Running head: SUBSTANCE USE RELAPSE AND READMISSIONS Substance Use Disorder Relapse and Readmissions Vincent Ekstrom Arizona State University 1 Abstract According to the National Institute on Drug Abuse (NIDA), tobacco, alcohol, and illicit drugs accounted for 820 billion dollars in costs related to crime, lost work productivity, and health care services. Nearly 20 million Americans suffer from substance misuse, but only 3.7 million received treatment. Of those who receive treatment, the risk of relapse is high, ranging from 4060% within a year of treatment. Improvement in the treatment of substance use disorders (SUD) is necessary to improve the health of our society. Current literature demonstrates that individualized recovery plans and follow-up care are effective in reducing relapse and readmission. Costs to the individual, institution, and healthcare system can be reduced. This project aimed to decrease the risk for relapse and readmission with recovery plan reviews at 72hrs, and two-weeks, post-discharge. The risk of relapse was measured by the Time-To-Relapse questionnaire and the UCLA loneliness scale. The project took place in a residential treatment facility in Phoenix, Arizona. There were five participants initially; two were lost at the two-week follow-up. Pre and post-test results were compared to measure potential predictability of relapse. The two-tailed paired samples t-test was performed to compare the means of the scores but yielded insignificant results. All participants maintained sobriety. Qualitative data via interview showed positive results demonstrated by statements from the participants. Recovery plan review with follow-up care is a promising evidence-based practice that can be implemented to help individuals maintain sobriety. Additional research is recommended to examine further the impact on the maintenance of sobriety over time. Keywords: substance use disorder, recovery plan, follow-up, relapse, readmission SUBSTANCE USE RELAPSE AND READMISSIONS 3 Substance Use Disorder Relapse and Readmissions According to the National Institute on Drug Abuse (NIDA), tobacco, alcohol, and illicit drugs accounted for 820 billion dollars in costs related to crime, lost work productivity, and health care services (National Institute on Drug Abuse [NIDA], n.d.). In 2018, the National Survey on Drug Use and Health reported that “approximately 20.3 million people aged 12 or older had a substance use disorder (SUD) related to their use of alcohol or illicit drugs in the past year” (Substance Abuse and Mental Health Services Administration [SAMSHA], 2019, p.1). Despite the significant need, only 3.7 million received any substance use treatment within the past year (SAMHSA, 2019). The gap in treatment is lethal, as 70, 237 people died in 2017 due to drug overdose, and an additional 88,000 deaths were related to alcohol (NIH; CDC). Due to the severity of the impact on the United States, treatment for substance use disorders must be more effective. Problem Statement Any drug taken in excess has a direct stimulation on the brain reward system (American Psychiatric Association [APA], 2013). Drugs of abuse produce feelings of pleasure and reinforce behaviors and memories of use. A significant reason SUD’s are challenging to treat is because of the pleasure derived from stimulation of the reward system. The Diagnostic and Statistical Manual of Mental Health Disorders (DSM) separates drugs into ten classifications: alcohol; caffeine; cannabis; hallucinogens; inhalants; opioids; sedatives, hypnotics, and anxiolytics; stimulants; tobacco; and unknown substances. The defining characteristic of a substance use disorder is a “cluster of cognitive, behavioral, and physiological symptoms indicating that the individual continues using the substance despite significant substance-related problems” (APA, 2013, p. 483). Remission/recovery occurs when an individual no longer meets the criteria for a SUBSTANCE USE RELAPSE AND READMISSIONS 4 substance use disorder, indicating improvement (APA, 2013). Relapse can be defined as the recurrence of symptoms of a disease after a period of recovery (Relapse, n.d.). Mental Health America (MHA), Substance Abuse and Mental Health Services Administration (SAMHSA), and the World Health Organization (WHO) have initiatives to reduce the negative impact of substance abuse, yet relapse rates for SUD’s are estimated between 40-60% within a year of obtaining sobriety (NIDA, n.d.). Thirty-day hospital readmission rates for substance use disorders are between 18-27% (Hines, Barrett, Jiang, & Steiner, 2014). The average length of stay for inpatient treatment of SUD is 4.5 days, with an average cost of $6,700 (Heslin, Elixhauser, & Steiner, 2006). Mental health or substance use disorders were among the conditions resulting in the most, all-cause, 30-day readmissions for Medicaid patients, resulting in a total of 113,100 readmissions for a cost of $832 million (Hines et al., 2014). Individuals with substance use disorders have a higher prevalence of 19 major medical conditions and higher disease burden than those without a substance use disorder (Bahorik et al., 2017). Patients undergoing acute treatment for detox may experience personal and medical crises that can exacerbate emotional, psychological, or mental conditions that can hinder the decision making and critical thinking abilities required throughout recovery to maintain sobriety (Miller & Kipnis, 2006). Healthy People 2020 initiatives for substance abuse include increasing the proportion of people who need and receive specialty treatment for abuse or dependence and reducing the number of deaths attributable to alcohol or drugs (Healthy People, 2020). Treatment Improvement Protocols (TIPs) were developed by the Center for Substance Abuse Treatment (CSAT) to prevent substance use readmissions, and are best practice guidelines for the treatment of substance use disorders. CSAT is a subdivision of SAMHSA that provides national leadership in substance use SUBSTANCE USE RELAPSE AND READMISSIONS 5 treatment. Its mission is to promote community-based substance abuse treatment and recovery services in every community. Additionally, the United Nations Office on Drugs and Crime (UNODC) created the International Standards for the Treatment of Drug Use Disorders to disperse good practice examples that are evidence-based to improve care in areas where treatment is ineffective and not supported by evidence. Within these documents, strategies for readmission and relapse prevention, along with many other aspects of treatment are provided. Individuals who have more social support report lower psychological distress (Segrin, McNelis, & Swiatkowski, 2016). Social support has an indirect effect on problem-drinking by reducing psychological distress. Loneliness is a form of psychological distress and has negative impacts on self-esteem and affect (Xia & Yang, 2019). Social isolation is associated with an increased risk of substance use. Individuals with lower social support experienced higher levels of drinking-related problems (Segrin, McNelis, & Swiatkowski, 2016). Socially isolated youth are more at-risk for engaging in alcohol or cigarette use (Nino, Cai, & Ignatow, 2016). Group therapy, support groups, and networks like Alcoholics Anonymous are beneficial because members see they are not alone in their struggle (Yalom & Leszcz, 2005). They find support, connection, and friendship in the groups and are antidotes to isolation, alienation, and loneliness. Through involvement in groups, individuals have the opportunity to observe others who have gone through the same problems and have improved their lives. Clients gain hope that they can get better by watching others that have changed their lives. The universality of human problems can provide a client with a significant source of relief (Yalom & Leszcz, 2005). SUBSTANCE USE RELAPSE AND READMISSIONS 6 Purpose & Rationale The purpose of this project is to improve the quality of recovery planning and follow-up care for the patient diagnosed with any substance use disorder, subsequently reducing their risk for relapse and readmission to a treatment facility. Background & Significance Outpatient services are associated with better outcomes in substance use treatment (Timko, Schultz, Britt, & Cucciare, 2016). Upon discharge from an inpatient setting, patients have outpatient substance use services coordinated to continue treatment; however, patient utilization of outpatient services upon inpatient discharge trends as low as 40-50% (Garnick, 2017; Marino, 2016; Timko, Schultz, Britt, & Cucciare, 2016). Medicaid analysis found that 67.7% of individuals treated for inpatient substance use had no residential or outpatient services within 14 days of hospital discharge (Reif, Avebedo, Garnick, & Fullerton, 2017). Patients have identified the distance to travel for services, other life responsibilities, and financial costs of treatment as barriers to follow up (Marino et al., 2016). Therefore, it is not surprising that inpatient readmissions were about 29.3% within 90 days of discharge, with many occurring in the first week (Reif et al., 2017). Treatment Improvement Protocols (TIPs) were developed by the Center for Substance Abuse Treatment (CSAT) to prevent substance use readmissions, and are best practice guidelines for the treatment of substance use disorders. Additionally, the United Nations Office on Drugs and Crime (UNODC) created the International Standards for the Treatment of Drug Use Disorders to disperse good practice examples that are evidence-based in hopes of improving care in areas where treatment is ineffective