Excellence PhD Dissertation and DNP Project Award Recipients

Biography: Dr. Emily McAllister began her registered nursing career as an oncology nurse in 1997 after graduating from the University of Oklahoma with a Bachelor of Science in Nursing. In 2003, she graduated with a Master of Science in Nursing from the University of Texas at Arlington as a Family Nurse Practitioner. Dr. McAllister has worked in primary care, retail health, physical medicine and rehabilitation, and occupational health, giving her a broad range of experience and knowledge. In 2019, she obtained a certification in occupational health nursing (COHN). Currently, she works in an occupational health setting where she is the provider for 4,000 employees in a manufacturing, research, and corporate environment. Her greatest academic accomplishment occurred in May 2024 upon graduating with a Doctor of Nursing Practice (DNP) at Texas Woman’s University. While pursuing her DNP, Dr. McAllister presented to national audiences and co-authored a publication in The Journal for Nurse Practitioners with her DNP Project Chair titled “Implications for Utilizing the Mountain Model for EBPQI in DNP Education.” Dr. McAllister’s DNP project titled “Implementing a Workflow Process for Human Immunodeficiency Virus Pre-Exposure Prophylaxis (PrEP): A Quality Improvement Initiative” was published in the September 2024 edition of The Journal for Nurse Practitioners.

Background and Problem Statement: The updated CDC HIV Pre-Exposure Prophylaxis (PrEP) guidelines recommend educating all sexually active adults about PrEP instead of counseling based on self-reported risk or health care provider's perception of HIV acquisition. PrEP counseling did not align with the updated guidelines in a student health clinic. The clinic lacked a systematic process for identifying those eligible for counseling, resulting in providers overlooking the necessity to educate during visits not deemed concerning for HIV acquisition.

Purpose/Objective(s): To measure the effectiveness of a DNP-led workflow process to bring a student health clinic in compliance with the updated PrEP guidelines by identifying and counseling all sexually active patients. This practice change aimed to implement a standardized method for identifying patients eligible for counseling and document counseling using a best practice alert (BPA) embedded in the electronic medical record. Lastly, the evaluation of patients' knowledge and acceptability of PrEP occurred before and after counseling.

Procedures: Quality Improvement (QI) strategies, Plan-Do-Study-Act (PDSA) cycles, during a workflow process change to standardize 1) identification of those eligible to receive counseling (sexually active, not taking PrEP and now living with HIV) and 2) counseling all eligible patients about HIV PrEP. The benchmarking of BPA electronic medical record (EMR) utilization was evaluated to determine its utility for sustainability. The BPA was a multifaceted tool that alerted the provider that counseling should occur. A counseling checklist was embedded in the BPA to support providers during counseling and serve as a reference. In addition, the BPA contained lab and prescription order sets to reduce time and streamline the process if a patient chooses to proceed with PrEP. Also, PrEP information populated into the discharge summary for patient reference after documentation that counseling occurred. A Knowledge and Acceptability Questionnaire measured the effectiveness of counseling on patients' knowledge and acceptability of HIV PrEP.

Outcome(s): This QI project improved the identification of patients eligible to receive counseling, the percentage of patients counseled, knowledge and acceptability of HIV PrEP, and robust utilization of the best practice alert. The new workflow process increased the accurate identification of patients eligible for counseling from 5.1% to 94.9%, and patients counseled increased from 3.0% to 97%. There was a statistically significant increase in the Total Knowledge question scores before counseling (M=3.09, SD=1.71) and after counseling (M= 4.61, SD=.60), t (53) = 7.09, p<.001. The mean increase was 1.52, with a 95% confidence interval ranging from 1.09 to 1.95. Cohen's d effect size (.964) was large. The Total Acceptability questions scores significantly increased from before counseling (M=50.65, SD=5.13) to after counseling (M=54.31, SD=7.04), t (53) =5.18, p<.001. Cohen's d-effect size (.704) was moderately large. Provider utilization of the BPA over 16 weeks was 79%.

