Clinical Chart Review Research Articles

P-1385. Improving Sexually Transmitted Infections Screening at a Quaternary Care Hospital Ryan White HIV Program: A Quality Improvement Initiative

Abstract Background Screening for sexually transmitted infections (STIs) such as gonorrhea, chlamydia, syphilis and hepatitis C (among men who have sex with men) are part of core performance measures and a reporting requirement for every Ryan White HIV/AIDS program. Dartmouth Health's Ryan White HIV program screening rates for these STIs as of August 2022 were ranging from 38% to 44%. The aim of this quality improvement initiative was to improve the screening rates to 80% by June 2024.Figure 1DH-Bedford Clinic Process Flow MapHighlighted in green are parts of the process where observation of clinical encounter and retrospective chart review were conducted. Methods We used principles of models for improvement recommended by Institute of Healthcare Improvement (IHI), clinical microsystem approach and other quality improvement (QI) techniques in this project. The project had two phases. The initial phase comprised of an audit of the program through observation of clinical encounters, semi-structured interviews and retrospective chart review, while the implementation phase involved three quality interventions: (1) Presentation of initial chart audit findings and educational session about STI guidelines, (2) Use of STI screening cards (main intervention), (3) Practice feedback and regular updates. Analysis and display of data were presented using statistical control process (SPC) charts to identify any significant changes in the system or processTable 1Summary of the Findings of Clinical Encounter Observation and Semi-structured Interviews to the Providers Results From August 2022 to March 2024, we increased our screening rates for gonorrhea from 38% to 66%, chlamydia from 38% to 62%, syphilis from 44% to 64% and hepatitis C (among MSM) from 39% to 82%. Several favorable special cause variations were observed in all outcome measures after the implementation phase.Figure 2p chart for Patients Being Screened for Gonorrhea This figure illustrates the proportion of patients seen in clinic each month who are screened for gonorrhea at least annually. The mean screening rates is 48%. The upper control limit is 62% while the lower control limit is 34%. There are two special cause variations (2 data points outside the UCL - red circle). Conclusion The use of quality improvement principles and methods to routinize sexual history taking can improve screening rates for sexually transmitted infections among adult patients in a Ryan White HIV program. Incorporation of the questions through the EMR is the next step for it to become more sustainable.Figure 3p chart for Patients Being Screened for Syphilis This figure illustrates the proportion of patients seen in clinic each month who are screened for syphilis at least annually. The mean screening rates is 53%. The upper control limit is 67% while the lower control limit is 39%. There are two special cause variations. First is upward trend (green circle) from 8/2023 to 2/2024. The second is a point outside the UCL in 2/2023. Disclosures All Authors: No reported disclosures

P-181. Limited oxygen therapy capacity against COVID-19 and the public health response in Zambia 2021

Abstract Background The coronavirus disease 2019 (COVID-19) pandemic revealed the limited capacity of intensive oxygen therapy in Zambia, especially in provincial medical facilities. Zambia COVID-19 treatment centers with newly installed HFNC and NPPV devices in 2021 Methods The Ministry of Health (MoH) conducted a clinical chart review of COVID-19 cases admitted to 11 COVID-19 treatment centers across Zambia to evaluate oxygen therapy practices and clinical outcomes. Then, MoH launched a multidisciplinary capacity-building project targeting four national COVID-19 treatment centers in the provinces aiming to deploy high-flow nasal cannula (HFNC) and non-invasive positive pressure ventilation (NPPV) coupled with training for healthcare workers. The chart review and the reporting of the project were approved by National Health Research Authority, Zambia (NHRA000006/15/10/2021, NHRA00003/25/04/2022), and Tohoku University, Japan (2021-1-983, 2022-1-287). Results Among 1,687 cases of severe/critical COVID-19, 84 cases (5%) were treated without oxygen, 687 (41%) were administered with oxygen only via nasal cannula, 759 (45%) via conventional masks, and 157 cases (9%) received intensive oxygen therapy, including HFNC, NPPV, and invasive mechanical ventilation. Compared to the capital Lusaka, COVID-19 cases treated in the provinces were more likely to receive oxygen administration only via nasal cannula and less likely to receive intensive oxygen therapy. To strengthen the intensive oxygen therapy capacity in the provinces, MoH installed HFNC and NPPV devices in four provincial medical facilities equipped with oxygen plants and piping in October and November 2021. MoH also provided a five-day training program for healthcare workers focusing on clinical knowledge/skills of HFNC and NPPV in COVID-19 and infection prevention and control in aerosol-generating procedures. Conclusion Our study identified the inequity of intensive oxygen therapy capacity between Lusaka and the provinces in Zambia and the need for further strengthening. A multidisciplinary approach coordinating national experts was practical in rapidly deploying HFNC and NPPV amid the COVID-19 pandemic. Disclosures All Authors: No reported disclosures

