Naloxone Telephone Outreach Program in an Outpatient Mental Health Clinic: A Quality Improvement Project

Abstract Over half of veterans diagnosed with OUD have experienced an overdose leading to hospitalization or death. Naloxone is an opioid receptor antagonist that reverses the effects of opioid overdose. Telephone outreach improves naloxone access in high-risk populations. In June 2022, 47.37% of at-risk veterans in an outpatient facility were prescribed naloxone which was significantly lower than the 65.10% national average of other facilities. The nurse-led intervention team implemented the telephone outreach from June 7, 2022, to September 7, 2022. The naloxone prescription rate for veterans diagnosed with OUD in the facility increased from 47.37% in June 2022 (n = 19) to 82.35% in September 2022. The rate for at-risk veterans increased from 20.0% in June 2022 (n = 90) to 58.24% in September 2022. The telephone outreach exceeded goals and increased the naloxone prescription rates for at-risk veterans at the facility in three months.


Introduction
Many veterans have been diagnosed with opioid use disorder (OUD) in the United States.Furthermore, overdoses affected 62.1% in 2016 (Peltzman et al., 2020).Naloxone reverses the effects of an opioid overdose and saves lives (Epocrates, 2022).Outreach programs have improved naloxone access in high-risk populations over the past 20 years, made it available outside medical settings, and improved accessibility (Haegerich et al., 2014).Consequently, people at-risk can correctly identify opioid overdose and use naloxone for treatment (Haegerich et al., 2014).There was no increase in drug use or high-risk behavior with naloxone use (Haegerich et al., 2014).Naloxone nasal spray, a user-friendly mediation, could be successfully administered by over 90% of laypersons without prior training (Ryan & Dunne, 2018).
Approximately 1080 veterans in this regional health system were at-risk for opioid overdose between 2020 and 2021, and 29% to 48.7% of those veterans filled naloxone prescriptions (Facility Intranet, 2022a, 2022b).In June 2022, 47.37% of veterans with OUD in this facility were prescribed naloxone, which was lower than the 65.10% national average among other facilities (FI, 2022c).Pharmacists in other facilities used a telephone outreach program to improve naloxone distribution to at-risk veterans as part of the Opioid Overdose Education and Naloxone Distribution (OEND) initiative (Hoefling et al., 2021;McQuillan, 2022;Yates et al., 2018).No comparable programs existed in this project site.Psychiatric mental health nurse practitioners (PMHNPs) had prescriptive authority over naloxone and were well positioned for telephone outreach.This quality improvement (QI) project aimed to use the Getting to Outcomes (GTO) framework to identify and implement a nurse-led intervention at the project site to improve naloxone access for at-risk veterans with OUD.No baseline data were obtained as historical data were unavailable for retrospective collection.Furthermore, the intervention team's schedule was limited.A fishbone diagram was used to depict the root-cause analysis (Figure 1).
Our paper makes a significant contribution to the literature because few previous studies examined nurse-led telephone outreach interventions or protocols to address naloxone prescription rates among veterans diagnosed with OUD.Even fewer studies published a telephone protocol for naloxone outreach.

Search process
The QI question was whether telephone outreach increased the number of naloxone prescriptions over a three-month period compared to non-telephone outreach.A literature search was performed using CINAHL Plus, APA PsychInfo, APA PsycArticles, Psychology, and Behavioral Sciences Collection, MEDLINE, and ERIC databases.Although no publication date limits were imposed, all papers were published within the past five years.A combination of search terms included "naloxone telephone outreach, " with specifications, such as "full text, " "peer-reviewed, " "English language, " and "academic journals." In total, 31,954,923 articles were retrieved.Search terms were changed to "naloxone, " and "and telephone, "and veteran" were added, which narrowed the results to 11 articles.After duplicates were removed, the search yielded seven articles.

