Health Systems Research
This section includes Class of 2022 Embark Projects within the Health Systems research areas. This is a wide range of projects including Quality Improvement, Patient Safety, and Health Systems studies.
Development of A Deep Neural Network for Synthesis of Non-Contrast Cranial T1-Weighted Magnetic Resonance Imaging
Agueda M. Taylor, B.S.1, Evan Porter, B.A.2, Thomas Guerrero, M.D./Ph.D.3
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Research Institute, Royal Oak, Michigan
3Department of Radiation Oncology, Beaumont Health System, Royal Oak, Michigan
INTRODUCTION
Although 122 out of 1000 people in the US have MRI’s done each year, there are over 4 million with contraindications that subsequently forgo the diagnostic benefits. Studies in recent years have implemented artificial intelligence (AI) algorithms such as deep neural networks (DNN) for production of synthetic medical imaging. The goals of this project are to develop a DNN, specifically a Generative Adversarial Network (GAN) that will predict synthetic Cranial T1 Weighted MRI from non-contrast CT, and to evaluate the model quality.
METHODS
A cohort of 390 sets of non-contrast MR and CT images from Beaumont Hospital’s Gamma Knife Center were utilized to develop a GAN composed of two competing systems: a generator and discriminator. The generator used the original CT images to produce the synthetic MR images, while the discriminator was tasked with determining which image was real when provided the real and synthetic images. synthetic and original MR images to determine if the provided image was real or synthetic. The GAN was created using TensorFlow (version 2.3) and trained on an Nvidia RTX 8000 GPU (graphics processing unit). Image similarity between synthetic and original MRI images was evaluated by calculating the SSIM (structural similarity metric), MAE (mean average error), and NRMSE (normalized root mean square error).
RESULTS
The synthetic MR images had a mean SSIM of 0.881; a value near 1, indicates a strong structural similarity. The MAE of the model was 0.0259; being a smaller value is indicative of higher model quality. The NRMSE of 0.0069 was indicative of the error in the model’s predictive performance.
CONCLUSIONS
The results support that the GAN model was successfully able to produce synthetic non-contrast cranial T1-Weighted MR images with well-defined ventricles, bone and meninges. Areas of improvement include white/grey matter differentiation, interpolating artifacts, and correctly predicting relative intensities.
Transcatheter Aortic Valve Replacement: Volume Impact on Length of Stay and Readmissions
Alexander M. Balinski, B.S.1, Patrick Karabon, M.S.1, Girish Pathangey, M.D.2, Amr E. Abbas, M.D.3
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Internal Medicine, Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire
3Department of Cardiovascular Medicine, Beaumont Health, Royal Oak, Michigan
INTRODUCTION
TAVR is an alternative to surgery in patients with severe aortic stenosis. Procedural volume impacts mortality but little is known of its effect on length of stay (LOS) and readmissions. We sought to examine the impact of transcatheter aortic valve replacement (TAVR) procedural volume on trends in LOS and hospital readmission rates from 2012 to 2016.
METHODS
Using the 2012-2016 Nationwide Readmissions Database (NRD), a total of 102,059 TAVR cases were identified and evaluated for any 30-day all-cause hospital readmissions. We defined annual hospital TAVR procedural volume as high-volume (≥ 100 cases/year), medium-volume (50-99 cases/year), and low-volume (< 50 cases/year). Numbers were discharge-weighted and the NRD survey design was employed. Rao-Scott chi-square tests, t-tests, and logistic regression were used for univariate and multivariate analysis.
RESULTS
From 2012 to 2016, TAVR procedural volume increased (6,924 cases to 39,914 cases), while LOS (9.34 days to 5.35 days; p = < 0.0001) and 30-day readmissions (19.12% to 15.42%; p = 0.0029) decreased. Readmission rate declined for high-volume (17.50% to 15.52%; p = < 0.0001), medium-volume (20.36% to 15.57%; p = 0.0754), and low-volume (19.60% to 15.55%; p = 0.0007) hospitals. Readmissions were significantly lower in high-volume hospitals versus low-volume and medium-volume hospitals combined (15.13% vs 17.71%; p = < 0.0001).
CONCLUSIONS
Higher annual TAVR procedural volumes were associated with decreased average LOS and lower 30-day readmission rates over five years. Hospitals performing TAVR procedures at high yearly volumes were associated with significantly decreased readmission rates. These data suggest improved efficacy of TAVR procedures with increasing procedural volume over time.
Assessing Geriatric Patients in the Emergency Department Using A New Trauma Protocol
Benjamin Travers, B.S.1, Ali Beydoun, B.S.2, Patrick Karabon1, Shanna Jones, M.D.2, Michael Opsommer, M.D.2, Aveh Bastani, M.D.2, David Donaldson, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Emergency Medicine, Troy Beaumont Hospital, Troy, Michigan
INTRODUCTION
The Level III trauma protocol was implemented to reduce time to diagnosis and treatment for elderly patients (defined as patients aged > 65 who are taking anticoagulant/antiplatelet agents) presenting in the ED with a potential head injury after a fall. In 2017, approximately 11.4% of the patients triaged by the new Level III trauma protocol had an intracranial bleed, which is a population that had previously been overlooked as they have a low index of suspicion for an intracranial bleed. This study sought to determine if the recent implementation of the Level III trauma protocol in the Troy Beaumont Emergency Department (ED) has led to better health outcomes for this population.
