Judged 3-Minute Oral Presentation Session III

Background/Significance:

Lack of a universally accepted definition of strict vs standard intake and output (Ia&nO) has led to variability in documentation and orders across the pediatric hospital medicine (PHM) team. A serious safety event (SE) occurred when severe dehydration in an infant was not recognized, resulting in hypovolemic shock. Chart review showed minimal output documentation and failure to recognize I&O imbalance.

Purpose:

We aimed to decrease the I&O SE rate by 10% over 2 years by improving the integration of I&O into the clinical care of general pediatric patients at two campuses of Advocate Children’s Hospital.

Methods:

A multidisciplinary team of physicians, nurses, and patient care technicians (PCTs) developed a key driver diagram, with primary drivers identified to be measurement, documentation, reconciliation, and discussion of I&O. Outcome measures were time between, and percentage of I&O SE. Process measures included appropriate flowsheet documentation, discussion of I&O on family-centered (FCR) and night rounds (NR) and correct I&O orders. Balance measures included nurse/PCT overtime and staff perceptions. Interventions included revising the IIntroduction O policy to define strict vs standard I&O and escalation criteria, standardizing nursing I&O report on rounds, adding order set and note hard stops, and placing visual reminders. Run and control charts were analyzed using standard rules.

Results:

The I&O SE rate is now less volatile, and the median time between events remained 18 days. Campus A documentation of intake improved from 91% to 98% and output sustained at 93%. Campus B documentation intake improved from 35% to 55% and output improved from 33% to 49%. Discussion of I&O went from 76% to 94% on FCR and 70% to 93% on NR. Percentage of I&O orders placed correctly improved from 50% to 79% with special cause variation achieved and sustained for the last 6 months of the project. Staff surveys show no perceived increase in time spent on documentation/rounds and there has been no change in overtime.

Conclusion:

This 2-year project improved I&O ordering, documentation, and discussion, meeting our aim of better integrating I&O into care and addressing SEs. Because the outcome measures rely on event reports, rates may decrease due to process improvement or increase due to staff awareness prompting increased reporting. Opportunity remains for improvement in documentation at Campus B. These interventions are generalizable to other inpatient pediatric settings.

Background/Significance:

Postoperative delirium is a significant problem in older adults undergoing surgery. The risk varies widely by surgery type; for elective orthopedic procedures delirium may occur in >40% of older adults. Many factors influence the risk of delirium; evidence indicates age and pre-existing cognitive impairment are the strongest predictors of postop delirium. Our institution does not have a robust preop screen for cognitive impairment and postop delirium risk in older adults undergoing elective orthopedic procedures.

Purpose:

To preoperatively administer a brief cognitive screening tool, the AD8, to older adults undergoing elective joint replacement surgery to (1) assess feasibility of administering the AD8 during preop education sessions and (2) determine if patients with a higher score on this tool were more likely to develop postop delirium.

Methods:

Patients aged 65 or older undergoing Joint Academy preop education session via phone call between 4/2/25 – 6/5/25 completed the AD8 questionnaire during the session. Questions were read aloud by provider or Spanish interpreter. Patient answers were documented directly into REDCap. Descriptive statistics are used to summarize findings.

Results:

122 questionnaires were administered; 94.3% were completed in 5 minutes or less. Seven patients were excluded from analysis due to data entry error (n=4) or canceled surgery (n=3). Most questionnaires were completed within 30 days of surgery (90.4%). Patients had a median age of 72 yrs (IQR 69-76.5), were majority female (67.8%), and were majority White non-Hispanic (80.9%). Twenty percent of patients (n=23) had an AD8 score ≥2, indicating mild cognitive impairment prior to surgery. Only two patients had documentation of delirium during the post-op period. Of note, the median length of stay was 25.5 hours (IQR 8-29) and only 17.4% of patients had LOS ≥48 hours.