Criteria and appraisal process
The search did not yield any systematic reviews, meta-analyses, or randomized controlled trials.Best evidence articles were identified via a search for experimental and non-experimental studies.Inclusion criteria were articles that discussed naloxone telephone outreach.Articles that did not discuss naloxone telephone outreach or reported naloxone-buprenorphine combination therapy were excluded.Combination therapy was excluded as buprenorphine was a controlled medication and required special waivers for prescription at the time of the project.We focused on naloxone as it presented fewer prescription barriers via telephone outreach.Of the seven results, five met the inclusion criteria.Of these, one was a retrospective quasi-experimental study and four were case studies.

Selection of intervention
The five studies utilized several outreach interventions for at-risk veterans.Hoefling et al. (2021) combined a mailed letter with a telephone outreach program.Meanwhile, McQuillan (2022) educated clients both in person and via the telephone.The other three studies used a telephone outreach program exclusively (Szydlowski & Caruana et al., 2018;Wu et al., 2021;Yates et al., 2018).A mailed letter required additional time and increased total costs with marginal benefit (Hoefling et al., 2021).In-person education was ideal for facilitating communication; however, barriers, such as travel and active pandemic conditions, significantly restricted feasibility.Telephone outreach was not always successful in reaching clients.Furthermore, it was inherently less effective in facilitating communication than in person outreach programs (Hoefling et al., 2021;McQuillan, 2022;Szydlowski & Caruana, 2018;Wu et al., 2021;Yates et al., 2018).However, it was a cost-effective intervention used in other facilities, with higher naloxone prescription rates for at-risk veterans than in this facility.Thus, the intervention team selected telephone outreach as the intervention for this QI project.

Rationale
Opioid overdose and OUD accounted for more than $1.02 trillion in 2017 in the United States (Florence et al., 2021).Naloxone reduced that cost by $43,600 per life saved or $56,699 per quality-adjusted years (Acharya et al., 2020;Nielsen et al., 2022).A practice gap existed as the project site lacked the naloxone telephone outreach used in other facilities as part of the OEND (Hoefling et al., 2021;McQuillan, 2022;Yates et al., 2018).The PMHNPs and mental health team facilitated education and prescribed naloxone to at-risk veterans.Telephone outreach improved client outcomes as it (1) reduced opioid overdose mortality, (2) increased the facility's naloxone prescription rate to match the national rate, (3) brought the facility into compliance with the current practice standards that were already used in other facilities, and (4) improved the facility's reputation among veterans and policymakers (Haegerich et al., 2014;Hoefling et al., 2021;McQuillan, 2022;Yates et al., 2018).
The Getting to Outcomes (GTO) Framework was a 10-step program developed by the RAND Corporation for QI implementation, evaluation, and continuous improvement (Rand Corporation, 2022).The framework began with a needs and resource assessment, followed by goals and desired outcomes, and enumerated the specifics that made a QI project successful.The framework also included process and outcome evaluations, and ended with an outline for sustainability.

Specific aims
This QI project aimed to increase the percentage of naloxone prescriptions for veterans, enrolled in the facility, who were at risk for an opioid overdose via an intervention team that conducted 15-minute telephone conversations over three months.The program aimed to increase naloxone prescriptions from 47.37% to the 65.10% national average (FI, 2022c).An at-risk veteran was defined as an U.S. Armed Forces veteran who had an (a) OUD diagnosis, (b) active opioid prescription, (c) discontinued opioid prescription in the past 180 days, or (d) opioid overdose in the past 12 months.Veterans were excluded if they (1) moved to another facility, (2) had a less-than-16-day opioid prescription for outpatient surgery with no other underlying overdose risk indicators, (3) were admitted to an inpatient rehab facility, or (4) had a naloxone prescription within the past 365 days.