Conclusion(s): The DNP-led QI project enhanced the delivery of PrEP counseling in a student health clinic by implementing a sustainable standardized workflow for counseling during office visits. A critical first step in the workflow was creating a mechanism, the BPA, for identifying those requiring counseling while supporting providers to counsel. The counseling provided positively impacted our patients' knowledge and acceptability scores of PrEP. If no workflow process is in place, it can be challenging to remember to counsel due to multiple competing demands and priorities.

Implications for Practice or System Change(s): Clinicians' nonadherence to the 2021 CDC HIV Pre-Exposure (PrEP) Guidelines leads to missed opportunities to counsel all sexually active patients about HIV to increase knowledge and acceptability of this valuable intervention for HIV prevention. Due to the project's success, this process is transferable to other settings, considering the resources, barriers, and challenges specific to the population and setting. For example, in adolescent health, women's health, primary care, internal medicine, and geriatric clinics. The new workflow process aided another clinic in evaluating the improvement of HIV PrEP education in their workflow, potentially becoming system-wide.

Background and Significance: Up to a million or more Americans live while traveling full-time in recreational vehicles (RVs). Many health policies are based on assumptions of stationary residency. Little study has been devoted to the healthcare experiences of Americans who are geographically mobile.

Purpose/Objective(s) and/or Specific Aims: The purpose of this study was to explore the experiences of RV-dwelling full-time American nomads when seeking healthcare in the U.S. and to identify barriers and facilitators of access to care. Levesque’s Conceptual Model of Access to Healthcare was used to generate questions for the interview guide. The principal investigator lived in an RV and traveled to a variety of campgrounds in several states to post recruitment  flyers.

Methods: The exploratory, qualitative design was informed by interpretive description. Interviews were recorded and transcribed verbatim. Thematic analysis was conducted with a phronetic iterative approach using Atlas.ti software.

Results/Findings: Participants included American adults who lived for more than six months of the year in an RV while traveling to different regions of the U.S. Participants were located in a variety of campgrounds in several states including Oregon, Wyoming, Colorado, New Mexico, Texas, and Arizona. Twenty-five RV-dwelling nomads participated in semi-structured interviews. Three overarching themes described the experiences of RVers: overwhelming logistics, don’t need healthcare, and orchestrating a web of care. Some of the barriers identified included difficulty vetting healthcare resources in unfamiliar locations, caregivers’ lack of understanding of the nomadic lifestyle, limited resources in rural areas, and rules limiting care and payment for care by state. Some of the facilitators identified included caregiver collaboration, telehealth, national chains with shared records, information and support from other RVers, and self-advocacy.

Conclusion(s): RV-dwelling nomads are an understudied population that face complexity and barriers to healthcare access related to seeking care in unfamiliar locations, misunderstandings about their lifestyle, and policies that prevent the portability of healthcare across state lines.

Background and Problem Statement: Adolescent mental health disorders are a growing public health issue globally. Though symptoms of mental health disorders were worsening before COVID-19, the pandemic exacerbated this crisis. Because adolescents spend most of their time in school, interventions focusing on mental health education and skill-building, reducing stigma, and increasing helping-seeking behaviors should be prioritized here. In a suburban Arizona school district, high school social workers and counselors are seeing increased social and emotional referrals and are looking for solutions. In 2021, the state of Arizona passed SB1376, mandating mental health education in public schools.

Purpose/Objective(s): This project aimed to evaluate the effectiveness of tMHFA (teen Mental Health First Aid) on high school students’ mental health literacy (MHL), mental health first-aid skills, and stigma perception. The project also qualitatively evaluated students’ opinions of tMHFA education and instruction. The implementation design included a phase two expansion and program sustainability by stakeholders within the existing district curriculum.