Conclusion:

The AD8 questionnaire was able to be administered quickly, and most patients had their education session within 30 days of surgery, making this an optimal time for preop cognitive screening. This patient group, however, had a very low rate of postop delirium and was typically discharged less than 48 hours after surgery. Given these patients are undergoing elective joint replacement surgery and they are discharged quickly, they have a lower risk of postop delirium and were not an ideal patient population to assess whether preop AD8 screening can predict postoperative delirium.

Background/Significance:

Artificial intelligence (AI) tools are rapidly being introduced into clinical and research workflows, including documentation support, data analysis, and decision augmentation. Successful integration of these tools will depend on user trust, perceived appropriateness, and ethical comfort. Misalignment between institutional implementation strategies and frontline clinician attitudes may influence adoption, safe use, and long-term impact on patient care and research integrity. However, little is known about how health system–affiliated clinicians, educators, and researchers perceive AI use across professional roles and training levels. To this end, we conducted a survey-based study to gather information on how AI tool use is perceived by our colleagues.

Purpose:

To collect data regarding attitudes towards AI use tool among individuals affiliated with research activities, clinical operations, or with residency and fellowship programs.

Methods:

A REDCap survey was sent to emails on the Advocate Health Midwest Scientific Day listserv as of 10/17/25 (n=2,238). Each email address received a unique link, sent up to 11 times between 11/21/25-2/17/26. Survey questions gathered demographics (job role, clinical area, age group), prior AI tool knowledge, and attitudes regarding AI tool use in medical education, research, and direct patient care.

Results:

A total of 363 participants answered at least one AI tool-related question on the survey; 8 did not answer all survey questions. Respondents included attending physicians (32%), residents and fellows (19%), scientists and research support staff (16%), GME faculty physicians (15%), and other clinical and research-affiliated individuals (18%). Most respondents had exposure to AI tools prior to the survey (89%). The majority of participants believed AI tools could be used in some capacity in medical education (88%), in research with disclosure of use (85%), and in direct patient care (94%). Individuals selecting a job role of resident, fellow, GME faculty, or attending were more favorable of the use of AI tools in medical education (94% vs. 73%), research (89% vs. 78%), and patient care (97% vs. 87%) compared to individuals selecting a job role of scientist or research support.

Conclusion:

Participants were familiar with AI tools and generally supportive of their use in various aspects of medical education, research, and direct patient care. Clinical-focused individuals overall were more likely to support AI tool use compared to research-focused individuals.

Background/Significance:

Patient no-show (NS) rates in residency-based primary care clinics are a common problem. Adverse effects include disruption in continuity of care, inefficiency, reduced productivity, delayed care, increased ED visits, and negative impact on resident learning. Studies examining approaches to reducing NS rates in the primary care setting have been conducted, but the best approach remains unclear and is likely highly dependent on unique clinic characteristics. At the Aurora Sinai Family Care Center (FCC) clinic, the overall NS rate was 27% in 2024. Currently, all FCC patients receive automated call reminders two days before a scheduled appointment. This project implemented an additional direct phone call reminder targeted to patients with historically high NS rates.

Purpose:

To evaluate the relationship between targeted direct visit reminder calls and subsequent appointment attendance outcomes.

Methods:

For each FCC clinic day over a 3-month period, direct reminder calls were made to 10 patients identified as having the highest historical NS rates two days prior to scheduled appointments. Call outcome (patient reached, left message, unable to reach) was documented. EMR was reviewed to determine appointment outcomes (show, NS, other). Descriptive and chi square statistics were used to calculate associations between call and appointment outcomes.

Results:

A total of 784 appointments with documented pre-visit targeted direct phone call outreach had a NS rate of 33.8% (265/784). Of the patients reached by phone, 55.9% (212/379) attended their appointment, compared with 44.5% (109/245) when a message was left and 36.9% (59/160) when unable to be reached. Patients unable to be reached had NS rates substantially higher than the overall average (53.1% vs 33.8%), whereas patients who were reached had NS rates below the overall average (22.7%). Differences in attendance counts and NS rates are due to the inclusion of “other” in appointment outcomes (e.g. cancellation, rescheduling). A chi-square test demonstrated a statistically significant association between call outcome and appointment outcome (χ2(4) = 50.9, p < 0.001).