METHODS
The researchers retrospectively reviewed the charts of patients who met the inclusion criteria and had an intracranial bleed on computed tomography (CT). The patients were split into two groups based on if they were triaged before (N = 12) or after (N = 56) the level III trauma protocol was implemented. Time spent in the ED and patient centered health outcomes were collected and t-test analyses were completed to compare the pre and post level III trauma protocol groups.
RESULTS
The average time from arrival in the ED to CT (2.37 hours) and time spent in the ED (ED LOS) (4.72 hours) for the pre-level III trauma group were significantly longer than the average time to CT (0.64 hours) and ED LOS (2.55 hours) for the post-level III trauma group (p < 0.01). There was insufficient evidence to conclude that there was any difference in health outcomes between the pre and post Level III trauma groups (p > 0.05).
CONCLUSIONS
This suggests that the Level III trauma protocol reduces the time to diagnosis, treatment, and ED LOS for the high-risk elderly population with an intracranial bleed without negatively impacting health outcomes.
Evaluation of Implementing ‘Team Strategies and Tools to Enhance Performance and Patient Safety’ in a Community Emergency Department
Bryson Caskey, B.S.1, Shanna Jones, M.D.2, Sarah Berry, DNP2, Heather Harris, BSN2, David Donaldson, D.O.2, Aveh Bastani, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Health Systems Troy Emergency Department, Troy, Michigan
INTRODUCTION
This study sought to provide an initial evaluation of implementing the program ‘Team Strategies and Tools to Enhance Performance and Patient Safety’ (TeamSTEPPS) in the William Beaumont Troy Hospital Emergency Department (ED). TeamSTEPPS is a program that integrates teamwork into medical practice by developing tools and strategies within leadership, situation monitoring, mutual support, and communication to increase team performance in order to improve quality, safety, and efficiency of healthcare.
METHODS
Validated survey data that assesses staff perceptions of TeamSTEPPS values and the safety culture within the Beaumont Troy ED that was collected before TeamSTEPPS training in January 2017 and compared to survey data that was collected after training in January 2018. All ED staff members and trauma staff who were trained were sampled. Responses were analyzed using descriptive statistics and a student 2-sample T-test.
RESULTS
While it was anticipated that TeamSTEPPS would improve survey responses, post training responses showed mild, yet statistically significant, reductions in average staff responses of the importance of teamwork, leadership, situation monitoring, mutual support, and communication. Analysis of the average difference among post training responses between different staff groups (physicians, physician assistants/nurse practitioners, nurses, and support staff) did not yield significant results. Additionally, staff perceptions of the culture of safety within the ED declined compared to pre-training responses.
CONCLUSIONS
Assessment of survey data alone was insufficient to suggest potential successes of TeamSTEPPS. The limitations of the study, including only using subjective survey data, smaller sample size of post training responses compared to pre-training, ongoing sample changes, and compounding variables affecting staff perceptions/earnestness of responses may explain the results. Future directions include using different methods of evaluation (i.e. patient outcomes-based) or re-administering the surveys to achieve a larger post training sample size, and searching the literature for methods of improving TeamSTEPPS implementation in the Troy ED.
COVID-19 Case Rates and Death Rates versus Cancer Screening Rates
Deanna Tran1, Patrick Karabon1
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
The coronavirus disease 2019 (COVID-19) began in December 2019, culminating into a worldwide pandemic with variable responses, case rates, and death rates between geographic regions and health care institutions. These differences may be partially attributed to distinctions in the health care infrastructures established prior to the pandemic. Primary prevention is focused on prevention of disease occurrence.2 Cancer screening rates reflect multiple elements of healthcare infrastructure, including access, perceived benefits, and social determinants.4 These factors can possibly elucidate differences in COVID-19 case rates and outcomes between different geographic regions and institutions. Our study evaluates the correlation of state-wide cancer screening rates and COVID-19 outcomes.
METHODS
This is a health policy retrospective study using Behavioral Risk Factor Surveillance System (BRFSS), the Center for Disease Control’s telephone survey system that collects state data on health risk behaviors. In the study, we will analyze and compare COVID-19 case rates and death rates to cancer screening rates in states. The general statistical framework for this study has been previously developed and used in examining the effect of medical marijuana laws on marijuana use at the state level.6 The differences-in-differences (DID) methodology accounts for state-level characteristics and time-variant characteristics that are not measured by BRFSS, reducing the potential for missing variable bias.
RESULTS
For Breast, Cervical, and Colorectal screening, states with higher rates of these screenings had reduced rates of COVID-19 cases; however, there was not enough evidence to conclude that these associations were statistically significant (all P > 0.05). For Prostate Cancer screening, states with higher rates of screening had greater rates of COVID-19 cases (P = 0.0364).
CONCLUSIONS
In all screenings, higher rates of screenings at the state-level were correlated with higher COVID-19 death rates. The findings were significant for Breast and Prostate cancer; however, they were not for Cervical and Colorectal cancer.