and not supported by evidence. Within these documents, strategies for readmission and relapse prevention, along with other aspects of treatment are provided. Immediately after admission, the development of goals and plans for substance use treatment post-discharge should be discussed SUBSTANCE USE RELAPSE AND READMISSIONS 7 daily (Busse, Gerra, Koutsenok, & Saenz, 2015). Additionally, effective treatment planning includes strategies for a successful transition to the next treatment, and support should be given to help navigate the social and health care systems (Busse et al., 2015). A psychiatric inpatient unit within the Veterans Affairs (VA) medical system, evaluated the implementation of a recovery-oriented model of care (Zuehlke, Kotecki, Kern, Sholty, & Hauser, 2016). In this initial model, treatment and discharge planning on the unit were provider-driven (Zuehlke et al., 2016). On another psychiatric inpatient unit with the VA system, the process was modified to include direct patient participation in treatment planning (Koval et al., 2016). Both units incorporated a peer support specialist to assist with individual recovery goal development and access to resources (Zuehlke et al., 2016). While readmission rates among one facility did not change significantly with the intervention, patients had higher satisfaction with the care provided (Zuehlke et al., 2016). The second facility demonstrated a decrease in readmission percentages after implementation (Koval et al., 2016). The intervention included direct involvement of the patient in their recovery, revision of policies and procedures to reflect the importance of recovery, and extensive staff training on recovery principles (Koval et al., 2016). This intervention demonstrates that including the patient directly in treatment planning is very useful for recovery. Bridging strategies implemented before discharge can be useful for increasing the odds of successful contact with outpatient resources (Taylor et al., 2016). One adult inpatient unit initiated a brief interview addressing goals and barriers immediately before discharge to increase engagement in aftercare and reduce early psychiatric readmissions. Patients who did not receive the intervention were twice as likely to be readmitted within thirty days (Taylor et al., 2016). Brief Critical Time Intervention (BCTI) is an Evidenced Based Practice (EBP) demonstrated to be highly effective in connecting patients to outpatient treatment (Shaffer et al., SUBSTANCE USE RELAPSE AND READMISSIONS 8 2015). Acute service coordinators (ASC) at a community health organization identify unmet patient needs and develop a personalized discharge plan. The ASC’s were directed to provide more intense recovery planning. They focused on strengths and connections to community resources to help patients develop autonomy. Individuals that received the BCTI had a readmission rate of 28% compared to 47% in those who did not receive the intervention (Shaffer et al., 2015). Another strategy for improving readmissions is a recovery interview. This intervention was developed by a medical center in Pennsylvania for use in inpatient substance use programs. The meeting is approximately 15-30 minutes and focuses on eight different topic areas relating to reasons for readmission, use of crisis plan, discharge planning, and barriers. The group receiving the intervention had lower readmissions and reduced odds of readmission (Hutchison et al., 2018). Maarevand et al. (2015) explored a community based relapse prevention plan with a motivational interview at discharge. The study was a randomized controlled trial with 71 participants. Drug tests were done at 45 and 90 days as a quantitative measure of abstinence. The motivational interview at discharge and the involvement of community members had a higher rate of abstinence than the control group. All of the individuals who relapsed in during the program did so within the first 45 days post-intervention. Another randomized controlled trial was conducted in Zambia and used treatment recommendations from the WHO mental health general action plan (mhGAP) (Sheikh et al., 2017). A single 20-minute motivational interview was conducted with the patient and the family member at discharge. There were significant reductions in the frequency of alcohol use for those that did not remain abstinent. In addition, the intervention group had an average abstinence period of 51 days versus 10 days in the control group. SUBSTANCE USE RELAPSE AND READMISSIONS 9 Ongoing studies are evaluating interventions for this population with the desired outcome of reducing readmission rates. One intervention is a patient navigator who provides a motivational interview at the time of discharge and assists patients in their transition to outpatient treatment. The intervention includes helping with a lack of transportation, lack of an ID card, and other external barriers. They also assist in resolving internal barriers to treatment like decreased motivation for treatment (NCARE, n.d.; NAVSTAR, n.d.). No outcome data is available for these interventions as they are still in progress, but patient navigators are of significant interest. The quality of network ties and efficiency amidst substance use programs are associated with readmissions (Spear, 2014). Higher efficiency programs get the patient connected to services more quickly. Programs with higher efficiency were associated with lower odds of readmission for compared to low-efficiency programs with higher readmission risk (Spear, 2014). Internal Evidence The project took place in a residential treatment facility in Phoenix, Arizona. The facility has not conducted formal quality improvement projects or research regarding substance use readmission rates. However, when meeting with clinical leaders, administration, and site champions, they described readmission rates for substance abuse disorders as a problem that has been observed by many staff members, which include nurses, social workers, and clinicians. Repeat admissions provide an opportunity to improve on the quality of care that is provided. The site champions have a desire to improve the outcomes for individuals diagnosed with substance use disorders. Additionally, reimbursement has changed in recent years, and readmission for substance use may become a marker in which insurance uses to calculate payments. Key stakeholders are concerned and want to protect the facility’s role in providing mental health services while also maintaining financial stability. SUBSTANCE USE RELAPSE AND READMISSIONS 10 PICO Question Substance use disorder readmissions are detrimental for this facility and the entire health care system. Based on the internal data of the organization and the desired outcome the following clinical question was developed: In adult hospitalized patients diagnosed with substance use disorder (P), how does a Relapse Prevention Plan review (I) compared to standard discharge instructions (C) affect self-efficacy and readmission rates (O)? Evidence Synthesis Search Strategy Respective to the PICO question, an exhaustive search was performed. Three databases were used: CINAHL, PsycINFO, and PubMed. Three grey literature resources were also exhausted using: Health Sciences Online, Virginia Henderson Global Nursing e-repository, and MedNar. Keywords included: Substance use disorder, addiction, substance abuse, rehospitalization, readmission, relapse, recovery, discharge, intervention. Limits were set to the English language, an origination date of 1-1-2014 or sooner, and published in a peer-reviewed journal and/or database. Initial results yielded 51 results in CINAHL, 99 results in PsycINFO, and 525 results in PubMed. After the evaluation of search results, terms were reduced to “substance use disorder” and readmission or rehospitalization to yield more relevant results. Producing 15 results in CINAHL, 25 results in PsycINFO, 29 results in PubMed, 415 results in Virginia Henderson Global Nursing e-repository, 265 results in MedNar, and two results in Health Sciences Online. Ten studies were selected to answer the PICO question, and each was independently reviewed. These studies were selected because of several factors: inpatient psychiatric setting, relevance to the desired outcome, and effective, feasible interventions. SUBSTANCE USE RELAPSE AND READMISSIONS 11 Critical Appraisal and Synthesis Ten studies were retained for this literature review. Retained studies include six longitudinal cohort studies (LCS) and four randomized controlled trials (RCT) (Appendix A). The Melnyk and Fineout-Overholt’s (2014) rapid critical appraisal tools were used to evaluate their validity, reliability, and applicability. Eight studies were conducted in the United States, one in Iran, and one in Zambia. Study length ranged from 45-days to three years. All study samples evaluated patients with a substance use disorder, a mental health disorder, or cooccurring substance use and mental health disorder (dual diagnosis). All samples contained patients aged 18 years or older, with some limiting the age group from 18-65 years of age. Studies that defined a comparison group had relatively balanced matching concerning gender. The balanced design was used to control for confounding variables. One study had a skewed sample, which reduces its ability to be generalized (Akerele et al., 2017). The sample being 80% African American, 62% homeless, and 65% male (Akerele et al., 2017). One study did not provide any demographics for the population studied but, because of the setting, some population demographics can be inferred (Zuehlke et al., 2016). Most of the