Procedures: Partnering with a large urban school district’s social work department, a teen Mental Health First Aid (tMHFA) pilot program was developed. Using the Iowa Model of Evidence-Based Practice (IMEBP) as a guide, the organizational priorities of teen mental health provided the trigger specific to this pilot project implementation. This model is fitting for working with a large public school organization and promoted expansion in the sustainability plan. tMHFA is a classroom-based mental health literacy and skill-building program for students aged 15-18 to promote mental health literacy, help-seeking behaviors, and attempts to reduce stigma while teaching adolescents to help a peer or friend through a mental health crisis. A "train the trainer" program was developed with key stakeholders (school district social workers) and the initial pilot project included eleventh-grade students who received three 90-minute classroom workshop sessions. The district chose one high school within the system for the pilot and provided class instruction time for the intervention. Parental consent and student assent were required to participate. tMHFA was delivered in three 90-minute classes on non-consecutive days. The initial pilot group included a class of nine (n = 9) adolescent students aged 17-18 years old. Using the Mental Health Literacy questionnaire (MHLq), a pre and post-test design was used to evaluate the program quantitatively and qualitatively. Descriptive statistics, two-tailed paired samples t-test, Shapiro-Wilk, and qualitative data were collected by student participants and project stakeholders.

Outcomes: Using the Mental Health Literacy questionnaire (MHLq), a pre and post-test design revealed a significant increase in the students’ mental health literacy, help-seeking behaviors, and decreased stigma (p<0.05). Students reported overall positive experiences based on a subjective question at the end of the survey. One student wrote, “I think all students should get a class like this.”  Project stakeholders will continue the tMHFA using project trained staff who are school district employed social workers to offer the program to all eleventh grade students in the project school. The district plans to expand the program to three additional schools in the next academic school year.  The tMHFA is now integrated into the district curriculum.

Conclusion(s): This pilot project using tMHFA significantly improved mental health literacy, mental health first-aid, and stigma perception. Students enjoyed and valued the programming, which is crucial when providing education to teenagers. The Iowa Model of Evidence-Based Practice provided implementation support to engage stakeholders, start the project with a small pilot, and develop a plan for expansion district-wide. After the original nine-student pilot, the district expanded to all junior-level students in one high school (800 students). It is now planning to deliver tMHFA to 2400 junior-level students in three of its high schools.

Background and Problem Statement: Unanticipated and adverse events are inevitable in healthcare. Second victim phenomenon (SVP), which happens when a clinician experiences personal or professional negative effects after a clinical event, can impact the psychological and physical health of the clinician and compromise subsequent care. One in five UTMB Department of Anesthesiology clinicians surveyed reported experiencing an emotionally distressing clinical event in the preceding 30 days. An alarming 88% knew of a colleague emotionally affected by a work event; only 20% felt these individuals had been adequately supported. Prior to the DNP project, no formal process was in place to facilitate peer support.

Purpose/Objective(s): This project served to implement evidence-informed strategies for emotional peer support and foster clinician resilience within the department of anesthesiology. A needs assessment survey indicated that over 90% of clinicians would likely look to a trusted colleague for support after an emotionally challenging clinical event. The project aim was that at least five UTMB Health Anesthesiology clinicians would participate in the emotional peer support process each week between September 26 – December 19, 2022.

Procedures: A diverse group of peer-nominated clinicians, led by the DNP project coordinator, completed specialized training to learn formal peer support. The pilot team, forU Peer Support, comprised all levels of anesthesia professionals, including physicians, APRNs, and residents. Trained peer supporters (TPS) collected de-identified information from interactions, including the precipitating event, provider type, and satisfaction. Encounter forms included an area for open-ended feedback for ideas to help improve the process, such as adding a badge-buddy with additional identification and tips. The department webmaster and project lead collaborated to add a Peer Support Resource hub to the intranet. Resources, including SVP awareness, coping strategies, emotional first-aid tips, a team roster, and professional mental health services contacts, were made available for 24-7 access. The project was promoted via multiple channels including grand rounds, email communications, and the use of branded badges and brochures. The proactive process included a standardized outreach email from forU to any provider involved with an internal morbidity and mortality (M&M) report. Data were collected and analyzed using a multi-pronged approach via surveys, training evaluations, website analytics, and encounter forms. The project coordinator followed encounter forms and monitored participation using a run chart which included the frequency of encounters and intranet resource access. The continuous quality improvement process utilized a rapid cycle Plan-Do-Study-Act (PDSA) approach to determine whether changes and additional interventions were necessary to achieve the project aim.