Identification of Key Predictors of Cryoglobulinemia Severity at Diagnosis: Threshold, Type, and Severity Score at Diagnosis.

Background: Cryoglobulinemia (CG) syndrome is a heterogeneous condition characterized by the presence of cryoglobulins in serum, often leading to vasculitis with protean clinical manifestations. Understanding the presentation of cryoglobulinemia-related symptoms based on cryoprecipitate levels, GC type, and severity at diagnosis is essential for effective management. Hence, this study aimed to provide a comprehensive analysis of patients with positive cryoglobulin detection to investigate these aspects. Methods: We conducted a retrospective review of clinical charts from patients with positive cryoglobulin detection at Colmar Hospital between May 2015 and April 2019. Results: Among 166 patients with positive cryoglobulins, the median cryoprecipitate value was 37 mg/L [IQR: 25-70], with 62% of patients below the 50 mg/L threshold. High cryoprecipitate levels were associated with C-virus hepatitis (p = 0.0007), increased fatigue (p = 0.001), fever (p = 0.0013), weight loss (p = 0.028), and musculoskeletal symptoms (p = 0.002). These patients also exhibited decreases in complement fractions (p-values 0.017 to 0.006). At the end of the one-year follow-up, they required frequent renal replacement therapy (p < 0.0001) and had a higher mortality rate (p = 0.02). Based on the CG type, patients with type I GC had splenomegaly (p = 0.039) and hemopathy (p = 0.001). According to severity at initial presentation, the severe patients had more purpura (p < 0.001), Raynaud's phenomenon (p = 0.039), and leukocytoclastic vasculitis on skin biopsy (p < 0.001), along with higher cryoprecipitate levels (p = 0.011). Multivariate analysis identified purpura (OR: 10.25), hematological malignancy (OR: 7.06), Raynaud's phenomenon (OR: 6.41), and cryoprecipitate levels (OR: 1.02) as significant markers of disease severity serving for the development of a severity score for clinical practice. Conclusions: This study identifies severity markers in patients with positive cryoprecipitate and proposes a score related to severity at diagnosis.

Microcystoid Macular Edema in Epiretinal Membrane: Not a Retrograde Maculopathy.

To investigate the incidence, clinical spectrum, and pathophysiology of microcystoid macular edema (MME) in 2 cohorts of patients with epiretinal membrane (ERM) and idiopathic full-thickness macular hole (FTMH). Single-center, retrospective, interventional, cohort study. Review of clinical charts, structural and en-face optical coherence tomography (OCT), and fluorescein angiography (FA) imaging of ERM and FTMH eyes that underwent surgery with pars plana vitrectomy and internal limiting membrane (ILM) peel, with a minimum follow-up of 6 months. Histopathology analysis of 3 specimens: 2 human retinas and 1 human ILM. One hundred seventy-two patients with ERM (123) and FTMH (49) were included in the study and followed up a mean of 9.1 ± 4.7 and of 8.2 ± 3.6 months, respectively. Preoperatively, MME was present in 27 of 123 eyes with ERM (21.9%), and in none of 49 eyes with FTMH (P < .001). MME was significantly associated with advanced ERM stages (P < .001). MME was typically located below continuous ERM-ILM adherence areas. FA in 46 ERM eyes showed capillary leakage in 36.4% of eyes without MME or cystoid macular edema (CME), in 39% of eyes with MME, and increased hyperfluorescence in CME. Postoperatively, new-onset MME appeared in 13 of 84 ERM eyes (15.5%) and in 1 FTMH eye (2%, P = .014). MME resolved in 7 of 40 ERM eyes with either preoperative or postoperative MME (17.9%) by 2.8 ± 1.5 months postsurgery. MME showed variable evolution postoperatively. The association between MME and postoperative best corrected visual acuity was significant only in univariate analysis (P = .037). Histopathology analysis showed anatomical continuity between Müller cells and ERM, suggesting a higher risk of iatrogenic damage in ERM eyes during peeling maneuvers. Postoperative MME was a frequent finding in ERM and a rare occurrence in FTMH, suggesting that ILM peeling alone may not be sufficient to cause MME. The morphology and clinical characteristics of ERM-related MME are unlikely related to neurodegenerative processes and rather attributable to Müller cell disruption and iatrogenic damage. The characteristics of MME and CME may overlap, blurring the differences between the 2 entities.