studies address possible limitations, but none reported any bias. Cohort studies have inherent selection bias. All six LCS’s are subject to selection bias. The quasi-experimental design allows them to be conducted with or without control/comparison groups. Three of the LCS’s do not have a comparison group; and self-identified as quality improvement (QI) projects; however, their goal was to generate new knowledge thus disqualifying them as QI projects. Measured outcomes were readmission rates and the likelihood of readmission which demonstrates objectivity, reduced bias, and higher reliability. Most studies used logistic regression analysis and reported results in the form of an odds ratio (OR). Using logistic regression allowed for greater control over numerous potential SUBSTANCE USE RELAPSE AND READMISSIONS 12 confounders. Also, it strengthens the reliability of the results. The analysis was further supported by significance tests for the models being used. Some studies used several significance tests to ensure a good fit, improving confidence that the statistical model represents the data collected. Two of the randomized controlled trials included in this synthesis have not yet been completed. They are ongoing clinical trials and have been included to demonstrate the relevance and need for this study. They also provide a reference for framework and design. Conclusion from Evidence Substance abuse relapse and subsequent readmissions are costly to the individual, healthcare system, and the community. Inpatient stays are often brief, and there is a lack of engagement in outpatient services. The evidence suggests that modifying discharge protocols and incorporating detailed recovery planning can reduce relapse and readmission rates to substance abuse facilities. In the studies that had an experimental and control group, the experimental group consistently demonstrated a decrease in the likelihood/odds of readmission and relapse. Most of the support for this change comes from level III and level II evidence. More randomized controlled trials are needed to build upon current literature and support future studies. Conceptual Framework & EBP Model The Neuman systems model (Appendix C) was chosen as the conceptual and theoretical framework as a way to describe interrelated concepts, understand and predict events, and guide the desired impact of the DNP project. Butts & Rich (2018) state that the model provides nursing with a comprehensive, systems-based guide. The model focusses on the response of the client to stressors, and the use of primary, secondary, and tertiary nursing prevention interventions for ideal client wellness. Stressors are disruptive forces and are categorized in three levels: Intrapersonal (occur within the person or family); Interpersonal (occur between individuals and SUBSTANCE USE RELAPSE AND READMISSIONS 13 their roles); and Extra-personal (occur outside the individual) (Butts & Rich, 2018). A person is viewed holistically and consists of five variables: physiological, psychological, sociocultural, developmental, and spiritual. Outcome and treatment planning within the model involves collaboration between the caregiver and the client. Neuman places significant emphasis on wellness and the central role that clients play in setting goals and identifying prevention interventions (Neuman & Fawcett, 2011). Neuman’s model is unique because of the inclusion of the perceptions held by the client and the nurse (Butts & Rich, 2018). Interactions between the client and the environment are significant because they can have a positive or negative effect on the other (Neuman & Fawett, 2011). The effectiveness of the interventions is based on whether the client’s goals were met or not met. The studies evaluated did not identify or use Neuman’s model, but the framework can be used to understand the synthesized evidence. Increasing patient involvement in treatment and discharge planning was a significant component in many of the studies evaluated. Individualized recovery planning is an example of tertiary prevention and client-centered care, described in Neuman's systems model. Engaging with the client in a detailed and specific manner of how goals are going to be met in various scenarios is preventative. The Iowa Model of Evidence-Based Practice to Promote Quality Care was selected to guide the development of the evidence-based project. It has been used in a variety of settings and includes input from the entire organizational system (Schaffer, Sandau, & Diedrick, 2012). The layout of the Iowa Model of Evidence-Based Practice to Promote Quality Care (Appendix D) integrates quality improvement and research utilization. It is a model that nurses find easily understandable (Gawlinski & Rutledge, 2008). Melnyk & Fineout-Overholt (2014) describe it as a step by step process that begins initially with a problem SUBSTANCE USE RELAPSE AND READMISSIONS 14 or knowledge-focused trigger. These triggers highlight the opportunity for improvement and lead to the questioning of current clinical practice standards (Melnyk & Fineout-Overholt, 2014). High readmission rates for individuals with SUD were considered to be both a problem and knowledge- focused trigger. The pilot intervention and subsequent evaluation is a crucial step in the process and determines whether the practice change is appropriate and effective (Melnyk & Fineout-Overholt, 2014). Lastly, the model expects the dissemination of results to contribute to professional learning. The Iowa model is straightforward and guides clinicians through the EBP process. The plan for follow-up care was selected based on the holistic view of the Neuman systems model. The goal of the intervention was to more concretely discuss how the patient will connect with resources in the community, in addition to assessing their needs within several domains. The three levels of stressors, described by Neuman, are integrated into the recovery plan by inquiring about personal relationships, roles and expectations, and extra-personal needs. Five of the domains come directly from the Neuman systems model and are: physiological, psychological, sociocultural, developmental, and spiritual. The two remaining domains: transitioning and prevention, are based on Neuman’s concept of extra-personal stressors or stressors that occur outside the individual. Methods Ethical Considerations and IRB Approval The evidence-based project was approved by the IRB board affiliated with Arizona State University (ASU) (Appendix E) and required one modification (Appendix F). Protection of human subjects included the creation of a specific participant ID as follows: first letter of first name; favorite color; patient selected two-digit number (e.g. Vgreen12). Participants included in SUBSTANCE USE RELAPSE AND READMISSIONS 15 the study were required to be 18 years of age or older, have a primary diagnosis of a Substance Use Disorder, and completed a minimum of 30 days in treatment without involuntary discharge. Minors, Adults who are unable to consent, Individuals who are unable to read or write in English, and prisoners, were not included. No compensation or credit was given to any participants. The potential benefits of participation in the study include decreased risk of relapse, improved self-efficacy, and reduced risk of readmission. There was no known risk greater than those that are associated with everyday types of activity. Minor psychological discomfort may be experienced during the discussion of difficult personal topics and the recovery plan. If the patient began to feel suicidal or wanted to self-harm, appropriate resources would be given. If at the 72hour or two-week follow-ups the patient was found to have relapsed or be in crisis, crisis resources would be given. The co-Principal Investigator (PI) met with facility staff to present and explain the study. Staff were provided with the Recruitment Criteria Flyer (Appendix G) and the Patient IRB Consent form and asked to present this information to any potential participants at the facility, which includes contact information for patients to reach the co-PI if they wish to participate. Confidentiality of data was ensured via the use of ASU’s Sensitive and Highly Sensitive Information policies and protocols. Electronic data was stored via local storage using VeraCrypt software. Physical data, such as paper surveys were stored in a locked cabinet in the manager’s office. Data was linked via a specific participant ID as described above. The Co-PI consented all the participants for the study. The consent process took place over the phone. Consent was inferred and obtained via verbal affirmation and completion of the surveys. Participants were given as much time as needed when deciding whether they would like to participate. Participants were allowed to have all questions answered before participation. Participants were able to state back to the Co-PI requirements for the project. SUBSTANCE USE RELAPSE AND READMISSIONS 16 The intervention consisted of a three-part process: (1) patient develops a recovery plan during treatment using the Self-Management and Recovery Training (SMART) system; (2) 72hour post-discharge follow-up phone call; (3) two-week post-discharge follow-up phone call. The general plan can be viewed in Appendix (H). During the 72-hour follow-up, participants completed: Time to Relapse Questionnaire (TRQ) (Appendix I), UCLA Loneliness Scale (Appendix J), and a Semi-structured interview (Appendix H). Upon discharge, the patient reviewed the recovery plan that was developed with facility staff. At 72-hours post-discharge, patients received a telephone