Outcomes: Fifteen percent of anesthesia providers at UTMB Health completed the formal training and continue to serve as support champions. Over three months, supporters recorded 33 encounters, with an average satisfaction rate of 4.76/5 stars. The intranet site, intended to destigmatize mental health resources and normalize SVP, was accessed over 100 times with high ratings. Before implementation, only 21% of clinicians agreed that adequate and timely emotional peer support resources were available, and post-implementation agreement increased to 77%. Many incidents warranting support were related to patient care and adverse events, but some were related to individual issues, such as work-life balance and workplace bullying. Survey results showed that awareness of the term second victim (advocated as the primary intervention in mitigating its negative effects) doubled. The project continues to have a strong team champion presence, with 25 clinicians being trained and identifiable by branded badges in the clinical setting.

Conclusion(s): This DNP-driven project was a leap forward in safeguarding clinician well-being and efforts have shown to be impactful and meaningful to clinicians. The project was collaborative and inclusive and used translational methods to improve outcomes for the healthcare team and enhance patient safety. While the incidence of challenging clinical events remain constant, the forU program models strategies to encourage substantial improvements in levels of perceived support. A thriving workforce is a necessary prerequisite to advancing patient safety and care, and the forU program helps ensure a culture of care that puts workforce well-being as a high priority.

Background and Significance: Postoperative pulmonary complications (PPCs) are associated with high rates of mortality and morbidity, intensive care admission, hospital length of stay, and resource utilization. PPCs are attributable to both modifiable and non-modifiable risk factors associated with characteristics of patients, surgery, and anesthesia. Although many PPC risks are not modifiable, intraoperative ventilation parameters can be adjusted to reduce risk. Lung protective ventilation (LPV) has been adapted for intraoperative use to protect pulmonary parenchyma against ventilator-induced lung injury (VILI). Questions remain regarding “how”, “when”, and for “whom” LPV can be used to minimize VILI leading to PPCs.

Purpose/Objective(s) and/or Specific Aims: The overarching goal of the dissertation was to identify personalized care strategies that optimize ventilation parameters in order to reduce PPCs for adult patients undergoing a non-cardiothoracic non-obstetric surgery with general anesthesia and mechanical ventilation. The specific aims were to identify: (Aim 1) phenotypes based on preoperative patient and surgical nonmodifiable risk factors to understand the risk of PPC; and (Aim 2) optimal intraoperative ventilator parameters associated with the most desirable outcome for each phenotype.

Methods: A retrospective cohort observational study leveraging electronic health records (EHRs) was conducted. The study included 34,910 adult surgical patients (≥ 18 years of age) who underwent general anesthesia with mechanical ventilation for non-cardiothoracic non-obstetric procedures between 1/1/2018 and 8/31/2019 at a health system in the Southeast of the United States. The PPC outcome in this study was composed of two properties: PPC diagnosis and postoperative respiratory status (PRS) severity. The International Classification of Disease, Tenth Revision diagnosis codes (e.g., acute respiratory distress syndrome, pneumonia) were used to identify patients who developed PPCs. PRS severity was rated by the level of postoperative requirement for supplemental oxygen and ventilatory support.

For Aim 1, the least absolute shrinkage and selection operator (LASSO) regression was first employed to select variables. Then, the classification and regression tree (CART) was conducted for PPC diagnosis and PRS severity to identify phenotypes. For Aim 2, the stratified logistic regression of PPC diagnosis and multinomial logistic regressions of PRS severity with stepwise selection and forced inclusion of tidal volume (VT) and positive end expiratory pressure (PEEP) were performed within each phenotype. The area under the receiver operating characteristic curve was computed to identify the optimal values of VT and PEEP.