Rapidly growing knowledge of Mycobacterium smegmatis: A case series and review of antimicrobial susceptibility patterns

Mycobacterium smegmatis is a rapidly growing nontuberculous mycobacterium that is rarely isolated from clinical specimens and is frequently considered to be a contaminant. We conducted a retrospective review of mycobacterial cultures positive for M. smegmatis from 1998 to 2023 at our institution to evaluate the clinical significance of recovering this mycobacterium. Antimicrobial susceptibility patterns were also determined. Twenty-two M. smegmatis isolates were identified from 17 patients, 12 of whom met criteria for clinical chart review. M. smegmatis was deemed a cause of infection in 5/5 isolates from skin or soft tissue, 3/3 from bone, 1/1 from blood, and 0/3 from respiratory specimens. All cases thought to be significant were treated with at least 2 active agents for periods varying from 2 weeks up to 8 months. 18 isolates had antimicrobial susceptibility testing performed and all were susceptible to doxycycline, imipenem, linezolid, moxifloxacin, trimethoprim/sulfamethoxazole, and tobramycin while all isolates were resistant to clarithromycin. When recovered in culture, the presence of M. smegmatis should be correlated with clinical presentation as it may represent a true infection.

Prevalence of unplanned extubation in a tertiary care neonatal intensive care unit.

Orotracheal intubation and mechanical ventilation (MV) have become routine practices in intensive care units. Unplanned extubation (UE) is one of the most important complications, particularly in premature infants and critically ill newborns. The objective of this study was to determine the prevalence of UE in a tertiary care neonatal intensive care unit (NICU). In this analytical cross-sectional retrospective study, all data, including perinatal data, indications for ventilatory support, days of MV at the time of UE, work shift, month of the event, reintubation, and postextubation complications, were obtained from the manual review of clinical charts. In total, 151 neonates, who received invasive MV, were included in this study. The prevalence of UE was 2.0/100 days of ventilation. The most affected were premature infants, with a gestational age of ≤ 32 weeks (54.7%) and a birth weight of ≤ 1500 g. The main cause for UE was deficient fixation of the endotracheal tube (ETT) (27.7%). Most UE events occurred during night shifts (48.1%). Reintubation was required in 83.3% of newborns. Immediate complications developed in 96.3% of the UE events, including desaturation (57.7%) and bradycardia (36.5%). The prevalence of UE was high, particularly in premature infants, with a high rate of reintubation and immediate complications. Standardized protocols for ETT care must be implemented to reduce these events.

A-241 Verification and Validation Study for a Host Response Assay Based on TRAIL, IP-10, and CRP: Application to Pediatric Emergency Department