VIEW POSTER
Pediatric Obesity: A Survey Study of Patients and Caregivers’ Concern and Interest in Specialized Programs
Dina Abdo, B.S.1, Ryan Rogers, B.S.1, Susan Edlibi, M.D.2, Aimee Pollak, M.D.2, Stacy Shubeck, M.D.2, Kerry Mychaliska, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Health, Royal Oak, Michigan
INTRODUCTION
Body mass index (BMI) ≥ the 85th percentile for age and sex is considered overweight in the pediatric population. Pediatric obesity is defined as a BMI at or above the 95th percentile. According to the Centers for Disease Control (CDC), obesity prevalence has tripled over the past 50 years; 1 in 5 school-aged children has a diagnosis of obesity. The purpose of this study is to survey caregivers’ perceptions and concerns about their child’s weight and patient and caregiver interest in a comprehensive healthy lifestyle program.
METHODS
This IRB approved study conducted a ten-question survey of patients aged two to seventeen who presented to the Beaumont Children’s Hospital Pediatric Outpatient Clinic between November 2020 and June 2021. Inclusion criteria were patients aged two to seventeen whose parents voluntarily completed the survey. Exclusion criteria were children under 2 years of age, over 17 years of age, and parents unable to complete the survey due to language barrier or incomplete surveys. Surveys were provided to the families by office staff. Responses were recorded in Excel. Medical history was verified by a physician.
RESULTS
Of 204 surveys administered, 156 met the inclusion criteria and were analyzed. The average age was 8 years old and 83 of 156 (53%) total participants were female. Seventy-four of 156 (47%) had a BMI ≥ the 85th percentile; 57 out of 74 (77%) met the criteria for obesity. Nineteen of the 74 (26%) parents endorsed concern regarding their child's weight. All 19 parents that expressed concern over their child’s weight had children that met the criteria for obesity. Three of 17 (18%) parents of overweight, but not obese, children expressed interest in having their child participate in a healthy lifestyle program, while 25/57 (44%) parents with obese children expressed interest in having their child participate in a healthy lifestyle program.
CONCLUSIONS
Twenty-six percent of surveyed parents with overweight children reported concern regarding their child’s weight. Forty-four percent of parents with obese children reported an interest in participating in a healthy lifestyle program suggesting that availability of such interventions could lead to improved health outcomes. This study highlights the lack of parental recognition of the detrimental implications of childhood obesity and demonstrates an opportunity to investigate health literacy barriers in pediatric obesity.
A Multimodal Approach to Reducing Opioid Administration in the Emergency Department
Eddie Ford, B.A.1, Vito Rocco, M.D.2, Shanna Jones, M.D.2, Patrick Karabon, M.S.1
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Emergency Medicine, Beaumont Health System, Troy, Michigan
INTRODUCTION
The United States is battling a deadly epidemic against opioid abuse and misuse, which began as a result of pharmaceutical companies purposely downplaying the addictive potential of their opioid medications. In 2016, a Colorado health care system developed a groundbreaking alternative to opioids (ALTO) protocol for acute pain relief in the Emergency Department (ED). This protocol was adapted and implemented in the Beaumont Troy ED and is the focus of this current study. The primary goal of this study is to characterize the impact of the ALTO protocol on the rate of opioid and non-opioid administration in the ED for acute pain relief.
METHODS
A retrospective chart review of ED admission/discharge data and medication administration data for adult patients was completed for the control cohort (pre-ALTO) and ALTO intervention group during the study period of April 2018 – August 2019. To ensure accurate comparison of medications used, a standardizing procedure of calculating morphine equivalent units (MEUs) was utilized. Variables were statically analyzed and presented through a log-linear regression model and monthly percentage change (MPC) over time.
RESULTS
During the study period, the rate of MEUs administered decreased by 1.05% per month MPC = -1.05%, P = 0.0018. Compared to the control group, ALTOs administered increased on average by 2.36% per month MPC = 2.36%, P = <0.0001 and the rate of ALTOs administered increased by 1.82% per month MPC = 1.82%, P = <0.0001.
CONCLUSIONS
The results support the hypothesis that the ALTO protocol would lead to a reduction in the rate of opioids used for acute pain relief while also increasing the rate of ALTOs used for acute pain relief in the ED. This research will contribute to the production of universally accepted and preferred non-opioid pain reduction protocols in order to limit future opioid misuse and abuse.
Quality Improvement Study on Human Papilloma Virus (HPV) Vaccination uptake in HIV patients
Hassan Akram, B.S.1, Paul Johnson, M.D.2, Trini Mathew, M.D., MPH2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Division of Infectious Disease and International Medicine, Beaumont Hospital, Royal Oak, Michigan
INTRODUCTION
HPV is strongly associated with cervical cancer and cancers of vagina, vulva, penis, anus, rectum. In 2019, WHO identified vaccine hesitancy as top 10 global threats and planned to increase the HPV vaccine coverage. In June 2019 Advisory Committee on Immunization Practices (ACIP) in the US recommended shared clinical decision-making regarding HPV vaccinations in adults aged 26 through 45 years. The goal of this project is to evaluate the uptake of HPV vaccination and challenges in the HIV patients at Beaumont Hospital.