Results/Findings: CART identified 14 subgroups for PPC diagnosis (Phenotype 0 though 6) and PRS severity (Phenotype A thought G). The optimal PEEP and VT values ranged from 4 to 12 cmH2O and from 5.98 to 9.39 ml/kg predicted body weight across phenotypes, respectively. The parameters for Phenotype 0, which associated with the lowest PPC incidence (3.9%), 6.25 ml/kg and 5 cmH2O. Phenotype 5 had the highest likelihood of developing PPCs (71.4%), and their optimal parameters were 7.35 ml/kg and 5 cmH2O. Phenotypes A and C had the highest likelihood of having the most severe PRS level; the optimal parameters for Phenotype A were identified as 7.44 ml/kg and 6 cmH2O, and those for Phenotype C were 7.73 ml/kg and 6 cmH2O. In contrast, Phenotype E was associated with the lowest likelihood (3.3%) of having the most severe PRS, and the optimal parameters were 5.98 ml/kg and PEEP of 8 cmH2O.

Conclusion(s): This dissertation was the first to use CART to classify phenotypic subgroups based on risks for PPCs and classified the adult surgical patients into 14 phenotypes based on non-modifiable preoperative risks. The intraoperative mechanical ventilator parameters should be adjusted based on a patient phenotype to optimize the postoperative outcomes. This dissertation (a) offered a foundation for individualized care to optimize patients’ outcomes, and (b) identified several directions for future research to advance our understanding of care individualization aimed at reducing incidence of PPCs and alleviating PRS severity.

Background and Problem Statement: Patients older than age 65 seeking UCSF Health Emergency Department (ED) services represent 32-34% of total visits, contrasting the 20% national average. Published return to care data in this population (recidivism) at 30 days accounts for 6-28% of visits. Addressing medication management challenges, in real-time, can impact self-efficacy and recidivism during ED medication reconciliation thus improving the patient experience. Contributing to the challenge was 50% of older adults reporting medication non-adherence. As part of the institutional motivation California state law SB1254 required hospitals to complete medication reconciliation for all high-risk patients at any point in the continuum of care. The emergency department, as portal of entry to the health system, was a logical starting point. Published patient awareness of their plan of care, after discharge, across 4 domains varies from 24-64%.

Purpose/Objective(s):  The aim of the project was to determine the impact of a structured medication reconciliation workflow on older adult medication self-efficacy as the primary outcome. As secondary outcomes, recidivism, and awareness of plan of care in the Emergency Department were also examined during data collection for a concordant period. As part of an evidenced-based study there is value in collecting qualitative data regarding interprofessional stakeholder experience. Post study lessons-learned was an opportunity for professional participants to share their insights.

Procedures: Following a pilot of the Medication Management Deficiencies in the Elderly (MedMalDE) a modified version was utilized during medication reconciliation. This was a convenience sample of 21 older adult patients assessed over four contiguous weeks in an academic 35-bed emergency department (ED) with 10 additional observation beds. Concurrently the Self-Efficacy for Appropriate Medication Use (SEAMS) confidence survey instrument was used as the outcome instrument applied prior to medication history or reconciliation in coordination with the pharmacy technicians. Additionally, 164 patients were assessed with only the MedMalDE over the same period without self-efficacy measurement due to a variety of factors. All patients were age 65 or older and primarily responsible for their medication management in this pre- and post- design. Primary language was not a barrier with video translation services at the bedside. Reconciliation interventions occurred at the index ED visit and the SEAMS was repeated via telephone at seven to fourteen days following their return to home whether discharged from the ED or post-hospitalization. A secondary outcome measure included a survey of 56 patient’s awareness of ED plan of care collected concurrently and during a pre-intervention period for comparison. Aggregate 30-day recidivism for study patients was another secondary outcome. It was assessed at 39 days prior to, during, and after the pre/post intervention to match the SEAMS four-week data collection and 7–14-day follow-up periods.  Descriptive, Wilcoxon Signed Rank, and Mann-Whitney U Tests were used for statistical analysis where appropriate.

Outcomes: Descriptive statistics revealed patient demographics were similar for intervention and aggregate ED patients. Mean SEAMS scores were significantly improved following MedMalDE guided interventions using the Wilcoxon Signed Rank Test (Z=-2.6989, p=.0077). 30-day ED recidivism for intervention patients was 9.5% versus aggregate patients at 12.3% but not significantly different using Mann Whitney U. The qualitative data from 7-14-day SEAMS follow-up highlight the continued gaps in health system support for the post-discharge experience. The thematic qualitative findings from interviewing the Department of Pharmacy professionals highlighted their satisfaction with improvement study, insights into common medication self-management challenges, and value of caregiver participation.