call to complete the TRQ and UCLA Loneliness scale, review receipt of any medications, discuss any upcoming appointments and barriers, assess the recovery plan's effectiveness thus far, and provide crisis resources if necessary. At two-weeks, postdischarge patients received another phone call to assess the effectiveness of the recovery plan, provide crisis resources if necessary, and complete TRQ and UCLA Loneliness scale. Instruments The outcomes measured used qualitative and quantitative data through the use of questionnaires and semi-structured interviews. The Time to Relapse Questionnaire was designed to assess the time from initial thought of drug use to actual use. It is a 9-item scale that classifies results into Sudden Relapse, Short Delay Relapse, or Long Delay Relapse. A higher score in one category indicates the patients predominate relapse style. The TRQ is not in widespread use but demonstrated validity and internal consistency (Adinoff et al., 2010). There are no restrictions for use, approval of its use was obtained by the creator, and it is in the public domain. It will provide quantitative information to answer if the discharge intervention can decrease relapse rates and reduce readmissions. Patient loneliness will also be evaluated, and a separate tool will be used. The University of California, Los Angeles [UCLA] loneliness scale measures subjective SUBSTANCE USE RELAPSE AND READMISSIONS 17 feelings of loneliness and social isolation (Russel, 1996). Three studies demonstrated validity and reliability, and many recent studies have used this tool, further confirming its credibility. The loneliness scale will also provide quantitative data for evaluating if the recovery plan is effective at increasing connectedness in the community. Incorporating an assessment of loneliness and social isolation meets Neuman's' idea of a holistic path to wellness. The semi-structured interviews addressed individual concerns regarding discharge needs and the effectiveness of the intervention. During the interview, barriers to accessing community resources, strategies to overcome barriers, detailed discharge plans, and current needs to facilitate the transition were discussed. The semi-structured interviews provided client perceptions, and according to Neuman, are an essential part of the nursing process when developing treatment plans (Butts & Rich, 2018). The desired outcome is determining the likelihood of substance use relapse and subsequent readmission and the impact of the recovery plan intervention. The demographics tool can be viewed in Appendix (K). Budget The proposed budget can be seen in Appendix (L). No funding was received for this study. Results Outcomes Twenty-three individuals were eligible for the study, but only five were reached and participated. Two individuals were unable to be reached at the two-week follow-up mark. Four were left messages but were never reached, twelve had either a wrong phone number or the number was disconnected, and two were still in treatment. This can be seen in the flowchart of participant enrollment and retention (Appendix M). As illustrated in Table (3), (Appendix N) SUBSTANCE USE RELAPSE AND READMISSIONS 18 frequencies and percentages were calculated for Gender, Age_Range, Relationship_Status, and Education. The most frequently observed category of Gender was Male (n = 3, 60%). data showed that the sample (N=5) had three males (60%) and two females (40%). The most frequently observed category of Age_Range was 30-39 (n = 4, 80%), the other participant was within the 21-29 age range. The most frequently observed category of Relationship_Status was Single, never married (n = 2, 40%). The most frequently observed category of Education was GED (n = 2, 40%). Frequencies and percentages for Outapatient_services, Working, and In_Crisis are presented in Appendix O. At 72 hours, all of the participants reported having some form of outpatient service. The most frequently observed category of Outpatient_services was Y (n = 5, 100%). None were considered to be in crisis or need crisis resources. The most frequently observed category of In_Crisis was N (n = 5, 100%). 80% of the participants were working. The most frequently observed category of Working was Y (n = 4, 80%). Two out of five had concerns about housing. At the two-week follow-up, only three of five were able to be reached. One male and one female. Of the remaining three reached, none were in crisis or needed crisis resources. All were still utilizing outpatient services and were working. One participant still had concerns about housing but already had help in place. The average time length of the phone call at 72hrs was 8.2 minutes and 4.6 minutes at two weeks. Statistical Significance The items for UCLA loneliness scale had a Cronbach's alpha coefficient of 0.91, indicating excellent reliability. Appendix (P) presents the results of the reliability analysis. A two-tailed paired samples t-test was conducted to examine the mean difference of UCLA_72hr and UCLA_2week. The result of the two-tailed paired samples t-test was not significant based on an alpha value of 0.05, t (2) = 1.00, p = .423. A table of the means is presented in Appendix (Q). SUBSTANCE USE RELAPSE AND READMISSIONS 19 A two-tailed paired samples t-test was conducted to examine the mean difference of Sudden_72hr and Sudden_2week. The result of the two-tailed paired samples t-test was not significant based on an alpha value of 0.05, t (2) = 1.73, p = .225. A table of the means is presented in Appendix (R). A two-tailed paired samples t-test was conducted to examine the mean difference of Short_72hr and Short_2week. The result of the two-tailed paired samples ttest was not significant based on an alpha value of 0.05, t (2) = -2.00, p = .184. A table of the means is presented in Appendix (S). A t-test for Long_72hr and Long_2week could not be conducted because there were no changes in scores. A linear regression analysis was conducted to assess whether Short_72hr, Sudden_72hr, and Long_72hr significantly predicted UCLA_72hr. The results of the linear regression model were not significant, F (3,1) = 0.32, p = .825, R2 = 0.49, indicating Short_72hr, Sudden_72hr, and Long_72hr did not explain a significant proportion of variation in UCLA_72hr. Since the overall model was not significant, the individual predictors were not examined further. Appendix (T) summarizes the results of the regression model. The mean difference between scores were as follows: UCLA_72hr (M= 3.67, SD= 0.58), UCLA_2week (M= 3.33, SD= 0.58) , TRQ-Sudden_72hr (M= 5.33, SD = 2.08), Sudden_2week (M= 4.33, SD= 1.15), TRQ- Short_72hr (M= 5.00, SD= 1.00), Short_2week (M= 5.67, SD= 0.58), no differences in scores were noted between Long_72hr and Long_2week. This can be viewed in Appendices Q-T. The project was based on Neuman’s systems model which measures the effectiveness of an intervention on whether the client’s goals were met or not met. Based on this view, the intervention was effective because all participants maintained sobriety. Statements from participants are included below. SUBSTANCE USE RELAPSE AND READMISSIONS 20 Participant 1: Having someone reach out to me was valuable because I felt more supported. I reflected on the questions asked, and I felt good because I saw differences in myself. Participant 2: The entire program changed my life and I know that I would have relapsed again without it. I was glad to have some additional support, and the follow-ups made sense. Participant 3: I don’t have any concerns about my recovery plan. If I do start thinking about using then I know that I need to go straight to a meeting. It was nice to have someone checking in with me. Clinical Significance Although statistical significance was not seen there is valuable clinical significance that can be derived from carefully viewing the project and the data. The mean for UCLA loneliness scale decreased, which may be supported by the qualitative statements, indicating that clients felt more supported. The two participants that were not reached at the two-week mark may be evidence of the rapidly changing status of individuals with substance use disorders. The decrease in mean score of the sudden relapse style and subsequent increase in mean of the short relapse style may indicate that individuals transitioned from a sudden to short style over this time period. This is also supported from qualitative statements indicating that the questions asked on the TRQ caused a self-reflection and influenced the participants thinking and perception. Knowing that the risk of relapse is high, and that individuals are particularly vulnerable early in recovery, clinicians should be aware that additional support is necessary and beneficial. This support also strengthens the therapeutic alliance and sense of connection for individuals; knowing that someone is actively there for them and ready to help. The environment and social networks that SUBSTANCE USE RELAPSE AND READMISSIONS 21 patients return to after treatment may not be supportive of sobriety. It is the responsibility of every healthcare provider to improve this. Implications for Change & Innovation Many key stakeholders can benefit from this project and include facilities, administrators, hospital staff, providers, patients, insurance companies, and the healthcare system as a whole. Patients would benefit from effective treatment and improved quality of life. Inpatient facilities and other components of the healthcare system would benefit from reduced costs associated with readmissions. The desired outcome and expected impact for the project was the reduction of relapse