METHODS
HIV+ patients (ages 18-45) who received care in the Infectious Disease Clinic at Beaumont Hospital were provided a questionnaire on HPV vaccine as part of routine care. Anecdotally patients in this clinic were not routinely offered the HPV vaccine prior to implementation of this student-led Quality Improvement project. The results of 9 items questionnaire, demographic data and HPV vaccination uptake, were analyzed from August 2019 to December 2020.
RESULTS
Thirty-two participants completed the questionnaire of which 12 participants confirmed interest in receiving vaccination and 9 subsequently received the HPV vaccination at the time of analysis (Figure 1). Of note, two other participants had received vaccination previously. 78% of the participants were males and the mean age of the participants was 33.6 years (SD: 5.6).
CONCLUSIONS
The study was limited by transition to tele-medicine clinic during COVID19 pandemic surge. However, it highlighted need for patient-provider discussions to improve HPV vaccination uptake, understand insurance coverage and patient’s perception of the vaccine cost to optimize HPV vaccine uptake.
Cardiac Arrest Patient Length of Stay and Survival in Michigan Hospitals
Jaemin Song1, David A Berger, M.D.2, James H Paxton, M.D.3, Joseph B Miller, M.D.4
Joshua Reynolds, M.D.5, Nai-Wei Chen, M.D.2,
Robert Swor, D.O.1,2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Hospital, Royal Oak, Michigan
3Detroit Receiving Medical Center, Detroit, Michigan
4Henry Ford Hospital, Detroit, Michigan
5Spectrum Health Hospital, Grand Rapids, Michigan
INTRODUCTION
Current guidelines recommend deferring prognostic decisions for at least 72 hours following admission after Out of Hospital cardiac arrest (OHCA). Most non-survivors experience withdrawal of life-sustaining therapy (WLST), and early WLST may adversely impact survival. We sought to characterize the hospital length of stay (LOS) and timing of Do Not Resuscitate (DNR) orders (as surrogates for WLST) to assess their relationship to survival following cardiac arrest.
METHODS
We performed a retrospective cohort study of probabilistically linked cardiac arrest registries (Cardiac Arrest Registry to Enhance Survival (CARES) and Michigan Inpatient Database (MIDB) from 2014 to 2017.
RESULTS
We included 3644 patients from 38 hospitals with >30 patients. The patients' mean age was 62.4 years and were predominately male (59.3%). LOS ≤ 3 days (ORadj = 0.11) and early DNR (ORadj = 0.02) were inversely associated with survival to discharge. There was a non-significant inverse association between hospital rates of LOS ≤ 3 days and survival (p = 0.11) and Early DNR and survival (p = 0.83). In the multilevel model, using median odd ratios to assess variation in LOS ≤ 3 days and survival, patient characteristics contributed more to variability in survival than between-hospital variation. However, between-hospital variation contributed more to variability than patient characteristics in the provision of early DNR orders.
CONCLUSIONS
We observed that LOS ≤ 3 days for post-arrest patients was negatively associated with survival, with both patient characteristics and between-hospital variation associated with outcomes. However, between-hospital variation appears to be more highly associated with the provision of early DNR orders than patient characteristics. Further work is needed to assess variation in early DNR orders and their impact on patient survival.
Differences in Small Bowel Obstruction Outcomes in an Academic vs Community Hospital
Kelly Fahey, M.S.1, Aveh Bastani, M.D.2, Shanna Jones, M.D.2, Philip Kilanowski-Doroh, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Emergency Medicine, Beaumont Health System, Troy, Michigan
INTRODUCTION
A small bowel obstruction (SBO) is a common, costly ailment causing admission to the hospital from the Emergency Department. The management of SBO includes both operative and conservative approaches, and wide variation exists across hospitals regarding the choice of primary admission service. Variation exists because it can be challenging to determine which patients will require surgical intervention and which can be treated successfully with conservative measures. Current studies support primary management of SBO by surgeons because admission to a medical service is associated with a delayed time to operative intervention, longer length of stay, higher cost, and a higher operative mortality rate.
METHODS
A retrospective chart review of 200 adult patients at Royal Oak or Troy with an admission diagnosis of intestinal SBO with adhesions was conducted. Equal numbers of patients were collected for chart review from each hospital. Outcomes of morbidity, mortality, and length of stay between surgical and medical admissions were analyzed.
RESULTS
Only one patient from Royal Oak was admitted to the Medicine service. The aims of the study were adjusted to compare outcomes of surgical service admissions between the two sites. For those who underwent surgery, the 30-day readmission rate was higher at Royal Oak (11.1%) compared to Troy (5.3%). Post-op complications including infection and hemorrhage were not statistically significant between the sites. For patients receiving medical management, the 30-day readmission rate was higher at Troy (21%) compared to Royal Oak (2%). Length of stay and mortality were the same between both sites.
CONCLUSIONS
Readmission rates for patients managed surgically were higher at Royal Oak while patients managed medically were higher at Troy. Future analysis comparing outcomes of patient’s managed by the surgical service vs the medical service should be completed.