Conclusion(s): A structured medication reconciliation for older adult ED patients can be incorporated in standard pharmacy workflows, identify common medication challenges addressed in real time, and improve self-efficacy. The qualitative findings collected during the post SEAMS telephone follow-up revealed the lack of uniformity of post-discharge problem-solving with patients and caregivers despite existing professional discharge support. Health systems need to adopt best practices to mitigate health care return to care by addressing gaps in the plan of care communication, assessment of functional, access, and knowledge deficits in medication management, and ensure patient discharge transitions are better supported.

Background and Significance:
Alzheimer’s disease (AD) affects not just the individual but also families, providers, and society. Early recognition and diagnosis of AD may reduce cost by reducing interaction with the health care system, earlier initiation of treatment, and prolonging time to long- term care. Primary care providers, the first contact for many with concerns for memory, are not fulfilling the potential of early diagnosis. Biomarkers of AD emerge 15 to 20 years before clinical diagnosis, yet currently established biomarkers are not easily available in the primary care setting. A growing body of literature is focused on identifying non-invasive early signs of AD.

Purpose/Objective(s) and/or Specific Aims:
The overall aims of this program of research were to understand factors contributing to AD diagnosis variability in primary care practice settings and to find ways to improve early AD diagnosis by primary care providers. Four projects were undertaken to meet the aims and each resulted in a published manuscript.

Methods:
Four studies were undertaken to achieve the aims. The first study reported the results of an integrated review estimating the prevalence of missed diagnosis in primary care when compared to trained raters’ diagnoses. The findings call to attention the difficulty primary care providers face to detect and diagnose AD at all levels of the healthcare system.

In order to understand the pain experience in those with worsening cognition, the second study was a secondary analysis of a cross-sectional age- and sex-matched two group cohort study and found that the experience of pain differs between males and females as a measure of cognition worsened suggesting a possible role of pain as a tool to distinguish those at risk for AD.

This finding led to the third study, which was a narrative review conducted to describe how alterations in senses have been associated with the diagnosis of AD. The results suggested differences in smell, taste, vision, hearing, and proprioception were associated with different levels of the AD continuum but points out an obvious gap in the literature concerning other senses. This led to the fourth study examining evidence that the e4 allele of Apolipoprotein E modifies the experience of pain in those individuals carrying the allele such that greater temperatures are required to elicit pain and the experience of that pain is more unpleasant.

Results/Findings:
The first published manuscript, an integrative review, highlights the complexity of barriers that exist at each level of the health care system contributing to limit diagnoses of AD by PCPs. One study suggested that the complexity of the patients’ presenting problems makes it difficult to diagnose AD. The results of the second study demonstrate that the experience of pain, a common comorbidity seen in AD, differs between males and females as a measure of cognition worsens suggesting pain may be a useful marker of AD. The third study, a narrative review demonstrates differences in senses have been associated with the different diagnostic categories of AD.The fourth study found individuals at increased risk of AD experience pain based on their APOE4 status at a greater temperature and found that pain was more unpleasant when compared to those without the increased risk of AD.

Conclusion(s):
The search to identify non invasive preclinical markers of AD would allow for earlier interventions and the potential use of disease modifying medications once approved in the primary care setting. Older adults and their loved ones often seek care from their primary care providers with initial concerns for memory. However, primary care providers’ diagnostic accuracy is poor and it takes about 17 years for research to catch up to in clinical practice (“2020 Alzheimer’s disease facts and figures” 2020). The evidence of this research suggests pain as a possible early marker of AD.

