where the participant is receiving care for their fracture (e.g. participating cluster location). This analysis will follow the same methods listed for the primary outcome, including appropriate regression models and adjustments for stratification variables. The investigators will conduct a third sensitivity analysis that will also vary the definition of return to the emergency room and urgent care centre to include all visits. They will follow mixed effects linear regression modeling to account for repeated measures, with treatment, time of assessment, and pre-specified variables prognostic of each outcome included as independent variables in fixed effects. The model will also include a cluster indicator as the random intercept. The outcome will be summarized as an adjusted mean difference between groups with corresponding 95% CIs. Subgroup Analyses The investigators plan to conduct subgroup analyses on gender and sex as potential factors influencing the effectiveness of the social worker intervention. Extensive literature spanning the past two decades suggests that patients tend to respond differently to offers of assistance and may perceive the experience of being helped differently depending on their gender identity. Unfortunately, definitions of sex and gender in older research are often used interchangeably. The investigators hypothesize that individuals identifying as women and those assigned female at birth may demonstrate a heightened receptiveness to assistance and engagement through social worker services. Using the currently accepted definitions, investigators will evaluate subgroups of female vs. male and women vs. men. Non-binary gender identities will be included in the baseline survey and will be analyzed separately where numbers permit. Additionally, the investigators plan to conduct subgroup analyses based on injury severity, pre-injury work status and opioid use at enrollment. These analyses will be approached and reported in accordance with best practices and guidelines for subgroup analyses. For positive subgroup effects, the investigators will use the criteria suggested by Schandelmaier et al. to guide inferences about the credibility of our subgroup analyses. Qualitative Analysis After the interviews have been conducted and transcribed, three Methods Centre personnel will independently review the text of each interview several times, in a process of inductive coding. Each reviewer will use these statements to identify larger subthemes individually, without consulting each other. They will then compare subthemes and work together to create overall themes that describe the experience of recovery from orthopaedic trauma. Researchers, including the interviewer, will, at some point during the research process, complete a written summary of their own experiences and context and situations that have influenced their experiences with respect to this research, making sure to cover all information required by the CORE-Q checklist. Researchers will engage in member-checking as part of the coding process, which will include sending themes and codes to participants to validate these findings. A separate SAP will fully describe the analysis plans for the qualitative components. DATA MANAGEMENT The Methods Centre will provide the research personnel at each institution with the trial case report forms (CRFs) prior to initiation of enrollment. Research personnel will submit the required data, as detailed on the CRFs, to the Methods Centre using the REDCap Cloud electronic data capture system. Research personnel will receive a unique login and password for the REDCap Cloud system and will be able to view and modify data for participants at their cluster(s). Interview data will be transcribed, and original interview recordings will be stored per institution records retention procedures. Transcriptions will be de-identified and labelled with participant number, then uploaded into Dedoose software (or a similar program) in a password protected project. Only those identified as performing the qualitative analysis of the data will have access to the project area. Some quantitative information will be included in the database such as age, gender, and treatment group, or in the case of health care professionals, position at the clinic. The Methods Centre data manager will program the EDC with multiple mechanisms for checking data at the time of entry including skip logic, range checks, and data type checks.26 Upon receipt of new data, Methods Centre personnel will query all missing, implausible, or inconsistent data. Research personnel will be able to review open queries in the system and will be required to respond promptly. SAFETY As this trial provides support and services in addition to usual care, the investigators do not anticipate any significant safety risks to participants. For patients receiving the social work intervention, discussing and addressing issues with a social worker may bring up challenging emotions, potentially leading to a transient increase in distress. For example, helping a participant to better understand their injuries and the recovery process could initially create heightened anxiety as they cope with upsetting information regarding their prognosis. To mitigate this risk, the social worker will perform standard safety and emotional wellness assessments as part of their interaction with the patient. These assessments will provide an opportunity for early assistance with any increased anxieties that the participant may be experiencing. Participants in the usual care arm will receive care according to current practices at each participating cluster, which does not typically involve care from a dedicated social worker. Consequently, they will not experience any increased risk due to their participation in the trial. Safety Monitoring This trial will not require a Data Safety and Monitoring Committee (DSMC) as the intervention is low risk to trial participants, and major morbidity, mortality, or other severe outcomes are not expected or evaluated. This decision aligns with the Food and Drug Administration guidelines for involving a DSMC in an RCT.