The impact of medical scribes on emergency physician diagnostic testing and diagnosis charting
Lucas Nelson, B.S.1, Brett R Todd, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Emergency Medicine, Beaumont Health System, Royal Oak, Michigan
INTRODUCTION
Since the widespread adoption of electronic medical records (EMRs), medical scribes have been increasingly utilized in emergency department (ED) settings to offload the documentation burden of emergency physicians (EPs). Scribes have been shown to increase EP productivity and satisfaction; however, little is known about their effects on the EP’s diagnostic process. We aimed to assess what effect, if any, scribes have on EP diagnostic test ordering and their documentation of differential diagnoses.
METHODS
We conducted a retrospective cohort study utilizing a chart review to compare diagnostic practices of EPs working both with and without scribes. We analyzed the number of laboratory and radiologic diagnostic studies ordered per encounter as well as characteristics of differential diagnosis documentation.
RESULTS
Scribes did not affect laboratory studies ordered per encounter (mean 6.31 by scribes vs. 7.35 by EPs, difference -1.04; 95% confidence interval [CI] -2.34 to 0.26) or radiologic studies ordered per encounter (mean 1.49 by scribes vs. 1.39 by EPs, difference 0.10; 95% CI -0.15 to 0.35). Scribes did not affect the frequency of documenting a differential diagnosis or the number of diagnoses considered in each differential, but they were associated with higher word counts in EP differentials (mean 72.29 by scribes vs. 50.00 by EPs, mean difference 22.79; 95% CI 6.77 to 38.81).
CONCLUSIONS
Scribe use does not appear to affect EP diagnostic test ordering but may have a small effect on their documentation of differential diagnoses.
Outcomes for Patients with Congestive Heart Failure and Chronic Kidney Disease Receiving Fluid Resuscitation for Severe Sepsis or Septic Shock
Melany Wiczorek, B.S.1, Ronny Otero, M.D.2, Steven Knight, M.D.2, Kaitlin Ziadeh3, James Blumline, B.S.1, Zachary Rollins, B.S.1
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Health, Royal Oak, Michigan
3Michigan State University, East Lansing, Michigan
INTRODUCTION
Sepsis core measures are an integral part of sepsis treatment. Current fluid administration guidelines consist of administering at least 30cc/kg of intravenous fluids (IVF) per ideal body weight (IBW) within the first three hours of sepsis diagnosis regardless of pre-existing comorbidities at risk for fluid overload. This study aims to evaluate the outcomes of patients with a history of congestive heart failure (CHF) and/or chronic kidney disease (CKD) who receive fluid resuscitation for the management of severe sepsis or septic shock.
METHODS
We performed a retrospective case-control study of Emergency Department patients treated for severe sepsis or septic shock. We identified subjects with a history of CHF or CKD (at-risk group [AR]) and a sample of patients without a history of CHF/ CKD (control group). We performed a structured chart review recording demographics, fluids received, airway interventions required, and outcome. Logistic regression analysis was used to compare the association between the amount of IV fluids received and the outcomes of interest.
RESULTS
Our cohort consisted of 745 patients with a history of CHF and/or CKD (AR group) and 570 patients without a history of CHF and/or CKD (control group). Overall patients in the AR group received less IVFs than the control group at 24 hours. There was a significant association between receipt of >30cc/kg of IVF in the AR group at 3hrs and 6hrs from ED arrival and the need for BiPAP. There was no statistically significant association between receipt of >30 cc/kg of IVF in the AR group compared to the control group in terms of in-hospital mortality.
CONCLUSIONS
We identified a significant association of >30 cc/kg IVF administration and the need for Bi-PAP in AR patients. We identified higher in-hospital mortality in the AR group, but this was not associated with the amount of IVF resuscitation received.
Investigating the Effects of a Mindfulness Based Stress Reduction Course on a Grouping of Oncology Patients, Health Care Workers and their Families on Measures of Stress and Burnout in Health Care Workers: A Randomized Control Trial
Michael Moussa, B.S.1, Patrick Herndon, B.S.1, Alyssa Heintschel, B.S.1, Scott Sabbagh, B.S.1, Ruth Lerman, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Center for Mindfulness, Beaumont Health System, Royal Oak, Michigan
INTRODUCTION
Mindfulness Based Stress Reduction (MBSR) can improve the mental health of health care workers (HCW). How will the inclusion of oncology patients (OP) in MBSR classes for HCW affect the mental health of HCW?
METHODS
The standard MBSR curriculum teaches meditation, yoga and stress management in 28 contact hours over eight weeks. 18 HCW and two of their family members (FM) were randomized to a study group which also included 27 OP and three of their FM. 19 HCW and two of their FM were randomized to control population. Study and control populations attended simultaneous MBSR courses taught by certified MBSR teachers. HCW completed pre, post, 4 month and 11-month f/u Perceived Stress Scales (PSS). and Maslach Burnout Inventories (MBI).
RESULTS
Average class attendance was 89.9%. Total PSS Scale Scores in the Treatment group significantly decreased between Class 1 and Month 11 (P = 0.0017), but did not significantly decrease in the Control group during this time frame (P = 0.1671).
The total MBI and its subscales portrayed no statistically significant differences between class 1 and month 11. The control group scores on one of 22 MBI questions “I don’t really care what happens to some patients” showed a significant increase while scores in the treatment group score did not (P = 0.0106).
