ChatGPT's Performance in Cardiac Arrest and Bradycardia Simulations Using the American Heart Association's Advanced Cardiovascular Life Support Guidelines: Exploratory Study.

BACKGROUND: ChatGPT is the most advanced large language model to date, with prior iterations having passed medical licensing examinations, providing clinical decision support, and improved diagnostics. Although limited, past studies of ChatGPT's performance found that artificial intelligence could pass the American Heart Association's advanced cardiovascular life support (ACLS) examinations with modifications. ChatGPT's accuracy has not been studied in more complex clinical scenarios. As heart disease and cardiac arrest remain leading causes of morbidity and mortality in the United States, finding technologies that help increase adherence to ACLS algorithms, which improves survival outcomes, is critical. OBJECTIVE: This study aims to examine the accuracy of ChatGPT in following ACLS guidelines for bradycardia and cardiac arrest. METHODS: We evaluated the accuracy of ChatGPT's responses to 2 simulations based on the 2020 American Heart Association ACLS guidelines with 3 primary outcomes of interest: the mean individual step accuracy, the accuracy score per simulation attempt, and the accuracy score for each algorithm. For each simulation step, ChatGPT was scored for correctness (1 point) or incorrectness (0 points). Each simulation was conducted 20 times. RESULTS: ChatGPT's median accuracy for each step was 85% (IQR 40%-100%) for cardiac arrest and 30% (IQR 13%-81%) for bradycardia. ChatGPT's median accuracy over 20 simulation attempts for cardiac arrest was 69% (IQR 67%-74%) and for bradycardia was 42% (IQR 33%-50%). We found that ChatGPT's outputs varied despite consistent input, the same actions were persistently missed, repetitive overemphasis hindered guidance, and erroneous medication information was presented. CONCLUSIONS: This study highlights the need for consistent and reliable guidance to prevent potential medical errors and optimize the application of ChatGPT to enhance its reliability and effectiveness in clinical practice.

Treatment satisfaction with Omnipod DASH in adults with type 1 diabetes: A non-blinded 1:1 randomized controlled trial.

OBJECTIVE: We compared treatment satisfaction with a tubeless insulin pump (Omnipod DASH® Insulin Management System) to usual care (multiple daily injections [MDI] or tubed insulin pump therapy [IPT]) in adults with type 1 diabetes using self-monitoring blood glucose (SMBG). RESEARCH DESIGN AND METHODS: Adults with type 1 diabetes on MDI (n = 40) or IPT (n = 25) from four diabetes centers in Australia were randomized in a 1:1 non-blinded manner to Omnipod DASH System (Omnipod group) or continue usual care (Usual Care group) for 12 weeks, followed by a further 12-week extension where all participants used the device. The primary outcome was treatment satisfaction assessed by change in Diabetes Technology Questionnaire 'current' (ΔDTQ-current) score at 12-weeks (study-end). Secondary outcomes included ΔDTQ-current following extension and other participant-reported outcomes (PROs) measuring quality of life, burden of disease treatment, glycemic and device-related outcomes at 12-weeks (study-end) and 24-weeks (end-extension). RESULTS: Treatment satisfaction improved more in Omnipod group vs. Usual Care group (ΔDTQ-current score of 16.4 [21.2] vs. 0.0 [12.8]; p < 0.001) at study-end. Significantly greater improvements in other PROs and HbA1c were also observed. Improvements in DTQ-current and other PROs comparing study-end and end-extension were similar. While %TIR change from baseline did not differ at study-end (-2.0 [12.7] %), it was significantly greater at end-extension (5.6 [10.9] %; p = 0.016). CONCLUSIONS: Omnipod DASH System resulted in greater treatment satisfaction at 12 weeks in adults with type 1 diabetes using SMBG which was sustained after 24 weeks of device use without compromising sleep quality and fear of hypoglycemia. Improvements in glycemia were also observed.

"You can hide it if you want to, you can let it be seen if you want to": A qualitative study of the lived experiences of Australian adults with type 1 diabetes using the Omnipod DASH® system.

