Generative Artificial Intelligence to Transform Inpatient Discharge Summaries to Patient-Friendly Language and Format.

Importance: By law, patients have immediate access to discharge notes in their medical records. Technical language and abbreviations make notes difficult to read and understand for a typical patient. Large language models (LLMs [eg, GPT-4]) have the potential to transform these notes into patient-friendly language and format. Objective: To determine whether an LLM can transform discharge summaries into a format that is more readable and understandable. Design, Setting, and Participants: This cross-sectional study evaluated a sample of the discharge summaries of adult patients discharged from the General Internal Medicine service at NYU (New York University) Langone Health from June 1 to 30, 2023. Patients discharged as deceased were excluded. All discharge summaries were processed by the LLM between July 26 and August 5, 2023. Interventions: A secure Health Insurance Portability and Accountability Act-compliant platform, Microsoft Azure OpenAI, was used to transform these discharge summaries into a patient-friendly format between July 26 and August 5, 2023. Main Outcomes and Measures: Outcomes included readability as measured by Flesch-Kincaid Grade Level and understandability using Patient Education Materials Assessment Tool (PEMAT) scores. Readability and understandability of the original discharge summaries were compared with the transformed, patient-friendly discharge summaries created through the LLM. As balancing metrics, accuracy and completeness of the patient-friendly version were measured. Results: Discharge summaries of 50 patients (31 female [62.0%] and 19 male [38.0%]) were included. The median patient age was 65.5 (IQR, 59.0-77.5) years. Mean (SD) Flesch-Kincaid Grade Level was significantly lower in the patient-friendly discharge summaries (6.2 [0.5] vs 11.0 [1.5]; P < .001). PEMAT understandability scores were significantly higher for patient-friendly discharge summaries (81% vs 13%; P < .001). Two physicians reviewed each patient-friendly discharge summary for accuracy on a 6-point scale, with 54 of 100 reviews (54.0%) giving the best possible rating of 6. Summaries were rated entirely complete in 56 reviews (56.0%). Eighteen reviews noted safety concerns, mostly involving omissions, but also several inaccurate statements (termed hallucinations). Conclusions and Relevance: The findings of this cross-sectional study of 50 discharge summaries suggest that LLMs can be used to translate discharge summaries into patient-friendly language and formats that are significantly more readable and understandable than discharge summaries as they appear in electronic health records. However, implementation will require improvements in accuracy, completeness, and safety. Given the safety concerns, initial implementation will require physician review.

Evaluating Whether an Inpatient Initiative to Time Lab Draws in the Evening Reduces Anemia.

BACKGROUND: Hospital acquired anemia is common during admission and can result in increased transfusion and length of stay. Recumbent posture is known to lead to lower hemoglobin measurements. We tested to see if an initiative promoting evening lab draws would lead to higher hemoglobin measurements due to more time in upright posture during the day and evening. METHODS: We included patients hospitalized on 2 medical units, beginning March 26, 2020 and discharged prior to January 25, 2021. On one of the units, we implemented an initiative to have routine laboratory draws in the evening rather than the morning starting on August 26, 2020. There were 1217 patients on the control unit and 1265 on the intervention unit during the entire study period. First we used a linear mixed-effects model to see if timing of blood draw was associated with hemoglobin level in the pre-intervention period. We then compared levels of hemoglobin before and after the intervention using a difference-in-difference analysis. RESULTS: In the pre-intervention period, evening blood draws were associated with higher hemoglobin compared to morning (0.28; 95% CI, 0.22-0.35). Evening blood draws increased with the intervention (10.3% vs 47.9%, P > 0.001). However, the intervention floor was not associated with hemoglobin levels in difference-in-difference analysis (coefficient of -0.15; 95% CI, -0.51-0.21). CONCLUSIONS: While evening blood draws were associated with higher hemoglobin levels, an intervention that successfully changed timing of routine labs to the evening did not lead to an increase in hemoglobin levels.

Standardized Preoperative Pathways Determining Preoperative Echocardiogram Usage Continue to Improve Hip Fracture Quality.

Introduction: The purpose of this study was to assess the hospital quality measures and outcomes of operative hip fracture patients before and after implementation of an anesthesiology department protocol assigning decision for a preoperative transthoracic echocardiogram (TTE) to the hospitalist co-managing physician. Materials and Methods: Demographics, injury details, hospital quality measures, and outcomes were reviewed for a consecutive series of patients presenting to our institution with an operative hip fracture. In May of 2019, a new protocol assigning the responsibility to indicate a patient for preoperative TTE was mandated to the co-managing hospitalist at the institution. Patients were split into pre-protocol and post-protocol cohorts. Linear regression modeling and comparative analyses were conducted with a Bonferroni adjusted alpha as appropriate. Results: Between September 2015 and June 2021, 1002 patients presented to our institution and were diagnosed with a hip fracture. Patients in the post-protocol cohort were less likely to undergo a preoperative echocardiogram, experienced a shorter time (days) to surgery, shorter length of stay, an increase in amount of home discharges, and lower complication risks for urinary tract infection and acute blood loss anemia as compared to those in the pre-protocol cohort. There were no differences seen in inpatient or 30-day mortality. Multivariable linear regression demonstrated a patient's comorbidity profile (Charlson Comorbidity Index (CCI)) and their date of presentation (pre- or post-protocol), were both associated with (P<0.01) a patients' time to surgery. Conclusion: A standardized preoperative work flow protocol regarding which physician evaluates and determines which patients require a preoperative TTE allows for a streamlined perioperative course for hip fracture patients. This allows for a shortened time to surgery and length of stay with an increase in home discharges and was associated with a reduced risk of common index hospitalization complications including UTI and anemia.