CONCLUSIONS
This research provides evidence that the integration of OP with HCW provides sustainable improvements in measures of HCW mental health and burnout while the HCW-only class participants did not experience these benefits.
Development of Verbal Behavior in Early, Intensive Behavioral Therapy for Autism Spectrum Disorder
Mithya Jayakumar, B.S.1, Lesly Hendershot, PsyD, LP, BCBA-D2, Lori Warner, Ph. D., LP, BCBA-D2, James F. Grogan, Ph.D.3
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Beaumont Children’s Hospital: Center for Human Development, Southfield, Michigan
3Department of Foundational Medical Studies, Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
The current study will explore the relationships between motor imitation (MI) and verbal requesting skills in children who are diagnosed with autism spectrum disorder (ASD). With nearly half of children with ASD evidencing minimal verbal skills, supporting language development is crucial in interventions. Early, intensive behavioral intervention (EIBI) is recommended to help children achieve better developmental outcomes. Functional language training often begins with the verbal requesting skill, MAND. Research suggests that a series of rapid motor imitations before a vocal model could facilitate vocal speech, but little is known about how MI can longitudinally influence functional language development. This retrospective, cohort study with MI skill level subgroups (low, medium and high) will explore whether the development of MI skills correlates with greater MAND progress in pediatric patients with ASD.
METHODS
Data will be gathered longitudinally using the Verbal Behavior Milestones Assessment and Placement Program (VB-MAPP), a criterion-referenced verbal behavior assessment tool and curriculum guide, to explore the relationship between the rate of acquisition with MI and MAND in an EIBI sample. Children ages 2-7 years old from the Ted Lindsay Foundation HOPE center at Beaumont Children’s will have a confirmed diagnosis of ASD, and would have received 15-30 hours of one-on-one applied behavior analysis treatment. VB-MAPP assessment scores include longitudinal, quantitative measures of MI and MAND. A linear mixed model will be used to explore longitudinal MI and MAND attainment.
RESULTS
There was a significant interaction between Baseline MI and Trend in MI Milestones, and there was not enough evidence to conclude an interaction between Baseline MI and Trend in MAND Milestones. Therefore, the trend in MAND milestones is not different between Baseline MI categories.
CONCLUSIONS
The results did not support the hypothesis that there is an intersection between MAND and MI in long-term EIBI for pediatric patients with ASD.
Risk Factors Affecting Delayed Diagnosis of Adult Acute Appendicitis
Nicholas W. Prewitt, B.S.1, Brett R. Todd, M.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Emergency Medicine, Beaumont Health, Royal Oak
INTRODUCTION
The diagnosis of appendicitis is time-sensitive and delays increase the risk of morbidity and mortality through complications, including perforation, gangrene, and abscess formation. Nearly all cases of appendicitis are diagnosed in and admitted through the emergency department (ED). Therefore, improvements in the time to diagnosis of acute appendicitis in the ED may improve patient outcomes. However, it currently remains unclear what factors contribute to prolonged diagnostic times of appendicitis. This study aimed to determine if demographic, environmental, or provider-related factors result in delays in the diagnosis of adult acute appendicitis in the ED.
METHODS
We conducted a retrospective cohort study to evaluate the causes of variability in diagnosis times of acute appendicitis utilizing chart review methodology. Charts were pulled from Epic for all patients diagnosed with acute appendicitis, 18 years and older, and from the Beaumont Hospital, Royal Oak ED in the time period of 2016 to 2018. Variables collected include age, race, sex, primary language, time of day, day of week, time of year, mode of arrival, referral to ED, return ED visit, and the provider experience. Time to diagnosis was defined as the time from ED arrival to diagnosis of appendicitis by radiologist read. Data was analyzed using a univariate generalized linear model.
RESULTS
907 patients met the inclusion criteria. Average time to diagnosis was significantly lower in white patients than nonwhite patients (3.96 versus 4.30 hours; p = 0.005) and in males than females (3.74 versus 4.32 hours; p < 0.001).
CONCLUSIONS
Female and nonwhite patients experience a prolonged time to diagnosis of acute appendicitis. This may reflect confusion with pelvic pathology in females and possibly the effects of implicit bias in healthcare workers. ED providers should be cognizant of these discrepancies in order to avoid increased morbidity and mortality in these population of patients with acute appendicitis.
Hepatitis A and Homelessness: A Systematic Review
Omar Abbas, B.S.1, Muhammad Waheed, B.S.1, Tracey A.H. Taylor, Ph.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Foundational Medical Studies, Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
Hepatitis A outbreaks have increased in frequency over the years, and homeless populations are at increased risk of Hepatitis A Virus (HAV) infection. The primary goals of this study were to summarize characteristics of patients with acute HAV infection, to determine the proportion of homeless patients (compared with non-homeless patients) within HAV outbreaks, and to identify differences in outcomes between homeless and non-homeless patients with HAV infection. We aimed to generate data that would support current HAV vaccination efforts targeted towards the homeless population, as recommended by the CDC.