Conclusion:

Targeted direct call reminders to patients with historically high NS rates improved visit attendance when patients were able to be reached. Further cost/benefit analysis of expanding this approach given personnel requirements is needed. Future areas for investigation may include targeting visit types with higher NS rates such as well-child visits.

Background/Significance:

Pulsed field ablation (PFA) for the management of atrial fibrillation is efficacious and safe. As adoption rapidly grows, there is interest in understanding how commercially available PFA platforms differ in operational characteristics and short-term clinical outcomes.

Purpose:

We compared procedural metrics and 30-day clinical outcomes between the first-time use of the Boston Scientific FARAPULSE pentaspline and Medtronic Affera Sphere-9 PFA systems.

Methods:

We analyzed the first cases who underwent PFA with no prior ablation using the pentaspline (n=35, 55.6%) or the Sphere-9 catheter (n=28, 44.4%) at our center. Continuous variables are presented as median (Q1, Q3) and categorical variables as n (%). A chi-square test or Fisher’s exact test was used for categorical variables and the Wilcoxon rank-sum test for continuous variables.

Results:

Baseline demographics and comorbidities were similar between both systems. There was no significant difference in total procedure time (pentaspline: 126.0 [102.2, 153.8] min, Sphere-9: 110.0 [90.5, 148.0] min, p=0.250). Procedures using the Sphere-9 catheter had shorter fluoroscopy duration (Sphere-9: 10.0 [6.6, 15.6] min, pentaspline: 18.6 [14.5, 25.2] min, p<0.001) and lower fluoroscopy dose (Sphere-9: 80.0 [35.5, 184.5] mGy, pentaspline: 214.0 [102.0, 374.0] mGy, p<0.001). Difference in left atrial (LA) dwell time was not significant but trended to shorter with the use of the Sphere-9 catheter (Sphere-9: 56 [15, 86] min, pentaspline: 75 [55, 89] min, p=0.053). Many cases included combined right-sided electrophysiology interventions. The Sphere-9 catheter had significantly more pulmonary vein isolation (PVI) plus cavo-tricuspid isthmus (CTI) ablations (Sphere-9: 14 [50.0 %] cases, pentaspline: 8 [22.9 %] cases, p=0.025) and less PVI plus any other lesion (Sphere- 9: 4 [14.3%] cases, pentaspline: 15 [42.9 %], p=0.014). Both catheters had similar frequency of PVI only cases. One Sphere-9 patient had a concomitant LA appendage occlusion procedure. No major adverse events occurred in either group. One patient (3.6%) who underwent ablation with the Sphere- 9 catheter had a minor groin hematoma that did not require intervention.

Conclusion:

With experienced operators, first-time use of the Sphere-9 catheter demonstrated markedly lower fluoroscopy time and dose compared with the pentaspline catheter, while total procedure time was similar. Both platforms demonstrated excellent acute safety.

Background/Significance:

Microvascular Fluorescence angiography (MFA) evaluates focal tissue perfusion and facilitates determination of ongoing or resolving ischemia. Our department utilizes MFA to validate or revise treatment plans for compromised surgical flaps in patients receiving hyperbaric oxygen therapy (HBOT). This study assessed whether the use of MFA impacted outcomes of patients with toe amputation compromised flaps who underwent HBOT guided by MFA compared to those who did not receive MFA.

Purpose:

To assess healing outcomes of MFA guided HBOT of compromised flaps of toe amputations.

Methods:

In this retrospective chart review study from 1/1/16 to 3/31/25, outcomes of healing, time to heal, additional amputations and readmission rates were compared for adult patients with compromised flaps after toe amputations. One cohort underwent MFA in conjunction with HBOT (n=129); the second cohort did not receive MFA (n=254). Analysis included Chi-square test, Fisher's test, and Wilcoxon rank sum test.