and readmission rates for individuals diagnosed with a substance use disorder. The primary goal was to evaluate the effectiveness of an individualized recovery discharge plan review with follow-up care on relapse and readmission rates. Changes at the practice and process level could be that nurses, social workers, or other staff develop the recovery plan with the patients and follow up with them after discharge. Discussing concrete details of how appointments will be met, how medications received, or what actions they might take if faced with cravings. The role may vary based on facility staffing and budget. Ideally, the process of recovery plan review and follow-up care would be done by an individual explicitly hired for that purpose. This would essentially model the role of a Heart Failure coordinator. The project borrowed strategies from the medical model of a heart failure coordinator and applied them to a behavioral health setting, like a “Cross-Pollinator”. The innovative leader may anticipate that reimbursement for substance use disorders may change in the future, because of the high cost, and begin to put systems in place to prepare for that change. Although the cost may be higher in the short term, leaders must look at the long-term goals and outcomes. At a systemic level, individualized recovery plans and intermittent follow-ups can become a standard of care for SUBSTANCE USE RELAPSE AND READMISSIONS 22 patients diagnosed with a SUD. The body of knowledge the project impacts is related to substance use disorders and evidence-based treatment practices that improve care. It will contribute to the current body of literature in support of individualized recovery plans with follow-up care. The project can be used as a model for any facility searching for strategies to reduce the risk of relapse and readmissions in patients with SUD. Several factors stimulated innovation for the project such as the observation of repeat admissions, the unknown process of what happens to the patient after discharge, and the limited ability of services to meet their needs. Providers must understand that individuals going through substance abuse treatment are vulnerable and need continued support. Making treatment recommendations without following through is doing patients a disservice and reflects a limited understanding of the acute and fragile nature of recovery, particularly early on. Discussion Limitations The study had several limitations and challenges. Ideally, more individuals within the treatment facility would have had phone numbers available. Exploring why more phone numbers were not available would be valuable for the facility. It would provide an opportunity to improve upon future projects or services that may benefit from having that information. The process of explaining the study and questionnaires was challenging to complete over the phone. Finding questionnaires that were relatively quick, easy to explain and understand, and demonstrated reliability was a challenging process. Participants may be distracted, pressed for time, or misunderstand information because of this communication modality. This gives reason to interpret the quantitative data with caution. Additionally, it provides opportunity for the development of questionnaires that can be easily used over the phone. Many of the current SUBSTANCE USE RELAPSE AND READMISSIONS 23 questionnaires that assess risk for relapse have 20 or more questions and would be cumbersome to explain over the phone. The sample size is also a severe limitation and may not reflect a larger population. Conclusion Substance abuse relapse and readmissions to inpatient facilities are costly to the individual, the community, and the healthcare system. Many treatment modalities can treat addiction and support recovery, but we must improve engagement in these services. If patients never use them, they are totally ineffective. The recovery-focused discharge plan with follow-up care can reduce readmission rates to inpatient facilities, decrease healthcare costs, and improve patient health outcomes. Further research is recommended to evaluate and identify best practices regarding recovery plan development and follow-up care for individuals with substance use disorders. SUBSTANCE USE RELAPSE AND READMISSIONS 24 References Adinoff, B., Talmadge, C., Williams, M. J., Schreffler, E., Jackley, P. K., & Krebaum, S. R. (2010). Time to Relaps Questionnaire (TRQ): a measure of sudden relapse in substance dependence. The American journal of drug and alcohol abuse, 36(3), 140–149. https://doi.org/10.3109/00952991003736363 Akerele, E., Lim, C., Olupona, T., Ojo, O., Co, N., & Lim, J. J. (2017). Reducing readmission rates in inpatient settings. International Journal of Mental Health, 46(3), 168-176. doi:10.1080/00207411.2017.1295782 American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing. Bahorik, A.L., Satre, D.D., Kline-Simon, A.H., et al. (2017). Alcohol, cannabis, and opioid use disorders, and disease burden in an integrated health care system. Journal of Addiction Medicine 11(1):3-9. Busse, A., Gerra, G., Koutsenok, I., & Saenz, E. (2015). International standards on the treatment of drug use disorders. Drug and Alcohol Dependence, (156), e32. Butts, J. B., & Rich, K. L. (2018). Philosophies and theories for advanced nursing practice (3rd ed.). Burlington, MA: Jones & Bartlett Learning. Garnick, D. W., Horgan, C. M., Acevedo, A., Lee, M. T., Panas, L., Ritter, G. A., … BeanMortinson, J. (2017). Influencing quality of outpatient SUD care: Implementation of alerts and incentives in Washington State. Journal of Substance Abuse Treatment, 82, 93–101. https://doi.org/10.1016/j.jsat.2017.09.007 SUBSTANCE USE RELAPSE AND READMISSIONS 25 Gawlinski, A., & Rutledge, D. (2008). Selecting a Model for Evidence-Based Practice Changes: A Practical Approach. AACN Advanced Critical Care, 19(3), 291-300. doi:10.4037/15597768-2008-3007 Healthy People 2020 [Internet]. Washington, DC: U.S. Department of Health and Human Services, Office of Disease Prevention and Health Promotion [cited [2/24/2020]]. Available from: https://www.healthypeople.gov/2020/topics-objectives/topic/substanceabuse/objectives Hedegaard, H., Warner, M., & Miniño, A. M. (2018). Drug overdose deaths in the United States, 1999-2017 (pp. 1-8). US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Health Statistics. Heslin, K. C., Elixhauser, A., & Steiner, C. A. (2006). Hospitalizations involving mental and substance use disorders among adults, 2012: statistical brief# 191. Hines, A. L., Barrett, M. L., Jiang, H. J., & Steiner, C.A. (April 2014). Conditions with the largest number of adult hospital readmissions by payer, 2011. Retrieved from http://www.hcup-us.ahrq.gov/reports/statbriefs/sb172-Conditions-ReadmissionsPayer.pdf. Hutchison, S. L., Flanagan, J. V., Karpov, I., Elliott, L., Holsinger, B., Edwards, J., & Loveland, D. (2018). Care management intervention to decrease psychiatric and substance use disorder readmissions in Medicaid-enrolled adults. The Journal of Behavioral Health Services & Research. http://dx.doi.org.ezproxy1.lib.asu.edu/10.1007/s11414-018-9614-y Koval, R., Mcdonagh, J., Grubaugh, A., Young, W., Corcoran, B., Lee, A., … Edlund, B. (2016). Implementation of recovery programming on an inpatient acute psychiatric unit and its SUBSTANCE USE RELAPSE AND READMISSIONS 26 impact on readmission. Journal of Addictions Nursing, 27(2), 101–108. https://doi.org/10.1097/JAN.0000000000000121 Maarefvand, M., Eghlima, M., Rafiey, H., Rahgozar, M., Tadayyon, N., Deilamizadeh, A., & Ekhtiari, H. (2015). Community-based relapse prevention for opiate dependents: A randomized community trial. Community Mental Health Journal, 51, 21-29. https://doi.org/10.1007/s10597-014-9772-1 Marino, L., Wissow, L. S., Davis, M., Abrams, M. T., Dixon, L. B., & Slade, E. P. (2016). Predictors of outpatient mental health clinic follow-up after hospitalization among Medicaid-enrolled young adults. Early Intervention in Psychiatry, 10(6), 468–475. https://doi.org/10.1111/eip.12206 Melnyk, B. M., & Fineout-Overholt, E. (2014). Evidence-based practice in nursing and healthcare: A guide to best practice (3rd ed.). Philadelphia, PA: Wolters Kluwer Health. Miller, N. S., & Kipnis, S. S. (2006). Detoxification and substance abuse treatment. A treatment improvement protocol–TIP 45. Substance Abuse and Mental Health Services Administration. Center for Substance Abuse Treatment. National Center for Alcohol Research and Education (NCARE): Transition to Recovery - FullText View - ClinicalTrials.gov. (n.d.). Retrieved from https://clinicaltrials.gov/ct2/show/NCT03737864 Trial Number NCT03737864 National Institute on Drug Abuse. (n.d.). How effective is drug addiction treatment? Retrieved from https://www.drugabuse.gov/publications/principles-drug-addiction-treatmentresearch-based-guide-third-edition/frequently-asked-questions/how-effective-drugaddiction-treatment SUBSTANCE USE RELAPSE AND READMISSIONS 27 Navigation Services to Avoid Rehospitalization (NavSTAR) - Full-Text View ClinicalTrials.gov. (n.d.). Retrieved from https://clinicaltrials.gov/ct2/show/NCT02599818 Trial Number NCT02599818 Neuman, B., & Fawcett, J. (2011). The Neuman systems model (5th ed.). Upper Saddle River, NJ: Pearson. Nino, M. D., Cai, T., & Ignatow, G. ( 2016). Social isolation, drunkenness, and cigarette use among adolescents. Addictive Behaviors, 53, 94-100. http://dx.doi.org/10.1016/j.addbeh.2015.10.005 Polit, D. F., & Beck, C. T. (2017). Nursing research: Generating and assessing evidence for nursing practice (10th ed.). Philadelphia, PA: Wolters Kluwer