Abstract Background Acute febrile illnesses (AFIs) represent a significant disease burden globally; however, the lack of reliable, rapid point-of-care testing makes diagnosis difficult. A simple tool for distinguishing bacterial versus non-bacterial infections would radically improve patient management and reduce indiscriminate antibiotic use. MeMed BV® (MMBV) is a blood test to reliably differentiate between viral and bacterial infections. Three host proteins (TRAIL, IP-10, and CRP) are measured quantitatively and a combinational score (0-100) is computed that indicates the likelihood of bacterial infection (or co-infection) versus viral infection (BV Score). MMBV time to result is 15 min on the MeMed Key® analyzer that conducts chemiluminescence-based immunoassays to measure protein concentrations and calculates the test score, using 100 µL of serum. Methods The verification study design was a comparison of known results to on-site testing for contrived samples. The validation study design was a prospective, single-center, observational, blinded diagnostic accuracy study conducted between 8/3/22 to 5/30/23. The population was pediatric patients with suspected serious infections presenting to Geisinger’s quaternary care Pediatric Emergency Dept. in Danville, PA. The comparator was a clinical chart review and adjudication of over 60 clinical and laboratory variables related to systemic infections. The data review team consisted of one pediatric emergency department physician, one pediatric infectious disease physician, and one microbiology laboratory director. If clinical decisions differed among the three, the team adjudicated with a full chart review including knowledge of the eventual outcome, which served as the final reference standard result. The study aims were to assess the analytical and clinical accuracy of MMBV testing and to assess the test’s potential impact to further inform the provider’s decision to admit or discharge the patient. Results Of 60 contrived samples there was three failed tests (5% failure rate) and 2 discordant results. Analytical accuracy was 55/57 (96.5%) for 28 bacterial, 26 viral, and 1 equivocal control. Of 50 pediatric patients recruited, 8 were excluded because they were trauma patients. Among 42 patients tested, 22/42 (52.4%) were interpreted as bacterial or co-infection, 16/42 (38.1%) as viral or non-infectious, and 4/42 (9.5%) as equivocal. For 37/38 (97.3%) of patients with bacterial/viral MMBV results, the result was concordant with expert adjudication. Of the 42 patients for which potential utility was assessed, all results were in alignment with the clinical decisions made for admission or discharge. Conclusions MMBV performed accurately in pediatric ED patients with symptoms of severe infection, both clinically and analytically. A larger sample size will be tested to assess the utility in ED admission and discharge decisions and antimicrobial stewardship. The prevalence of viral and bacterial or coinfection cases within the sample population is representative of the pediatric population samples.

“Implant-associated infection after hip fracture surgery in elderly patients: Risk factors and mortality”

AimsOur study aimed to evaluate the prognosis of implant-related infection following hip fracture and pre-surgical, surgical and microbiological factors modifying outcomes. MethodsWe retrospectively included patients aged 65 and older treated for infection following surgically managed hip fractures between 2012 and 2022. Periprosthetic and high-energy fractures were excluded. Data was obtained by clinical chart review, including: age, gender, Barthel index, Charlson Score, National Hip Fracture Database (NHFD) Mobility Score, surgery and infection type, causative pathogens, antimicrobial resistance, anti-biofilm antibiotic treatment and clinical situation at one-year follow up. The outcomes evaluated were failure/resolution of infection and one-year mortality. ResultsWe included 80 patients (67 women, 83.8 %) aged a median of 85 years (interquartile range: 78 – 88 years). Treatment failed in 38 (47.5 %) patients, and the one-year mortality was 37.5 %. Patients dying within 12 months after treatment were more likely to suffer acute vs. chronic infections (OR = 3.29 [95 %CI: 1.20–9.04]), be older and have more comorbidity, but baseline function and ambulation were not predictive. Treatment failure was higher among patients receiving non-antibiofilm controlling surgery, specifically surgical lavage (OR = 3.79 [95 %CI: 1.38–10,37]), as well as in older, more dependent patients. Receiving anti-biofilm antibiotics for more than 2 weeks was associated with less treatment failure (OR:0.32; [95 %CI: 0.13–0.80]) and 12-month mortality (OR:0.22 [95 %CI: 0.08–0.60]). ConclusionsAntibiofilm-controlling surgery and antibiotics improve treatment success following implant-related infection after hip fracture and should be considered regardless of fracture type. Acute infections are a “second hit” for frail patients recovering from initial surgery and are associated with increased mortality.

Predictors of left atrial appendage emptying velocity: Derivation and validation of CHIRP3(M-1) score

BackgroundAtrial fibrillation (AF) leads to impaired left atrial appendage contractility, increasing the risk of thromboembolic stroke. The left atrial appendage emptying velocity (LAAev) measured on transesophageal echocardiogram (TEE) is a marker of increased thromboembolic risk. ObjectivesThe purpose of this study was to evaluate predictors of reduced LAAev for identifying individuals at increased risk for cardioembolic stroke. MethodsThis was a single-center retrospective review of TEEs and clinical charts. Predictors of LAAev <30 cm/s were identified using logistic regression. A risk prediction model was created using stepwise selection in a derivation set (n = 695) and separately tested in a validated set (n = 300). ResultsWe included TEEs on 995 patients (age 71.3±12.7 years; female 38.1%; history of AF 82.1%; in AF at evaluation 27.7%; CHA2DS2-VASc score 4.1 ± 1.9; LAAev 41.6 ± 21.0 cm/s). Significant multivariable predictors of LAAev <30 cm/s in derivation set were used to create the CHIRP3M-1 score containing 8 variables: Coronary artery disease (1), congestive Heart failure (1), Increased left atrial volume index ≥42 mL/m2 (1), current Rhythm AF (1), Paroxysmal AF (2), Persistent AF (3), longstanding Persistent/permanent AF (4), and greater than moderate Mitral regurgitation (-1). In the validation set, as compared to intermediate scores (3–4), those with low scores (≤2) and high scores (≥5) had odds ratios for LAAev <30 cm/s of 0.41 (0.21, 0.78, P = .007) and 2.58 (95% confidence interval 1.45–4.61, P = .001), respectively. ConclusionWe developed and validated a novel risk stratification system to predict reduced LAAev using clinical and echocardiographic variables. This may help refine the stratification of cardioembolic stroke risk.