METHODS
We conducted a systematic review of the literature on Homelessness and HAV using PRISMA guidelines. Following removal of duplicates and based on our inclusion/exclusion criteria, 16 articles were used for our study from an initial search of 401. Data on patient characteristics were tabulated and subsequent analysis of data was completed. A Z-test was performed to discern the statistical significance of our data on hospitalization.
RESULTS
Data extraction from the 16 studies that met the inclusion criteria yielded a total of 6046 patients with acute HAV infection. 27.9% of the patients were homeless, 65.2% of the patients were male, 42.1% had a history of intravenous drug use, and the average age of patients was 44.86. There was a statistically significant difference in hospitalization rates between homeless (77.5%) and non-homeless patients (50.9%).
CONCLUSIONS
Homeless individuals account for a significant portion of acute HAV infection, and have higher rates of hospitalization compared to non-homeless patients infected with HAV. Our data re-emphasizes the importance of HAV vaccination efforts targeted towards the homeless population.
Characterization of the Clinical Impact of a Celiac Disease Algorithm on Diagnostic Workup of Patients at Beaumont Hospital Royal Oak
Omid Vadpey, B.S.1, Gabriel Maine, Ph.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Pathology and Laboratory Medicine, Beaumont Health, Royal Oak, Michigan
INTRODUCTION
Advancements in medical technology and laboratory testing have prompted room for more accurate diagnostic procedures that can aid in diagnostic efficiency, reducing time to diagnosis, unnecessary test ordering, and costs. Celiac Disease is a complex gastrointestinal autoimmune disorder characterized by symptoms that mimic other bowel diseases, such as diarrhea, steatorrhea, and malnutrition. The primary goal of this pilot study is to determine if the implementation of a diagnostic serologic algorithm differs from the current standard of practice for patients that are suspected of celiac disease at Beaumont Health-Royal Oak (BH-RO).
METHODS
94 patient data points were randomly selected from a larger pool of 12,603, who were suspected of celiac disease, within the year 2020 at BH-RO. Patient inclusion criteria includes age greater than or equal to 18, and completed duodenal biopsy via EGD and screened celiac disease serology of IgA tissue transglutaminase antibody, which were collected by respective ICD10 codes. These patients further serologic data were collected, and qualitatively compared to the algorithmic orders that were expected based on an algorithm adapted from Mayo Clinic for Celiac Disease diagnosis. The final data was sorted into groups based on each procedure, and underwent chi-square test for statistical assessment.
RESULTS
Compared to observed practice at BH-RO, chi-square testing demonstrates that the algorithmic approach statistically rejects the null hypothesis, with a difference in order numbers for ordering duodenal EGD biopsies (p=7E-34) and antibody orders for endomysial antigen (p=0.02), deamidated gliadin peptide antigen (p=0.005712).
CONCLUSIONS
The results support the hypothesis that instituting an algorithmic approach to serologic evaluation of celiac disease will reduce the number of orders placed. This data elicits both further concern as EGDs in these patients proved to elucidate other urgent disorders; however, the reduction of serologic tests could benefit lower costs and time for patients’ care.
Investigating the Effectiveness of a Mindfulness Based Stress Reduction Course Within Healthcare: A Qualitative Methods Study
Patrick Herndon , B.S.1, Michael Moussa , B.S.1, Alyssa Heintschel, B.S.1, Scott Sabbagh, B.S.1, Ruth Lerman, M.D.2,3
>1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Internal Medicine, Beaumont Health, Royal Oak, Michigan
3Beaumont Center for Mindfulness, Beaumont Health, Royal Oak, Michigan
INTRODUCTION
Mindfulness practice has emerged as a promising means to address stress in healthcare workers (HCW). Virtually all HCW studies of the standardized, validated Mindfulness Based Stress Reduction (MBSR) curriculum have been conducted using homogeneous (HCW only) populations. This work examines the novel approach of comparing the effects of heterogeneous MBSR classes: HCW, oncology patients and family members (HCW-HE) as compared to homogeneous classes HCW and their family members only (HCW-HO).
METHODS
During 2018, 37 HCW, 22 patients, and 6 family members completed MBSR courses at Beaumont Hospital. HCW had been randomized to HCW-HE or HCW-HO classes. Following the course, written narrative responses to open-ended questions underwent qualitative analysis using Grounded Theory.
RESULTS
15 of the 18 HCW-HE (83%) and 7 of the 19 HCW-HO completed the questionnaires and were included in the data analysis. Narrative questions were selected for analysis based on the relevance to the overall research question, quality of responses, and inter-rater reliability. Themes common to both populations included learning, bonding/shared purpose and pleasantness of course.
Only HCW-HO commented on the difficulty of the course. All HCW-HE expressed a preference for including OP in their classes while the HCW-HE felt that a heterogenous class would not have been preferable. The HCW-HE expressed a stronger sense of bonding, had a higher survey completion rate, and included overwhelming positive reflections as compared to HCW-HO.
CONCLUSIONS
The results demonstrate common and divergent themes between HCW-HE and HCW-HO MBSR courses. Consideration of including patients and families along with HCW in MBSR courses is suggested.