Health. Reif, S., Acevedo, A., Garnick, D. W., & Fullerton, C. A. (2017). Reducing behavioral health inpatient readmissions for people with substance use disorders: Do follow-up services matter? Psychiatric Services, 68(8), 810–818. https://doi.org/10.1176/appi.ps.201600339 Relapse [Def.2]. (n.d.). In Merriam Webster Online, Retrieved from https://www.merriamwebster.com/dictionary/relapse Segrin, C., McNelis, M., & Swiatkowski, P. (2016). Social support indirectly predicts problem drinking through reduced psychological distress. Substance Use & Misuse, 51(5), 608615. https://doi.org/10.3109/10826084.2015.1126746 Schaffer, M. A., Sandau, K. E., & Diedrick, L. (2012). Evidence-based practice models for organizational change: overview and practical applications. Journal of Advanced Nursing, 69(5), 1197-1209. doi:10.1111/j.1365-2648.2012.06122. Shaffer, S. L., Hutchison, S. L., Ayers, A. M., Goldberg, R. W., Herman, D., Duch, D. A., … Terhorst, L. (2015). Brief critical time intervention to reduce psychiatric SUBSTANCE USE RELAPSE AND READMISSIONS 28 rehospitalization. Psychiatric Services, 66(11), 1155–1161. https://doi.org/10.1176/appi.ps.201400362 Sheikh, W. A., Paul, R., Banda, H., Agath, K., & Luty, J. (2017). Impact of brief relapse prevention intervention in patients with alcohol dependence in Zambia. Journal of Substance Use, 22(1), 113-117. https://doi.org/10.3109/14659891.2015.1090494 Spear, S. E. (2014). Reducing readmissions to detoxification: An interorganizational network perspective. Drug and Alcohol Dependence, 137, 76–82. https://doi.org/10.1016/j.drugalcdep.2014.01.006 Substance Abuse and Mental Health Services Administration. (2019). Key substance use and mental health indicators in the United States: Results from the 2018 National Survey on Drug Use and Health (HHS Publication No. PEP19-5068, NSDUH Series H-54). Rockville, MD: Center for Behavioral Health Statistics and Quality, Substance Abuse and Mental Health Services Administration. Retrieved from https://www.samhsa.gov/data/ Taylor, C., Holsinger, B., Flanagan, J. V., Ayers, A. M., Hutchison, S. L., & Terhorst, L. (2016). Effectiveness of a brief care management intervention for reducing psychiatric hospitalization readmissions. The Journal of Behavioral Health Services & Research, 43(2), 262–271. https://doi.org/10.1007/s11414-014-9400-4 Timko, C., Schultz, N. R., Britt, J., & Cucciare, M. A. (2016). Transitioning from detoxification to substance use disorder treatment: Facilitators and barriers. Journal of Substance Abuse Treatment, 70, 64–72. https://doi.org/10.1016/j.jsat.2016.07.010 SUBSTANCE USE RELAPSE AND READMISSIONS 29 Xia, M. & Yang, C. (2019). The relationship among social support, self-esteem, affect balance and loneliness in individuals with substance use disorders in China. Journal of Community Psychology, 47, 1269-1281. https://doi.org/10.1002/jcop.22190 Yalom, I. & Leszcz, M. (2005). The theory and practice of group psychotherapy (5th ed.). New York, NY: Basic Books. Zuehlke, J. B., Kotecki, R. M., Kern, S., Sholty, G., & Hauser, P. (June 2016). Transformation to a recovery-oriented model of care on a veteran’s administration inpatient unit. Psychiatric Rehabilitation Journal, 39(4), 361. https://doi.org/10.1037/prj0000198 SUBSTANCE USE RELAPSE AND READMISSIONS 30 Appendix A Quantitative Studies Table 1 Evaluation Table Citation A Akerele et al., (2017). Reducing readmission rates in inpatient settings. Funding: None stated Bias: Inferred selection bias. No stated conflict. Country: United States Theory/ Conceptual Framework/ Model It is inferred to be a Maximumlikelihood theory based on the use of the likelihood ratio and odds ratio. Logistic regression model. Inferred Health Promotion Model Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Quantitative n: 1,707 CG: N/A Pre-INT RAR calculated from September 2014August 2015. Simple logistic regression analysis. Likelihood ratio test. SAS used. Odds of readmission post-INT was 29.1% lower. OR = 0.71. pvalue = 0.004. LOE: III Design: Longitudinal Cohort Study. Quasiexperimental design. IV: RED and use of patient navigators. The intervention includes a face-to-face “Warm Handoff,” prior authorization with insurance before DC, provision of medications on DC, f/u phone call within 72hrs, weekly call by Patient Navigator for up to 30 days. Purpose: To evaluate the efficacy of EBP interventions such as ReEngineered Discharge (RED) and patient navigators on RAR. EG: 1,707 Setting: A community IP psychiatric unit in Brooklyn, NY. Demographic s: 18-65 yrs old; hospitalized b/t September 2015 and August 2016; 80% AA, 11 % HSP, 62% homeless Post-INT RAR calculated from September 2015August 2016. CI: 0.05 or 95% RAR reduced by 27%. Strength: INT was significant in reducing RAR. Weakness: No control group. Retrospective study. Conclusion: Detailed intervention and significant findings. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS 31 Citation Theory/ Conceptual Framework/ Model Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Hutchison et al., (2018). Care management intervention to decrease psychiatric and substance use disorder readmissions in Medicaidenrolled adults NS It is inferred to be a Maximumlikelihood theory based on the use of the likelihood ratio. Logistic regression model. Quantitative N: 1,724 n CG: 481 RAR based on MH service paid claims data. Chi-square test. H-L (p= 0.99) and Cox & Snell R2 (0.03) for goodness of fit. Logistic regression analysis. SAS 9.3. RAR in EG = 4% compared to 7% in CG. Follow up rates to OP care, EG = 32%, CG = 25%. The odds of readmission for patients with DD were 1.7x (OR) higher than those with an MH or SUD only. LOE: III Design: Quality improvement; QuasiexperimentalTime series design (higher selection bias). IV: 15-30minute recoveryfocused interview + usual care. The interview focused on eight topics: reasons for current readmission, barriers to increasing community tenure, strategies to overcome barriers, plans for DC, strategies for accessing and using medications, use of recovery plan, factors Funding: None stated Bias: Inferred selection bias. No stated conflict. Country: United States Inferred Health Promotion Model Purpose: To improve the use of OP resources and reduce hospital RAR. n EG: 1243 Sample data were collected from 76 MH organizations across Pennsylvania. February 1, 2015December 31, 2016. Setting: IP substance abuse and residential detox facilities Demographic s: -18-64yrs CI: .05 or 95% SD = N/A Application to practice Strength: Large sample size. Specific intervention. The statistical model reflects the desired outcome/measur e. Weakness: No randomization and limited observed variables. Higher selection bias. Cohort study. Conclusion: Current SUD or Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS -Readmission to a psych facility, IP SUD facility, or residential detox within 30 days of a prior admission. Multiple prior IP treatment & detox services. CG: Ma64.9%, EA77.5%, DD73.2% EG: Ma59.7%, EA76.8%, DD74.6% Citation Koval et al., (2016). Implementation of recovery programming on an inpatient Theory/ Conceptual Framework/ Model Plan-DoStudy-Act (PDSA) Model. PDSA model is effective 32 for personal safety, and IP needs for transition to the community. DV: Usual care; defined as DC planning, referral to OP resources, caremanagement supports, and community services. This group did not receive the recovery interview. detox services do not provide much time for adequate and effective DC planning. This intervention may help to bridge that gap. Results demonstrate improvement in RAR. Would be feasible. Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Quantitative n: 1 n CG: N/A RAR. The “Readmission rate by discharge diagnosis related group for thirtyday time frame” Percentages of readmission s to the IP unit were evaluated at Preintervention RAR= 13%. (1 year, 2013) Post LOE: III Design: Quality improvement; Quasiexperimental- IV: 38-page veterans recovery selfhelp book and worksheet. Recovery n EG: N/A Application to practice Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS acute psychiatric unit and its impact on readmission. Funding: Office of Academic Affiliations Bias: Inferred selection bias. No stated conflict. Country: United States for quality improvement s. Time series design (higher selection bias). Recoveryoriented model. Purpose: To evaluate the impact of a recovery program on RAR. Setting: 16 bed acute psychiatric IP unit Demographic s: Pts with mental health diagnoses. No other demographics used. CG: N/A EG: N/A 33 principles added to Kardex and nursing report sheet. Daily unit schedule changed to 4 hours of recovery programming. report was used to gather information from this specific VA medical center. “Admissions, discharges, and transfers” report was used for data comparison. Data gathered from the U.S. Department of VA three different times: 2013, 2014, and 2015. Initial data collected prior to intervention . intervention RAR = 9.0% (2 year, 2014) post intervention RAR= 7.4% (3 year, 2015) SD = N/A Strength: Data was collected over a three Yr period. A steady decrease in RAR. Weakness: No control group. Quality improvement study. Potential for many confounding variables. No demographics/ patient information was obtained. No info was shared from IP to OP treatment providers. No demographics. Limited quantitative analysis of data. Conclusion: Results indicate a reduction in RAR but because the Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS Citation Maarefvand et. al, (2015). Communitybased relapse prevention for opiate dependents (CBRP): A randomized communitycontrolled trial. Funding: Rebirth Society. Bias: Blinding not possible. No stated conflict. Country: Iran Theory/ Conceptual Framework/ Model None stated. Inferred Health Promotion Model Design/ Method Design: Double-arm randomized controlled trial Purpose: To evaluate the effectiveness of a relapse prevention intervention among opiatedependent patients. 