Context
The project site was a clinic in a facility that was part of a nationwide hospital system.Bronfenbrenner's ecological model was applied to the cultural context of the project site (Blackstock et al., 2022).At the macro-system level, the facility was part of a nationwide network of hospitals and clinics.At the meso-system level, it served the outpatient veteran population with mental health issues within driving distance of Bexar County, Texas, USA.The QI project existed at the micro-system level within the facility's psychiatric and mental health outpatient clinic.The psychiatrist worked with the PMHNPs, who managed the clinic staff, which included members of the QI project.
The intervention came from collaboration between the student QI project champion and PMHNP QI project committee member, an employee of the facility.The intervention source was internal to the facility.A barrier was the intervention's strength as based on evidence acquired from one non-randomized experimental study and four case studies.To mitigate this barrier, the QI project team referred to the naloxone telephone outreach programs in other facilities.
Evidence showed a strong correlation between telephone outreach and increased naloxone prescription rates.While COVID-19 isolation protocols limited in-person visits, it had little impact on telephone outreach.State funding for naloxone distribution declined in Texas during the intervention period, another potential barrier (Barragan, 2022).Meanwhile, national support for naloxone distribution and other OUD harm reduction methods increased by $1.6 billion via federal funding (Substance Abuse and Mental Health Services Administration (SAMHSA), 2023).
The intervention's primary facilitator was the high national priority of naloxone distribution as part of the OEND initiative.In addition the external policy incentives were favorable.Naloxone access to at-risk veterans was nationally prioritized.Furthermore, the intervention already existed in other facilities.The facility's leadership and staff were strong facilitators of the intervention's implementation.The cost burden was lessened as the telephone outreach was completed by the Veteran Association (VA) Nurse Practitioner Residency Program's PMHNPs and a doctoral nurse practitioner (DNP) student.

Intervention
The intervention team comprised two PMHNP prescribers and a DNP student project champion who was not a prescriber.The team shadowed a clinical pharmacist from another facility, over the facility intranet, to learn the specific telephone outreach protocol.The pharmacist outlined and demonstrated the protocol.The team followed the protocol under the pharmacist's supervision to ensure faithfulness.It included education of at-risk veterans and family or friends who were willing to learn of opioid overdose and its treatments.The education included identification of the signs and symptoms of opioid overdose, naloxone administration for treatment, calling emergency services for support, and follow-up with a mental health provider.
After the team shadowed the pharmacist, they completed telephone outreach independently.Owing to schedule constraints and room availability, telephone outreach was completed every Tuesday.Specific schedules were designed with the intent of meeting at least 12 h per week.Implementation occurred over the course of three months and aimed to reach for approximately 30 clients.The onsite QI project mentor determined the number of clients.Once the project was approved by the Clinic Chief Psychiatrist, the on-site QI project mentor supervised the telephone outreach.The intervention team performed the telephone outreach and evaluated each other to ensure adherence.
The telephone outreach protocol used the facility database to identify at-risk veterans.The team called them, identified themselves, and explained that the call was to help them with naloxone.If the veteran did not answer, the team left a voicemail with a HIPAA-compliant message and phone number to return the call.A maximum of three attempts were made on different days.The team documented whether the client was reached on a shared spreadsheet on the facility's secure intranet.
The implementation team held a debriefing session at the end of each shift.They identified barriers, facilitators, and any deviations.The team also reviewed the total number of new naloxone prescriptions, renewed naloxone prescriptions, discussed at-risk veterans successfully reached within three phone calls, and those attempted.At the end of the 3-month period, the team and on-site advisor met to evaluate the overall process, discuss ways to overcome barriers and maximize facilitators, and apply changes for the next iteration to maximize outcomes.

Study of the interventions
Although randomized, double-blind studies were the gold standard for determining causation, this intervention was based on case studies.Studies showed a correlation between telephone outreach and increased naloxone prescription rates among at-risk veterans (Hoefling et al., 2021;McQuillan, 2022;Szydlowski & Caruana, 2018;Wu et al., 2021;Yates et al., 2018).Conversely, low naloxone prescription rate was correlated with a lack of outreach.Outreach opportunities existed at the prescriber level.