AIMS: Understanding the lived experience of using a tubeless insulin pump and how this differs compared to usual care (tubed insulin pump therapy (IPT) vs multiple daily injections (MDI)). METHODS: Interviews were conducted after 12-weeks of using the Omnipod DASH Insulin Management System (Insulet, Acton, MA) and analysed using thematic analysis. RESULTS: Fifty-eight adults (35 female; mean age 42;SD 13 years; 35 previous MDI) were interviewed. Most (84 %) wanted to continue using the device. Experiences fit two themes: 1. Taking back control of my diabetes: many previous MDI users perceived improved glycaemic control, explained by more "nuanced" control, with some reporting positive effects during exercise and sleep. Many previous MDI and IPT users endorsed positive experiences in concealing or disclosing their diabetes to others. However, some previous MDI users reported negative psychosocial experiences due to feeling continuously "attached" to their diabetes. 2. Barriers and facilitators of device acceptability: both MDI and IPT users cited wearability, alarms and the financial cost impacted their choice to continue device use. IPT users reported positive wearability experiences. CONCLUSIONS: The tubeless pump improved diabetes management perceptions for both MDI and tubed pump users. However, participants' prior glucose management affected perceptions of its advantages and disadvantages.

Proxy use of patient portals on behalf of children: Federally Qualified Health Centers as a case study.

Objective: This study examined the proxy use of patient portals for children in a large Federally Qualified Health Centers (FQHC) network in Texas. Methods: We used de-identified individual-level data of patients, 0-18 years, who had 1+ visits between December 2018 and November 2020. Logistic regression was used to examine patient-, clinic-, and geographic-level factors associated with portal usage by an assumed proxy (i.e. parent or guardian). Results: The proxy portal usage rate increased from 28% in the pre-pandemic months (November 2018-February 2020) to 34% in the pandemic months (March-Nov 2020). Compared to patients 0-5 years, patients aged 6 to 18 years had lower odds of portal usage (6-10 OR: 0.77, p < 0.001; 11-14 OR: 0.62, p < 0.001; 15-18 OR: 0.51, p < 0.001). Minoritized groups had significantly lower odds of portal usage when compared to their non-Hispanic White counterparts (non-Hispanic Black OR: 0.78, p < 0.001; Hispanic OR 0.63, p < 0.001; Asian OR: 0.69, p < 0.001). Having one chronic condition was associated with portal usage (OR: 1.57, p < 0.001); however, there were no significant differences in portal usage between those with none or multiple chronic conditions. Portal usage also varied by service lines, with obstetrics and gynecology (OR: 1.84, p < 0.001) and behavioral health (OR 1.82, p < 0.001) having the highest odds of usage when compared to pediatrics. Having a telemedicine visit was the strongest predictor of portal usage (OR: 2.30, p < 0.001), while residence in zip codes with poor broadband internet access was associated with lower odds of portal usage (OR: 0.97, p < 0.001). Conclusion: While others have reported portal usage rates as high as 64% in pediatric settings, our analysis suggests proxy portal usage rates of 30% in pediatric FQHC settings, with race/ethnicity, age group, and chronic disease status being significant drivers of portal non-usage. These findings highlight the need for appropriate and responsive health information technology approaches for vulnerable populations receiving care in low-resource settings.

Genetic risk scores identify people at high risk of developing diabetic kidney disease: A systematic review.

CONTEXT: Diabetic kidney disease (DKD) is the leading cause of end-stage renal disease (ESRD). Measures to prevent and treat DKD require better identification of patients most at risk. In this systematic review, we summarise the existing evidence of genetic risk scores (GRSs) and their utility for predicting DKD in people with type 1 or type 2 diabetes. EVIDENCE ACQUISITION: We searched MEDLINE, Embase, Web of Science and Cochrane Reviews in June 2022 to identify all existing and relevant literature. Main data items sought were study design, sample size, population, single nucleotide polymorphisms (SNPs) of interest, DKD-related outcomes, and relevant summary measures of result. The Critical Appraisal Skills Programme checklist was used to evaluate the methodological quality of studies. EVIDENCE SYNTHESIS: We identified 400 citations of which 15 were included in this review. Overall, seven studies had positive results, five had mixed results and three had negative results. Most studies with the strongest methodological quality (n = 9) reported statistically significant and favourable findings of a GRS' association with at least one measure of DKD. CONCLUSIONS: This systematic review presents evidence of the utility of GRSs to identify people with diabetes that are at high risk of developing DKD. In practice, a robust GRS could be used at the first clinical encounter with a person living with diabetes in order to stratify their risk of complications. Further prospective research is needed.