Dimensional changes in free gingival grafts at implant sites in the reconstructed mandible: a retrospective study.

To evaluate the dimensional changes in free gingival grafts (FGG) at implant sites in mandibular reconstruction patients. Patients who received FGG 4 months after implant placement in the reconstructed mandible with no keratinized mucosa (KM) present were invited for re-examination after 36.7 ± 16.8 months (3.06 ± 1.4 years). Immediately after graft extraction (T0), graft width (GW), graft length (GL), graft thickness (GT), graft dimension (GD), and vertical bone height were documented. Re-examination (T1) included clinical examinations (GW, GL, GD, peri-implant probing depths, and modified Sulcus Bleeding Index), radiographic examination (marginal bone level), and medical chart review. Twenty patients and 62 implants (47 in fibula flaps and 15 in iliac flaps) were included. A significant decrease in GW (51.8%), GL (19.2%), and GD (60.2%), were found between T0 and T1 (p < .001). The univariate analysis showed that GW change was not significantly associated with reconstruction technique, baseline GL, baseline GT, baseline GD, implant location, or type of prosthesis. Implant survival rate of 100% was observed at follow-up. Within the limitations of the study, free gingival grafts at implant sites in the reconstructed mandible undergo dimensional change that result in a reduction of approximately 60% of the original graft dimension. Graft width decreased over 50%. FGG is the standard of care intervention for increasing the amount of KM around implants. This study was the first to evaluate the dimensional change in FGG at implant sites in mandibular reconstruction patients after a medium-term follow-up. Clinical trial registration is not applicable as this study comprehends a retrospective analysis.

Anifrolumab treatment improves patient-reported quality of life and decreases disease activity and corticosteroid use in patients with systemic lupus erythematosus: A qualitative study in Denmark.

Anifrolumab is a new therapeutic approach for individuals with systemic lupus erythematosus (SLE) directed at blocking the type 1 interferon pathway. Despite the expanding body of literature on Anifrolumab, an essential aspect remains absent: the subjective patient experience of treatment effects and implications on patients' health-related quality of life (HRQoL). The present study aimed to fill this void by elucidating the nuanced perspectives of SLE patients receiving Anifrolumab treatment by conducting qualitative in-depth interviews (IDIs). SLE patients at Aarhus University Hospital who had received at least three infusions of Anifrolumab were approached for inclusion in the study, which comprised two main elements: (1) qualitative IDIs and (2) collection of patient data from electronic medical records (EMRs). The IDIs were semi-structured and based on a discussion guide that included open-ended and close-ended questions. Verbatim transcripts were coded and analysed using qualitative software to understand concepts important to patients and to understand patients' own experiences before and after Anifrolumab therapy. A clinical chart review was conducted using EMR data at baseline, 3months, and 6months after Anifrolumab initiation. IDIs were completed with 14 patients, and EMR data was collected from 16 patients (treatment days range: 62-474). Of the 23 symptoms spontaneously reported by patients prior to Anifrolumab treatment, fatigue, joint pain, sun sensitivity, joint stiffness, skin rashes, and hair loss were the most common. Most symptoms improved, and none worsened during treatment. Patients reported significant impacts of disease on daily life before treatment: day-to-day activities, social life, emotional aspects, physical activity, concentration/memory, work/employment, and family/romantic relationships. Patients reported improvements in all aspects after treatment but were still impacted. From the EMR data, we observed a fall in disease activity after treatment initiation with a concomitant reduction in the use of corticosteroids. This study provides valuable insights into the subjective experiences of SLE patients treated with Anifrolumab, and the findings collectively contribute to a comprehensive understanding of the treatment's efficacy from the patients' perspective and its tangible effects on both subjective and objective parameters in SLE patients.