Measures
The project aimed to increase naloxone prescriptions among veterans at risk for an opioid overdose via a 15-minute telephone intervention over 3 months.All process goals were met as planned and the project was completed on schedule.The primary outcome measure was to increase the percentage of naloxone prescriptions from 47.37% to 65.10% (the national rate) by September 7, 2022.The process measure involved (a) three telephone attempts to 90% of the veterans, and (b) a second intervention team member who verified that 100% of the electronic medical record (EMR) documentation was completed without deviation from the set documentation process.The balance measure was to decrease the (1) median duration of phone calls to less than 10 min and (2) percentage of EMR workarounds to 5%.

Analysis
Naloxone prescription rate was the primary outcome measure used to assess the impact of the intervention.If telephone outreach correlated with an increase in naloxone prescription rate from baseline, the intervention was considered successful.If there were no increase in the naloxone prescription rate, the intervention was considered unsuccessful.

Ethical considerations
The implementation involved data collection from veterans, a vulnerable client population (U.S.Department of Veterans Affairs, 2021).Reported data were HIPAA-compliant and nonspecific to individuals.Furthermore, data collection did not pose any ethical concerns.Data were aggregated into the total number of clients who were (1) diagnosed with OUD designated for outreach for new naloxone prescriptions, (2) successfully reached within three phone calls, (3) requesting a new naloxone prescription, and ( 4) not reachable within three phone calls.In addition, the entire data collection, review, and analysis process occurred on an encrypted intranet system.Client information was accessible only through a facility computer with an individual access card.Any attempt to connect an unapproved storage device automatically encrypted and locked it.Furthermore, the information system security team was automatically alerted.Telephone calls were made via encrypted remote access to protect client confidentiality.The Institutional Review Board (IRB) determined that no approval was necessary as the QI project was not regulated research.No conflicts of interest were identified.

Results
An intervention modification was made to expand the definition of at-risk veterans.The initial definition was limited to individuals diagnosed with OUD (Figure 2).During the intervention, the team discovered that the facility served only 19 veterans with OUD, which was less than the original goal of 30 clients.The team considered it unethical to withhold known benefits from other veterans at-risk for opioid overdose.Subsequently, the goal was increased to include all veterans who met the expanded at-risk definition during the second week (Figure 3).
The team expanded the definition of at-risk veterans to include a(n) (1) OUD diagnosis, (2) active opioid prescription, (3) discontinued opioid prescription in the past 180 days, or (4) opioid overdose in the past 12 months.This expanded definition identified 90 at-risk veterans.Naloxone prescription rates for veterans with OUD and at-risk veterans was illustrated in two separate run charts.

Summary of the outcomes
This project aimed to increase the facility's naloxone prescription rate of 47.37% to match the 65.10% national prescription rate.Naloxone prescription rate for OUD clients was 82.35% at the end of the 3-month intervention period.The prescription rate of at-risk veterans increased almost three-fold to 58.24%.All process and balance measure goals were met.Regarding process measures, the percentage of clients who were reached within three telephone calls was 92.5%.Furthermore, the intervention team verified 100% of the EMR documentation.For the balance measures, the median duration of phone calls was 6 minutes.No EMR workarounds occurred throughout the project.