Background and Problem Statement:
Despite the safety and efficacy of vaccines, US immunization rates have been dropping for decades. In Washington state, MMR vaccination rates had fallen so low that measles outbreaks occurred in 2017. Washington was one of few states allowing personal exemptions for MMR, vaccine mandates for school attendance were not widely enforced, and non-medical exemptions were among the highest in the country. In 2019, state leaders removed personal exemptions for MMR to protect public health. To implement the law, SPS nurses used evidence-based strategies to move families along the vaccine hesitancy continuum, resulting in 97% adherence for six childhood vaccines. The purpose of this project is to describe the tools used and present nurse opinion on effectiveness, to analyze demographic and immunization data to identify equity issues for the 3% non-adherent students, and to disseminate experiences of overcoming vaccine hesitancy in the hope of aiding vaccination efforts in these pandemic times.

Purpose/Objective(s):

• Develop and disseminate a nurse survey on use frequency and effectiveness of campaign tools, capture nurse expert opinion, and glean insight for improvements.
• Characterize nurse ratings of tools and strategies, summarize observations of campaign and risk factors for non-adherence, and summarize insights for improvement.
• Create a demographic and immunization database for two years following state vaccine law implementation.
• Analyze student data, using descriptive statistics and tests for differences.
• Characterize differences in non-adherence compared to all students and describe traits that may predispose students to non-adherence.
• Recommend improvements to protect health and prevent school exclusion.
• Disseminate findings to support vaccination efforts.

Procedures:
The Knowledge To Action and Phenomenological frameworks guided evaluation of the SPS 2019-20 immunization campaign. Since this campaign was conducted primarily by school nurses as end-users, their voices are needed to improve ongoing immunization efforts. The Phenomenological framework guided the search for nurses’ meaning about the campaign. Project objectives were met through the following process:

1. Problem Identification: Impetus was SPS’ desire to evaluate the immunization compliance campaign, seek nurses’ wisdom, analyze non-adherent cohorts for inequity-driven barriers, inform future SPS improvements, and share findings with the broader community.
2. Knowledge Inquiry
   • Reviewed literature and studied 17 articles and RCTs
   • Catalogued tools and strategies used in the SPS campaign
   • Created Excel database of de-identified SPS demographic and Washington State Immunization Information System (WAIIS) immunization data for all non-adherent students
   • Sought guidance and expert opinion from local, state, and national vaccine experts
3. Knowledge Creation
   • Surveyed SPS nurses about tool use/effectiveness and observed risk factors for non-adherence. Open narrative option captured anything nurses wished to convey. 
4. Knowledge Synthesis
   • Analyzed and characterized qualitative nurse survey data with REDCap
   • Analyzed and characterized quantitative demographic/immunization data in Python using descriptive statistics and ANOVA tests for differences (p-value<0.05) among non-adherent subgroups
     • District-wide data were compared with non-adherent student data before and after 2019 laws.
     • Data: gender, grade, ethnicity, primary/home language, birth country, English language learner, disability, advanced learner, Running Start, foster, homeless, immigrant, equity tier, and adherence for DTaP, Tdap, Polio, HepB, MMR, and Varicella.
   • Created and disseminated communication tools collaboratively with SPS

Outcomes:
School Nurse Survey (60% response rate): Most frequent strategies: nurse outreach (families, students, providers). Most effective strategies: Head Start collaboration, conversations with families, Instructional Assistant outreach. Risk factors for non-adherence: family beliefs, low/high incomes, non-English home language, person of color, immigrant, homeless, distrusting health care. Non-adherence often result of barriers to access.

Student Data:
Non-adherence decreased across time for all groups. Traits significant for non-adherence: Grade Level, immigrant, English Language Learner, Homeless, Ethnicity, Primary/home language, Equity Tier. Hispanic, African American/Black, and Native Hawaiian/Pacific Islander students not coming into adherence as quickly as White, Asian, Multiracial, and Alaska Native/Native American peers.

Conclusion(s):
Practice inquiry shows nurses employed evidence-based practices to reach 97% immunization adherence. A variety of tools were used in a strategy of tailored outreach with increased intensity to prevent school exclusion. Nurse perceptions of risk factors for non-adherence were supported by data. Nurses express concern that enforcing laws requiring adherence to remain in school may have damaged relationships with students and families. Data support that many non-adherent students come from communities with historic and current experiences justifying distrust in health care systems. Non-adherence may be more a result of barriers to access than negative beliefs about vaccines.