Assessing the effectiveness of “BETTER Women”, a community-based, primary care-linked peer health coaching programme for chronic disease prevention: protocol for a pragmatic, wait-list controlled, type 1 hybrid effectiveness-implementation trial

IntroductionThe Building on Existing Tools to Improve Cancer and Chronic Disease Prevention and Screening in Primary Care (BETTER) programme trains allied health professionals working in primary care settings to develop...

Bi-institutional analysis of microbiological spectrum and therapeutic management of parotid abscesses

BackgroundA parotid abscess (PA) is a complication of an acute bacterial parotitis with a potentially life-threatening course. To date, data on the diagnosis and therapy of PA is sparse and mostly consists of case reports or case series. Therefore, this study aimed at comprehensively analyzing the microbiological spectrum and the therapeutic management in a bi-institutional setting.MethodsA retrospective clinical chart review was performed to identify all patients surgically treated for PA at two tertiary care centers in Germany. Data on demographics, clinical management and microbiological data including species identification, pathogenicity, type of antibiotic therapy, adjustment of antibiotics, antibiotic sensitivity testing, and smear test results were extracted. Intervention-related variables and etiology were analyzed for their statistical association with outcome variables.ResultsOverall, 85 patients were included. Most patients (92.9%) underwent surgical incision. Around half of the patients (45.9%) were treated under local anesthesia. No facial nerve palsy was observed. The most frequently detected pathogens were Streptococci (n = 23), followed by Staphylococcus aureus (n = 6) including one case of methicillin-resistant Staphylococcus aureus. Most patients (68.2%) received an aminopenicillin ± beta-lactamase inhibitor as empiric antibiotic therapy. In 6 cases the antibiotic therapy was modified after receiving the antibiogram. Four patients (5.2%) presented with recurrent PA. Etiology was idiopathic (42.4%), followed by tumorous (12.9%), obstructive, and immunosuppressive (each 11.8%). Patients with a dental focus (p = 0.007) had a longer duration of hospitalization.ConclusionThe results show that the surgical therapy of PA under local anesthesia is safe. A dental examination should routinely be performed to rule out a dental focus. Obtaining a microbiological specimen in order to modify antibiotic therapy if necessary and a histopathological specimen to rule out a tumorous etiology is obligate.

Revealing inconsistencies between Epworth scores and apnea-hypopnea index when evaluating obstructive sleep apnea severity: a clinical retrospective chart review.

A common practice in clinical settings is the use of the Epworth Sleepiness Scale (ESS) and apnea-hypopnea index (AHI) to demonstrate the severity of obstructive sleep apnea (OSA). However, several instances were noted where there were discrepancies in the reported severity between Epworth scores and AHI in our patient sample, prompting an investigation into whether OSA severity as demonstrated by AHI or predicted by ESS quantification of sleepiness is primarily responsible for inconsistencies. Discrepancies were examined between Epworth scores and AHI by categorizing patients into two categories of inconsistency: individuals with either ESS < 10 and AHI ≥ 15 events/h or ESS ≥ 10 and AHI < 15 events/h. The potential influence of sex on these categories was addressed by assessing whether a significant difference was present between mean Epworth scores and AHI values for men and women in the sample. We investigated BMI both by itself as its own respective variable and with respect to the sex of the individuals, along with a consideration into the role of anxiety. Furthermore, we tested anxiety with respect to sex. In the first category of inconsistency the average ESS of 5.27 ± 0.33 suggests a normal level of daytime sleepiness. However, this contrasts with the average AHI of 32.26 ± 1.82 events/h which is indicative of severe OSA. In the second category the average ESS of 14.29 ± 0.47 suggests severe daytime sleepiness, contradicting the average AHI of 9.16 ± 0.44 events/h which only indicates mild OSA. Sex, BMI (both as a variable by itself and with respect to sex), and anxiety (both as a variable by itself and with respect to sex) contributed to observed inconsistencies. The findings of our study substantiate our hypothesis that Epworth scores should be de-emphasized in the assessment of OSA and a greater importance should be placed on measures like AHI. While Epworth scores offer insights into patients' daytime sleepiness levels and the perceived severity of their OSA, the inconsistencies highlighted in our results when compared to AHI-based OSA severity underscore their potential inaccuracy. Caution is advised when utilizing Epworth scores for evaluating OSA severity in clinical settings.