Interpretation of outcomes
When the intervention began, 47.37% (sample size n = 19) of the veterans at the facility were diagnosed with OUD.In addition, 20.00% (sample size n = 90) of all at-risk veterans received naloxone kits.Each phone call ranged from 5-17 min and required 15 min for screening, 15 min for documentation, and 5 min for documentation verification in advance.Hence, a single new naloxone prescription required 40-52 min.
The facility's naloxone prescription rate sharply increased after the intervention began.This increase was observed in the OUD and at-risk client groups, which suggested that the intervention was effective.After the intervention period ended, the prescription rate declined within three weeks.However, naloxone prescription rate in the OUD group continued to increase for 5 weeks (Figure 2).
An explanation for this unexpected increase was that providers could have changed their practices due to telephone outreach.Additionally, a new cohort of PMHNPs arrived at the facility.Hence, they may have had increased awareness of prescribing naloxone owing to of the visibility of the QI project.While these two possibilities could explain the increased prescription rate, the expanded at-risk group showed a steady decline in their naloxone prescription rate for seven weeks (Figure 3).The at-risk group included lower-risk clients who had an active opioid prescription or discontinued opioid prescription in the past 180 days.Consequently, the providers were less likely to provide naloxone prescriptions to this group.
The primary outcome was consistent with those of other studies.A 2021 case study in Pittsburgh, Pennsylvania increased the average annual naloxone prescription rates by 445% after telephone outreach over two years (McQuillan, 2022).A pharmacy team in Saginaw, Michigan also increased their naloxone prescription rate from 14% to 77% per month after telephone outreach (Szydlowski & Caruana, 2018).Similarly, pharmacy staff at a VA hospital in Madison, Wisconsin also reported increased naloxone prescription rates after telephone outreach (Yates et al., 2018).

Limitations
Ideally, baseline trends should be established several weeks before the intervention.However, historical data were unavailable and the intervention period needed to begin immediately owing to resource and time constraints.The implementation required time to ramp up.In addition, changes in prescription rates were not apparent until after the first 4 weeks.
Selection bias existed since the clients were selected based on the inclusion and exclusion criteria.Blinding and randomization mitigated the risk of selection bias.However, neither the implementation team nor clients were blinded.In addition, the clients were not randomized either.
Denying a known benefit to a population would be unethical.Thus, the implementation team added new clients as the QI project progressed.In total, 31 clients were added to the OUD and at-risk groups between June 7 and September 7, 2022.Simultaneously, 32 clients were lost to attrition.The weekly attrition rate was approximately equal to the gain rate.The final sample size for both groups on September 7, 2022, was n = 91 in the at-risk group and n = 17 in the OUD group.Reasons for attrition were sometimes found in EMR and included (a) moved away, (b) institutionalization, (c) homelessness without access to telephone communication, or (d) death.However, no information was available regarding whether death was due to an overdose.
Given the small sample size, attrition bias was a concern, particularly in the OUD group (n = 19).Three clients in the OUD group were also lost to attrition.However, one client was added to the OUD group.The small sample size made the results and interventions less generalizable.
The telephone outreach protocol was based on a template used by a clinical pharmacist at another facility.A protocol should be based on evidence, such as systematic reviews.However, the literature review identified no evidence-based telephone protocol for naloxone outreach.The telephone protocol served as a script through which the implementation team formed fluid conversations with individual clients.Minimal deviations occurred during the conversations, particularly when the client asked for clarifications.

Conclusions
Telephone outreach was a practical intervention that correlated with a significantly increased naloxone prescription rate.The facility's naloxone prescription rate increased within the first 4 weeks.Furthermore, the rate plateaued for 7 weeks afterward and formed an S-curve.Owing to the relatively slow decline after the intervention, maintenance costs should be significantly lower.Instead of continuing team intervention, a single person could perform monthly outreach maintenance.
Providers who wish to duplicate these results in their facility should perform telephone outreach once a week for two months.Sustainment would be less intensive and costly.Providers can also proactively prescribe naloxone to clients diagnosed with OUD.In this project, the naloxone outreach was expanded to include those in the at-risk group.However, providers should have conversations with this lower-risk group and establish their needs based on individual clients.The current literature details the mortality rate for veterans who died from opioid overdose, but the number of these deaths specific to the clinic is not available.Future updates to facility databases should include this information.In addition, current studies only show that telephone outreach is correlated with an increase in naloxone prescription rates.Future studies should move toward randomized controlled trials to elucidate whether telephone outreach causes an increase in naloxone prescription rates.

Figure 1 .
Figure 1.Fishbone diagram of root cause analysis.

Figure 3 .
Figure 3. Facility naloxone prescription rates for at-risk veterans (expanded definition).