Results:

A total of 383 patients underwent a median of 17 (IQR 9-25) HBOT sessions: a median 15 (IQR 7-23) in the non- MFA group vs. 20 (IQR 14-28) with MFA (p <0.01). Over half of patients (56%) healed in both groups (non-MFA and MFA). Readmission rates were similar for non-MFA (42%) and MFA (36%) groups (p=0.64). Median time to heal from the 1st HBOT was 96 days (IQR 46, 161) in non-MFA group vs 75.5 days (IQR 42.5-148.5) in the MFA group (p=0.36). Median time to heal from last HBOT was 68 days (IQR 28.5-142.5) in non-MFA group vs 54.5 days (IQR 22.5-121.5) in the MFA group (p=0.18). Of the non-healed patients (n=169), a higher amputation was completed for 79% of non- MFA vs. 93% of the MFA group (p=0.001).

Conclusion:

Driven by objective data, patients receiving MFA underwent more HBOT sessions and demonstrated favorable trends towards shorter healing times, lower re-admission rates, and fewer subsequent procedures among non-healed groups. Even though the wounds in the MFA group showed a higher acuity and flap compromise, the overall healing rates between groups were similar. MFA fosters definitive early identification of a non-viable flap which substantiates the need for a more proximal amputation to ensure closure and healing. These findings suggest that MFA may assist in optimizing HBO utilization and limb salvage. Further prospective studies are warranted to clarify which patient populations derive the greatest benefit from MFA-guided HBOT.

Background/Significance:

Hyperbaric oxygen therapy (HBOT) requires substantial patient commitment, often 20–40+ outpatient sessions, and premature discontinuation along with missed treatments are associated with poorer wound healing and limb salvage outcomes. Psychological and social determinants of health (SDOH) including anxiety, claustrophobia, and social connectedness, may influence adherence. Multiplace chambers allow peer interaction and continuous staff support that may mitigate anxiety, whereas monoplace chambers may increase perceived isolation. Limited data exists comparing treatment completion between chamber types.

Purpose:

To compare treatment compliance between monoplace and multiplace HBOT chambers and explore associations between adherence metrics in an outpatient hyperbaric program.

Methods:

A retrospective chart review was conducted on adult outpatients receiving HBOT at 3 monoplace sites and 1 multiplace site from 09/01/21 to 09/30/25. Demographics, indications, prescribed and completed sessions, no-shows, aborted runs, and anxiolytic use were collected. Statistical analysis utilized Chi-square, Fisher’s exact, and Wilcoxon rank-sum tests.

Results:

The cohort included 172 patients (monoplace n=110, multiplace n=62). Median prescribed sessions were higher in the monoplace group (40 [IQR 30–40] vs 30 [10–30], p<0.001). Median completed sessions were also higher in monoplace when compared to multiplace chambers (33 [21–40] vs 20 [4–30], p<0.001). No-show rates were significantly higher in monoplace (74.6%) vs. multiplace (48.4%; p=0.001). Aborted runs were more frequent in monoplace (17.3%) vs. multiplace (8.1%, p=0.19). When controlling for variability by limiting diagnosis to chronic refractory osteomyelitis, diabetic foot ulcer Wagner stage > 3, delayed radiation injury, and compromised surgical flap, similar treatment completion patterns were observed.

Conclusion:

Across all treated diagnoses in an outpatient HBOT cohort, monoplace patients were prescribed and completed more overall sessions but had higher no-show rates and aborted treatments. These findings suggest compliance challenges may be more prominent in the monoplace setting. Persistent trends after controlling diagnosis support the belief that chamber environment and associated social interaction may influence adherence behaviors. Prospective studies incorporating standardized anxiety and SDOH screening are warranted to guide targeted interventions aimed at improving treatment completion and outcomes.