Use of computer-assisted self-interview in detection of and referral for depression among adolescents living with HIV at an urban HIV treatment clinic in Uganda: a quasi-experimental study

BackgroundDepression is common among adolescents living with HIV (ALHIV) and impacts their quality of life. However, it is not routinely detected and treated due to a lack of screening tools,...

Implementing a natural language processing pipeline to expedite clinical chart review to assess germline genetic testing in veterans with breast cancer.

e13640 Background: The Making Advances in Mammography and Medical Options for Veterans Act of 2022 mandated that the Department of Veterans Affairs (VA) perform a study of germline genetic testing in Veterans with breast cancer (BC). Assessing whether a Veteran with BC was offered germline genetic testing based on electronic health records requires time-consuming review of unstructured clinical notes by clinical experts who are also extremely busy providing care. To expedite this review process, we implemented a productionized natural language processing (NLP) pipeline to identify relevant concepts and highlight them within a custom-made chart review interface. Methods: Clinical subject-matter experts (SMEs) were convened to identify keywords suggestive of genetic testing. These were categorized into 3 sets: gene-related concepts, VA-specific services, and exams and tests. We then used entity recognition and entity linking to link these concepts to a knowledge base. These were enriched with entities similarly extracted from an open-source reference clinical text on germline genetic testing in BC. The resulting concepts were visualized as a word cloud and reviewed by the SMEs. We implemented a Power BI and Power Apps interface to display clinical notes from 200 randomly selected Veterans, with relevant concepts highlighted. Each Veteran’s notes were reviewed independently by 2 of the 5 SMEs, who assessed whether the Veteran was offered testing, referred to a genetic counseling service, or otherwise had evidence of testing. Disagreements were adjudicated by the whole team. Results: 11622 notes (mean 58/Veteran, 3028 characters/note) were reviewed. Table shows the most prevalent entities among genes, mutations, and exam names. Of the 5 reviewers, 3 annotated 80 patients each in a single session, lasting less than 50 minutes on average. Reviewers had an inter-annotator agreement of 85.5%. Mention of target concepts was associated with germline genetic testing with an accuracy of 80.4%, precision of 76.4%, recall of 99.1%, and F1 of 86.3%. Germline genetic testing rates are detailed in a separate submission. Conclusions: Modern informatics tools such as productionized NLP, the Microsoft Power Platform, and high-performance computing environments allow for rapid implementation of health information technology tools that can enable clinicians to gain critical insights quickly and efficiently from electronic health record data. [Table: see text]

#1595 Use of machine learning directed interventions to reduce fluid related hospital admissions in hemodialysis patients

Abstract Background and Aims Patients with end-stage kidney disease (ESKD) are hospitalized nearly two times per year on average; volume overload, infections and other cardiovascular complications are amongst the most common causes of hospitalization. We developed a machine learning (ML) model to predict haemodialysis (HD) patients at imminent risk of hospitalization due to fluid overload within the next 7 days. Patients identified as high risk by the model were assessed, intervened, and triaged on by group of telephonic nurses. The goal of this analysis is to evaluate the impact of the ML model-directed interventions on fluid related hospital admissions and all cause hospital admissions. Method We used data from a group of ESKD patients on HD modality who were identified as high risk by the ML model between baseline months April 2022 and June 2023. These high-risk patients were eligible to receive an assessment by telephonic nurses. The follow-up months were defined as the month immediately following the baseline month. Intervention group was identified as the high-risk patients where a clinical chart review was performed with needs identified, clinical chart review was performed and patients were managed by the clinic, some contact was made over phone/mail by the telephonic nurse, or some clinical recommendation was made. High-risk patients without any intervention or activity documented by the telephonic nurse served as the control population. We used linear regression analysis with repeated measures to account for within patient variability to produce estimated outcome rates with 95% confidence intervals adjusting for baseline factors such as average ML model risk score, average albumin, gender, race, age, vintage, comorbidities like chronic heart failure, diabetes, hypertension, ischemic heart disease, and primary insurance. Results There were 15,878 patients (36,932 patient months) in the group that received intervention; there were 51,800 patients (398,594 patient months) in the group that did not receive intervention. The pooled analysis (Table 1) showed a 38% reduction in all cause hospital admission rate between the baseline and follow-up months for the group that received intervention and there was a 34% reduction in all cause hospital admission rate for the group that did not receive any intervention. There was a 36% reduction in the fluid related hospital admission rate between the baseline and follow-up months for the group that received intervention and there was a 32% reduction in the fluid related hospital admission for the group that did not receive any intervention. Conclusion There is a larger magnitude reduction in fluid related hospital admissions and all cause hospital admissions between the baseline and follow-up months for the high-risk group as identified by the ML model that received intervention from the nurses compared to those that did not receive any intervention. Further studies are necessary to prove this association.