Comparing Incidence of Opioid Prescription in Response to Similar Pain Scales Based on Differences in Race
Ramiz Memon, B.A.1, Jacob Keeley, M.S.1, Tracey A.H. Taylor, Ph.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Foundational Medical Studies, Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
The opioid epidemic is one of medicine’s most widely discussed topics. Previous studies have suggested that the incidence of opioid prescription is higher in white patients compared to black patients. The primary goal of this study is to identify if there are persisting differences in opioid prescription rates between non-Hispanic white and non-Hispanic black patients from 2011-2015. A secondary goal is to examine if private insurance serves as a normalizing factor.
METHODS
Using the National Hospital Ambulatory Medical Care Survey (NHAMCS) database, data was collected from patients’ visits in the emergency department between 2011 and 2015, and organized according to pain scale, race/ethnicity, and prescription of opioids. Specifically, after determining the three chief complaints with the highest incidences of opioid prescription, patients with these chief complaints were organized into non-Hispanic white or non-Hispanic black groups with pain scales of 1-4, 5-7, or 8-10. The corresponding groups were compared with whether or not those patients were prescribed an opioid. These data were analyzed again in patients with private insurance. A complex sample chi square test was used for statistical analysis, using complex survey design techniques considering the weighting procedures included in the dataset.
RESULTS
Compared to non-Hispanic whites, non-Hispanic blacks were less likely to receive opioids in every pain scale subgroup. Specifically, in the 1-4 group, approximately 6% less patients were prescribed, with approximately 10% less in the 5-7 and 8-10 group (p < 0.017). For patients with private insurance, 6%, 13%, and 19% less patients were prescribed opioids in the respective pain groups.
CONCLUSIONS
The results indicate that white patients are favorably prescribed opioids compared to black patients, thereby raising further questions about healthcare inequities in America and subjective treatments in the emergency department which might require further research to establish objective treatment guidelines.
How Is Empathy Articulated, Conceptualized, And Operationalized In Different Health Care Professions? A Systematic Review.
Rossini Batino, B.S.1, Stephen Loftus, Ph.D.2
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
2Department of Foundational Medical Studies, Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
“Empathy” is often defined in similar but not universal ways. This is because it is an abstract term, and is, therefore, highly subjective. Nonetheless, empathy influences the delivery of care in clinical practice. Though there is a consensus that empathic behavior is a vital element in clinical practice there are varying opinions on how empathy develops throughout health professional training. Studies reporting an improvement or decline of empathy might be inappropriate since there are varying definitions of “empathy." Therefore, evaluating, assessing, or comparing empathy requires a sophisticated appreciation of these contextual factors if we are to understand it.
METHODS
This research project conducted a systematic review using five electronic databases to search for literature focusing on empathy in medical, nursing, and other allied health education training. In each article, the definition of empathy, methods used, perspectives and themes that emerged, along with major findings were extracted.
RESULTS
The results yielded 27 articles with 16 studying a population of medical students, and 8 studying nursing students. Only 13 of the 27 retrieves defined empathy. Cross-sectional studies lead the most common methods used by these studies. This revealed that though empathy is uniquely defined, it is articulated in similar ways regardless of the health care professions. The themes that emerged among different populations also did not vary greatly.
CONCLUSIONS
The medical student curriculum is more accustomed to integrating “empathy” via teachings of ethics and classroom simulations, while nursing and other allied health training are more conducive for “empathy training” via clinical setting. This will help us understand how health professionals can maintain and develop empathetic behaviors throughout their education and practice. Further research is required to demonstrate the best way to allow students to further cultivate empathetic behaviors throughout training.
Prostate Cancer Screening in Hospitalized Patients: Results from The Nationwide Inpatient Sample
Thu Nguyen, B.S.1, Patrick Karabon, M.S.1
1Oakland University William Beaumont School of Medicine, Rochester, Michigan
INTRODUCTION
Prostate cancer (PCa) is the most common non-skin cancer in American men. Our study aims to determine the nationwide prevalence, and predictors of inpatient PCa screening encounters in patients with average risk of PCa using the National Inpatient Sample database.
METHODS
Data from 2006 to 2014 was used to evaluate PCa screening among hospitalized male patients between the ages of 45 and 69 at average risk for PCa in the United States. The outcome was whether a patient had an encounter for prostate cancer as noted on their discharge record. Variables analyzed included demographic factors, hospital characteristics, and other concomitant diagnoses for prostate or male urinary problems.
RESULTS
The prevalence of inpatient PCa screening was 2.57 per 100,000 hospital discharges. The following were significant factors associated with greater odds of inpatient PCa screening: Medicare (P = 0.0016), Uninsured patients (P = 0.0371), rural (P = < 0.0001) or urban nonteaching hospitals (P = < 0.0001), receiving care in the Midwest (P = < 0.0001), a diagnosis for urinary tract infections (P = 0.0367), GU symptoms (P < 0.0001), hyperplasia of prostate (P = 0.0006), or other male genital disorder (P < 0.0001).
CONCLUSIONS
We have concluded that prostate specific antigen screening should include shared decision making between physicians and patients. There exist disparities in Uninsured and rural patients for screening. Screening tools such as PSA are minimally invasive modalities in the inpatient setting that can help screen individuals at increased risk for the development of prostate cancer, allowing for early detection, prevention, improved rates of cure and ultimately, decreased rates of mortality.