34 Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results n: 71 IV: CBRP consists of five components: patient and family engagement (MI), community assessment, community mobilization, organizing community team, and CBRP planning. CBRP intervention plus f/u phone calls for 3- Family and social support questionnaire. BIOSENS- rapid on-site testing to determine the amount and type of narcotic at 45 and 90 days post-discharge. T-test analysis comparing demographi cs of CG v EG. Variables: age, education, age of abuse initiation. CG: 45 days post DC maintained abs: 41.7%, 90 days: 44.4%. Nearly all in CG relapsed within 45 days post DC. n CG: 36 n EG: 35 Setting: Seven shortterm residential abs-based treatment centers. Demographic s: 18yrs or older. Successful completion of a 28-day residential abs Results from the drug tests were considered the main outcome. Analyzed EG: 45 days post DC maintained abs:77.1%, 90 days: 77.1% design is not strong, confidence is limited. The specific intervention allows for feasibility. Decision for Use LOE: II Application to practice Strength: Detailed INT. RCT. Evaluated two different time intervals. The study supports literature with similar INT. Weakness: Potential for many confounding variables. Limited Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS program. DSM-IV criteria for opiate dependence. Navigation Services to Avoid Rehospitalization (NavSTAR) Full-Text View ClinicalTrials.go v. (n.d.). using Chi square tests. DV: F/U services as usual + F/U phone calls 3months postDC. CI: .05 or 95%. Twotailed test. Qualitative data via interviews w/ community members. SD = N/A demographic data provided. Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Design: Randomized controlled trial n: 400 Time to rehospitalization and 30-day RAR. N/A N/A LOE: II Purpose: To compare the effectiveness of Navigation Services to Avoid Rehospitalizatio EG: N/A IV: The Patient Navigator will work with patients for up to 3 months post-hospital discharge to resolve internal barriers (e.g., EG: Theory/ Conceptual Framework/ Model Andersen’s theoretical model of health service utilization. months postDC. Conclusion: Demonstrates benefits of detailed plans and community involvement. The scope of INT is vast and involves many components that may make it difficult to implement and control variables. Decision for Use CG: Citation 35 CG: N/A Setting: Large urban hospital. Demographic s: 18yrs or Strength: Detailed intervention. RCT. Demonstrates need to improve RAR. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS Funding: Friends Research Institute, Inc. n (NavSTAR) vs. Treatmentas-usual (TAU) for patients with co-occurring medical problems and SUD. Bias: N/A Country: United States Citation Theory/ Design/ Method older, current SUD diagnosis. Sample/ Setting 36 ambivalence about treatment; low motivation; competing life demands, etc.) and external barriers (e.g., lack of transportation; lack of ID card, etc.) to appropriate utilization and engagement in addiction treatment and medical care. Interventions include motivational interventions and patient navigation with proactive case management, tailored to participants' specific needs. Major Variables & Definitions Weakness: Study is still in progress, so no data on the effectiveness of the intervention is available. Conclusion: Pt navigators f/u for three months which would be difficult. Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS National Center for Alcohol Research and Education (NCARE): Transition to Recovery - FullText View ClinicalTrials.go v. (n.d.). Conceptual Framework/ Model None stated. Funding: University of Colorado, Denver Design: Randomized controlled trial n: 700 Purpose: To evaluate the effectiveness of patient navigation for increasing enrollment in SUD programs and preventing readmission. EG: N/A CG: N/A Setting: Unknown Demographic s: 18yrs or older. Bias: N/A Country: United States Citation Theory/ Design/ Method Sample/ Setting 37 IV: A single 45-60 minute 1:1 session of MI provided by patient navigators at discharged focused on transitioning patients to treatment after detoxification plus patient navigation for 30 days, or until the patient is successfully enrolled in substance abuse treatment, or readmission to detoxification occurs, whichever occurs first. Major Variables & Definitions Transition to substance abuse treatment and readmission to detoxification. N/A N/A LOE: II Strength: Detailed intervention. RCT. Demonstrates need to improve RAR. Weakness: The study is still in progress, so no data on the effectiveness of the intervention is available. Conclusion: Intervention may be too lengthy therefore more challenging to implement. Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS Shaffer et al., (2015) Brief critical time intervention to reduce psychiatric rehospitalization Funding: Community Care Behavioral Health Organization Bias: Inferred selection bias. No stated conflict. Country: United States Conceptual Framework/ Model NS It is inferred to be a Maximumlikelihood theory based on the use of the likelihood ratio. Logistic regression model. Inferred Health Promotion Model Quantitative n: 373 Design: Quasiexperimental investigation; Time series design (higher selection bias). n CG: 224 Purpose: The purpose of the study was to examine the association between BCTI and its impact on RAR. n EG: 149 Setting: Six community based behavioral health organizations. Demographic s: >18yrs old DD CG: Ma-48%, Fe-52%; AA43%, CAU56% EG: Ma-57%, Fe-43%; AA42%, CAU56% 38 IV: brief critical time intervention (BCTI). BCTI was broken into three phases of implementatio n. Phase 1: Assessment of immediate needs and resources. Phase 2: connection to community resources. Phase 3: Transition from ASC to community MH services. RAR Chi-square test. H-L (p= 0.87) and Cox & Snell R2 (0.06) for goodness of fit. Logistic regression analysis. SAS 9.3. CI: .05 or 95% RAR days 1-30: 28% was EG, CG was 47%. RAR for days 31180: EG: 44%, CG: 52% Those in CG were 2.83x (OR) more likely to be readmitted (p< .001). SD = N/A LOE: III Application to practice Strength: Detailed INT. Similar, welldefined sample in regards to CG & EG. Evaluated RAR at two different time intervals. The statistical model reflects the desired outcome/measur e. Weakness: Use of small sample and nonrandomized quasiexperimental design. Cohort study. Higher selection bias. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS 39 Conclusion: Person-centered approach leads to better outcomes. BCTI is effective for reducing early readmissions. Feasible. Citation Sheikh et al., (2017). Impact of brief relapse prevention intervention in patients with alcohol dependence in Zambia. Funding: None reported. Bias: None reported. Theory/ Conceptual Framework/ Model Inferred Health Promotion Model Community reinforcemen t approach Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Quantitative n: 114 Design: Randomized controlled trial n CG: 56 Audit score at intake and 8 weeks post-dc. SPSS V. 20. No other data analysis provided. EG: average time to first drink 51days. Purpose: Evaluate the effectiveness of a brief relapse prevention intervention. Setting: Chainama Hills Hospital. (Alcohol-use d/o leading cause of admission) IV: Treatment as usual + a brief relapse prevention intervention from the WHO mhGAP (education, brief motivational interviewing, involving friends and family) Strengths: Brief 20-minute interview that requires minimal training to administer. Pre and post intervention scores measured. n EG: 58 Demographic s: Meet DSM- DV: Treatment as usual which CG: average time to first drink 10 days. Average audit score for Limitations: Sample was predominately male. Sample may not be indicative of those who have Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS Country: Zambia Citation Taylor et al., (2014) Effectiveness of a brief care management intervention for reducing psychiatric hospitalization readmissions. Funding: Community Care Behavioral Health Organization IV criteria for SUD. 96.5%- M, 1865yrs old, 53.5%- Single. Theory/ Conceptual Framework/ Model It is inferred to be a Maximumlikelihood theory based on the use of the likelihood ratio. Logistic regression model. Inferred Health Promotion Model 40 was detoxification with diazepam and vitamin supplementatio n. CG: frequency of etoh intake at 8 week f/u: EG: 1.3, CG: 8.9. EG: P<.001 limited family support (all had family support). Design/ Method Sample/ Setting Major Variables & Definitions Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Design: Longitudinal Cohort Study. Quasiexperimental design. n: 195 IV: Recoveryfocused bridging strategy which was a one-time interview lasting about 10-20 minutes. Six areas were discussed: the reason for admission; barriers to increasing community tenure; strategies to overcome barriers; crisis RAR Chi-square test. H-L (p= 0.08) and Cox & Snell R2 (0.27) for goodness of fit. Logistic regression analysis. SAS 9.3. CG was 2.44x more likely to be readmitted than those in EG. LOE: III Purpose: To increase engagement in aftercare and reduce early psychiatric admissions. CG: 108 EG: 87 Setting: Large psychiatric IP specialty hospital Demographics : Medicaideligible adults readmitted to IP psychiatric care within 30 days prior. Sociodemograph ic information and MH service utilization was obtained through administrative data and paid MH claims. CI: .05 or 95% Application to practice Strength: Detailed INT. The study supports literature with similar INT. Weakness: Data collection from one IP facility. Limited demographic. No Bias: Inferred selection bias. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS No stated conflict. 18-64yrs old Admissions between April 2011November 2012 Country: United States CG: 54% M, 57% CAU, EG: 46% M, 62% CAU Citation Zuehlke et al., (2016). Transformation to a recovery- Theory/ Conceptual Framework/ Model Recoveryoriented model Design/ Method Sample/ Setting Quantitative n: 352 Design: Quality improvement; n CG: N/A 41 plan development; factors to keep individual safe; and current needs that would assist with the transition. DV: usual care; defined as DC planning, referral to OP resources, caremanagement supports, and community services. This group did not receive the recovery interview Major Variables & Definitions IV: Interdisciplinar y recovery team meetings randomization of the sample. Conclusion: Brief care management is effective in reducing RAR. The study is cost effective and easy to implement. Measurement/ Instrumentation Data Analysis Findings/ Results Decision for Use Outcome measures were: Staff satisfaction, restraints/ Independen t samples T-tests to assess for RAR: p= 0.75. LOE: III Application to practice. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS oriented model of care on a veterans’ administration inpatient unit. Funding: No funding information provided. Bias: Inferred selection bias. No stated conflict. Country: United States Inferred Health Promotion Model QuasiexperimentalTime series design (higher selection bias). Purpose: Examine the impact of recovery interventions on Pt outcomes and IP RAR. n EG: N/A Setting: 15bed acute psychiatric unit Demographic s: Veterans admitted to psychiatric IP unit. CG: None provided. EG: None provided. 42 and unit community meetings (pt.’s also had input); Staff recovery intervention education; Direct patient treatment planning; Recoveryoriented group programming; Peer support via CPS. seclusion use, and RAR (data collected via chart review). changes b/t baseline and intervention . Repeated measures univariate ANOVA. SPSS. Staff satisfaction: p= 0.001. Restraints: p= 0.03 RAR: SD = 2.32 (baseline) SD = 1.40 (interventio n periods) Strength: Data was collected over one year. Detailed intervention. Weakness: Lack of experimental CG. Quality improvement project. Potential for many confounding variables. No demographics/ patient information was obtained. No change in RAR. Limited quantitative analysis of data. Conclusion: INT was very detailed; however, data were inconclusive related to its impact on RAR. Key: AA- African American; Abstinence- abs; ANOVA- analysis of variance; ASC- acute service coordination; BCTI- brief critical time intervention; B/Tbetween CAU- Caucasian; CG- control group; CI- confidence interval; CPS- certified peer specialist; DC- discharge; DD-Dual Diagnosis (both MH and SUD diagnosis); DO-disorder; DV-dependent variable; EA- European American EG- experimental group; Fe- female; F/U- follow-up; HSP- Hispanic; H-L- HosmerLemeshow; INT- intervention; IP- inpatient; IV- independent variable; Ma-male; MH- mental health; MHGAP- mental health general action plan; MImotivational interview; N-number of studies (if SR) or participants in study; n- number of participants (if SR) or number of participants in subset; N/A- not applicable or not available; NavSTAR- Navigation Services to Avoid Rehospitalization; NCARE- Native Center for Alcohol Research and Education; Ptpatient; SAS - Statistical Analysis Software; SD- standard deviation; SUD- substance use disorder; SPSS - Statistical Package for the Social Sciences; VAVeterans affairs; OP- outpatient; OR- odds ratio RAR -Readmission rates; RCT- Randomized controlled trial; RED- Re-Engineered Discharge; WHO- world health organization; Yr/s – Year/s SUBSTANCE USE RELAPSE AND READMISSIONS 43 Appendix B Quantitative Studies Table 2 Synthesis Table Basics Author Akerele et al. 2017 LCS/ III Year Design/ LOE Hutchison et al. 2018 LCS/ III Koval et al. 2016 LCS/ III Maarefvand et al. 2015 RCT/II NavSTAR NCARE (n.d.) RCT/ II (n.d.) RCT/ II Shaffer et al. 2015 LCS/ III Sheikh et al. 2017 RCT/II Taylor et al. 2014 LCS/ III Zuehlke et al 2016 LCS/ III 114 18-65 LR 195 18-64yrs LR 352 N/A MR X X X X Study Characteristics Sample Size Age Bias 1,707 18-65yrs LR Pt navigator Recovery plan RF interview IND RP added to nursing report MI Direct Pt TP Pt education X X 1,724 18-64yrs LR X N/A N/A MR X X 71 18yrs + LR 400 18yrs + N/A 700 373 18yrs + 18yrs + N/A LR Interventions X X X X X X X X X X X X X X X X X X X X X X X Major Findings OOR INT pre- N/A N/A Key: IND- Individual; LCS- Longitudinal cohort study; LOE- Level of evidence; LR- Low risk; MI- Motivational interview; MR- Moderate risk; N/A- Not available/ applicable; NS- Not significant; OOR- Odds of readmission; Pt- Patient; RAR; Readmission rates RCT- Randomized controlled trial; RF- Recovery focused; RP- Recovery principles; TP- Treatment planning; Yrs- years; SUBSTANCE USE RELAPSE AND READMISSIONS OOR postINT RAR Likelihood oof relapse       44 N/A N/A  N/A N/A   NS NS Key: IND- Individual; LCS- Longitudinal cohort study; LOE- Level of evidence; LR- Low risk; MI- Motivational interview; MR- Moderate risk; N/A- Not available/ applicable; NS- Not significant; OOR- Odds of readmission; Pt- Patient; RAR; Readmission rates RCT- Randomized controlled trial; RF- Recovery focused; RP- Recovery principles; TP- Treatment planning; Yrs- years; SUBSTANCE USE RELAPSE AND READMISSIONS 45 Appendix C Figure 1 The Neuman systems model SUBSTANCE USE RELAPSE AND READMISSIONS Appendix D Figure 2 The Iowa Model of Evidence-Based Practice to Promote Quality Care 46 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix E Document 1 Internal Review Board Approval 47 SUBSTANCE USE RELAPSE AND READMISSIONS 48 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix F Document 2 Internal Review Board Modification Approval 49 SUBSTANCE USE RELAPSE AND READMISSIONS 50 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix G Document 3 Participant Recruitment Leter 51 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix H Document 4 Quality Improvement General Timeline 52 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix I Document 5 UCLA Loneliness Scale 53 SUBSTANCE USE RELAPSE AND READMISSIONS 54 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix J Document 6 Time to Relapse Questionnaire 55 SUBSTANCE USE RELAPSE AND READMISSIONS 56 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix K Document 7 Demographics Questionnaire 57 SUBSTANCE USE RELAPSE AND READMISSIONS 58 SUBSTANCE USE RELAPSE AND READMISSIONS 59 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix L Document 8 Proposed Budget 60 SUBSTANCE USE RELAPSE AND READMISSIONS Appendix M Figure 3 Flowchart of Participant Enrollment and Retention 61 SUBSTANCE USE RELAPSE AND READMISSIONS 62 Appendix N Table 3 Frequency Table for Nominal Variables Variable Gender Female Male Missing Age_Range 21-29 30-39 Missing Relationship_Status Divorced Married Single, cohabiting Single, never married Missing Education GED High School Less than high school degree Some college Missing Note. Due to rounding errors, percentages may not equal 100%. n % 2 3 0 40 60 0 1 4 0 20 80 0 1 1 1 2 0 20 20 20 40 0 2 1 1 1 0 40 20 20 20 0 SUBSTANCE USE RELAPSE AND READMISSIONS 63 Appendix O Table 4 Frequency Table for Nominal Variables Variable In_Crisis N Missing Working N Y Missing Outpatient_services Y Missing Note. Due to rounding errors, percentages may not equal 100%. n % 5 0 100 0 1 4 0 20 80 0 5 0 100 0 SUBSTANCE USE RELAPSE AND READMISSIONS 64 Appendix P Table 5 Reliability Table for UCLA 72hr Scale No. of Items α Lower Bound Upper Bound UCLA 72hr 3 0.91 0.81 1.01 Note. The lower and upper bounds of Cronbach’s α were calculated using a 95.00% confidence interval. SUBSTANCE USE RELAPSE AND READMISSIONS 65 Appendix Q Table 6 Two-Tailed Paired Samples t-Test for the Difference Between UCLA_72hr and UCLA_2week UCLA_72hr UCLA_2week M SD M SD t p 3.67 0.58 3.33 0.58 1.00 .423 Note. N = 3. Degrees of Freedom for the t-statistic = 2. D represents Cohen’s d. Figure 4 The means of UCLA_72hr and UCLA_2week d 0.58 SUBSTANCE USE RELAPSE AND READMISSIONS 66 Appendix R Table 7 Two-Tailed Paired Samples t-Test for the Difference Between Sudden_72hr and Sudden_2week Sudden_72hr Sudden_2week M SD M SD t p 5.33 2.08 4.33 1.15 1.73 .225 Note. N = 3. Degrees of Freedom for the t-statistic = 2. d represents Cohen's d. Figure 5 The means of Sudden_72hr and Sudden_2week d 1.00 SUBSTANCE USE RELAPSE AND READMISSIONS 67 Appendix S Table 8 Two-Tailed Paired Samples t-Test for the Difference Between Short_72hr and Short_2week Short_72hr Short_2week M SD M SD t p 5.00 1.00 5.67 0.58 -2.00 .184 Note. N = 3. Degrees of Freedom for the t-statistic = 2. D represents Cohen’s d. Figure 6 The means of Short_72hr and Short_2week d 1.15 SUBSTANCE USE RELAPSE AND READMISSIONS 68 Appendix T Table 9 Results for Linear Regression with Short_72hr, Sudden_72hr, and Long_72hr predicting UCLA_72hr Variable B SE CI β t (Intercept) -0.28 7.17 [-91.38, 90.81] 0.00 -0.04 Short_72hr 0.34 1.59 [-19.92, 20.60] 0.28 0.21 Sudden_72hr 0.36 1.20 [-14.89, 15.62] 0.40 0.30 Long_72hr 0.08 1.48 [-18.75, 18.90] 0.06 0.05 2 Note. CI is at the 95% confidence level. Results: F(3,1) = 0.32, p = .825, R = 0.49 Unstandardized Regression Equation: UCLA_72hr = -0.28 + 0.34*Short_72hr + 0.36*Sudden_72hr + 0.08*Long_72hr p .975 .865 .814 .967