Otolaryngology outcomes of infants with conductive hearing loss identified through universal newborn hearing screening

ObjectiveInfants and children diagnosed with a conductive hearing loss (CHL) are often referred for otolaryngology assessment. Although this is also a regular occurrence for infants diagnosed with a CHL through Universal Newborn Hearing Screening (UNHS), less is known about these infants and their outcomes. Using a cohort of infants diagnosed with CHL through UNHS and referred to otolaryngology, this study aimed to investigate the relationship between specific demographic or clinical characteristics and 1) triage category 2) middle ear diagnosis and intervention and, 3) service-related factors at otolaryngology. MethodsRetrospective analysis through clinical chart review was performed on all infants born between January 2014 and December 2017 who referred on UNHS, diagnosed with a CHL and referred to the Queensland Children's Hospital. Descriptive analysis and Chi squared analysis was conducted on data from 95 records. ResultsAnalysis between all infants referred from UNHS and those who referred, diagnosed with CHL and then referred to otolaryngology suggest that bilateral referrals/medical exclusion, preterm and infants with ≥1 risk factors are more readily associated with referral to otolaryngology for CHL. Nearly all (92.86 %) infants who were referred to otolaryngology had a primary diagnosis of OM and most infants (89.66 %) received grommets as an intervention. The average age of first appointment at otolaryngology was 427 days, the average age of intervention was 579 days and the average occasions of service at otolaryngology was 6.72. ConclusionThis paper provides a snapshot into the journey and outcomes of infants referred from UNHS, diagnosed with CHL, and referred to otolaryngology. Further investigation in both general and UNHS populations is needed to better understand and apply these findings.

Abstract PO5-02-04: Irish National Analysis of the Clinical and Economic impact of 21-gene Oncotype DX® testing in Early-Stage, 1-3 lymph node positive, Hormone Receptor positive (HR+), HER2-Negative (HER2-), Breast Cancer (BC)

Abstract PO5-02-04: Irish National Analysis of the Clinical and Economic impact of 21-gene Oncotype DX® testing in Early-Stage, 1-3 lymph node positive, Hormone Receptor positive (HR+), HER2-Negative (HER2-), Breast Cancer (BC)

Next generation of electronic medical record search engines to support chart reviews: A systematic user study and future research direction

ObjectiveLittle research has been done on the user-centered document ranking approach, especially in a crowdsourcing chart review environment. As the starting point of designing and implementing the next generation of Electronic Medical Record (EMR) search engines, a systematic user study is needed to better understand the users' needs, challenges, and future research directions of EMR search engines. Materials and methodsOne primary observation during the user study is the need for a ranking method to better support the so-called "early stopping" reviewing strategy (i.e., reviewing only a subset of EMRs of one patient to make the final decision) during the clinical chart reviews. The authors proposed two novel user-centered ranking metrics: "critical documents" and "negative guarantee ratio," to better measure the power of a ranking method in supporting the “early stopping” requirements during clinical chart reviews. ResultsThe evaluation results show that i) traditional information retrieval metrics, such as the precision-at-K, have limitations in guiding the design and development of EMR search engines to better support clinical chart reviews; ii) there is not a global optimal ranking method that fits the needs of different chart reviews and different users; iii) a learning-to-rank approach cannot guarantee a stable and optimal ranking for different chart reviews and different users; and iv) A user-centered ranking metric, such as the negative guarantee ratio (NGR) metric is able to measure the “early-stopping” performance of ranking methods. ConclusionsUser-centered ranking metrics can better measure the power of ranking methods in supporting clinical chart reviews. Future research should explore more user-centered ranking metrics and evaluate their impact on real-world EMR search engines.