Increasing Rates of Prone Positioning in Acute Care Patients with COVID-19.

BACKGROUND: Prone positioning improves mortality in patients intubated with acute respiratory distress syndrome and has been proposed as a treatment for nonintubated patients with COVID-19 outside the ICU. However, there are substantial patient and operational barriers to prone positioning on acute floors. The objective of this project was to increase the frequency of prone positioning among acute care patients with COVID-19. METHODS: The researchers conducted a retrospective analysis of all adult patients admitted to the acute care floors with COVID-19 respiratory failure. A run chart was used to quantify the frequency of prone positioning over time. For the subset of patients assisted by a dedicated physical therapy team, oxygen before and after positioning was compared. The initiative consisted of four separate interventions: (1) nursing, physical therapy, physician, and patient education; (2) optimization of supply management and operations; (3) an acute care prone positioning team; and (4) electronic health record optimization. RESULTS: From March 9, 2020, to August 26, 2020, 176/875 (20.1%) patients were placed in prone position. Among these, 43 (24.4%) were placed in the prone position by the physical therapy team. Only 2/94 (2.1%) eligible patients admitted in the first two weeks of the pandemic were ever documented in prone position. After launching the initiative, weekly frequency peaked at 13/28 (46.4%). Mean oxygen saturation was 91% prior to prone positioning vs. 95.2% after (p < 0.001) in those positioned by physical therapy. CONCLUSION: A multidisciplinary quality improvement initiative increased frequency of prone positioning by proactively addressing barriers in knowledge, equipment, training, and information technology.

Smoking and hepatocellular carcinoma mortality.

The association between cigarette smoking and mortality from hepatocellular carcinoma (HCC) is ambiguous. We analyzed the association between smoking and mortality in HCC patients seen at our center. We collected data retrospectively on patients diagnosed with HCC between 2002 and 2009. We estimated the association of smoking history with demographic, clinical and treatment factors. We then modeled these factors as predictors of mortality. Among smokers, we analyzed the effects of pack-year history and cessation times on survival. Two hundred and twenty-three out of 444 patients with HCC had a history of smoking. Smokers were more likely to be younger at diagnosis, to have α fetoprotein (AFP) values less than the median, and to have had surgery (p=0.04) compared to non-smokers. In a Cox model, younger age, lower AFP and Child's Class were all independently predictive of survival, but smoking was not. Smokers with over 20 pack-years did not have worse survival than lighter smokers, and cessation times also did not affect survival after controlling for age. We found a significant interaction between smoking and drinking. In our data, smoking was not independently associated with HCC survival in a multivariable model. Smoking was associated with favorable prognostic features which likely outweighed any independent effect of smoking.

Diabetes, body mass index, and outcomes in hepatocellular carcinoma patients undergoing liver transplantation.

UNLABELLED: For many cancers, features of the metabolic syndrome, such as diabetes and obesity, have been associated with both increased risk of cancer development and poor outcomes. METHODS: We examined a large retrospective cohort of 342 consecutive patients who underwent liver transplantation for hepatocellular carcinoma between January 1999 and July 2010 at our institution. We evaluated the relationship between diabetes, obesity, hepatocellular carcinoma (HCC) recurrence, and overall survival. RESULTS: We found that a body mass index (BMI) higher than 30 was an independent predictor of poor overall survival in a multivariable Cox model, approximately doubling the risk of death after transplantation. A BMI higher than 30 was also a predictor of recurrent HCC, although this was of borderline statistical significance (hazard ratio for recurrence, 1.9; 95% confidence interval, 0.9-4.1). We also found increased BMI to be an independent predictor of microvascular invasion within HCC tumors, lending a possible explanation to these results. Those with diabetes showed worsened overall survival compared with those without diabetes in univariate but not multivariable analysis, possibly related to longer wait times. CONCLUSIONS: Our findings suggest a relationship between higher BMI, tumor vascular invasion, increased recurrence, and worsened overall survival. These findings may help explain why those with high BMI have worse outcomes from their cancers. A better understanding of the role of obesity and diabetes in patients with cancer should help develop better predictors of outcome and improved treatment options for patients with HCC.