An overview of COVID-19 in people with diabetes: Pathophysiology and considerations in the inpatient setting.

INTRODUCTION: The coronavirus disease (COVID-19) pandemic has continued to have a devastating impact on health worldwide. There has been a rapid evolution of evidence, establishing an increased risk of morbidity and mortality associated with diabetes and concurrent COVID-19. The objective of this review is to explore the current evidence for inpatient assessment and management of diabetes during the COVID-19 pandemic and highlight areas requiring further exploration. METHODS: A literature search of databases was conducted to November 2020 using variations on keywords SARS-CoV-2, COVID-19, SARS, MERS and diabetes. Information relating to the impact of diabetes on severity of COVID-19 infection, the impact of COVID-19 infection on diabetes management and diabetes-related complications was integrated to create a narrative review. DISCUSSION: People with diabetes and COVID-19 are at an increased risk of morbidity and mortality. It is important that people with both known and previously unrecognised diabetes and COVID-19 be promptly identified and assessed during acute illness, with close monitoring for clinical deterioration or complications. People with diabetes may require titration or alteration of their glycaemic management due to the potential for worse outcomes with hyperglycaemia and COVID-19 infection. Comprehensive discharge planning is vital to optimise ongoing glycaemic management. CONCLUSION: Further understanding of the risk of adverse outcomes and optimisation of glycaemic management for people with diabetes during COVID-19 is required to improve outcomes. Increased glucose and ketone monitoring, substitution of insulin for some oral anti-hyperglycaemic medications and careful monitoring for complications of diabetes such as diabetic ketoacidosis should be considered.

Considerations for people with diabetes during the Coronavirus Disease (COVID-19) pandemic.

INTRODUCTION: The severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) continues to cause havoc globally, resulting in unprecedented healthcare, societal and economic disruption. People with diabetes have been shown to be at higher risk of complications and death when exposed to pneumonia, influenza and other coronaviruses. Despite pandemic scale infection, there is currently limited understanding on the potential impact of coronavirus disease (COVID-19) on people with diabetes. AIMS: (1) To characterise the outcomes of COVID-19 for people with diabetes and (2) add value to current recommendations for healthcare providers and people with diabetes to encourage optimal management. METHODS: A search of PubMed, Embase and MEDLINE to March 2020 was undertaken, using search terms pertaining to diabetes, coronavirus and acute respiratory distress syndrome (ARDS). We briefly reviewed the epidemiology and pathophysiology of SARS-CoV-2 in the context of diabetes. CONCLUSION: People with diabetes are at greater risk of severe infection and death with COVID-19. COVID-19 has significantly impacted the daily lives of individuals living with diabetes through financial implications, food and medication scarcity and its burden on mental health. In Australia, delivery of medical care has been adapted to reduce the risk of transmission, with a particular emphasis on telehealth and remote monitoring.

Implementation of an Extended-Infusion Piperacillin-Tazobactam Dosing Protocol: Unexpected Findings when Monitoring Safety and Compliance with Smart Pump Technology.

Compliance with recommended infusion rates was evaluated before, during, and after the implementation of extended-infusion (EI) piperacillin-tazobactam at an academic medical center. Software-controlled infusion-pump alert data were studied for piperacillin-tazobactam administrations before and after implementation of a four-hour EI protocol. Compliance was analyzed 16 weeks before (pre-EI), two weeks after (peri-EI), and an additional 16 weeks after (post-EI) protocol implementation. We defined potential harm as a programmed infusion rate exceeding the recommended rate, possible harm as a programmed infusion aborted by the user, and compliance as reversion to recommended rates. Potential and possible harm were standardized to 1000 patient days. Overall, 3110 alerts were identified during the period. Potential harm per 1000 patient days for pre-, peri-, and post-EI were 0, 6.12, and 1.05 (p < 0.001). Possible harm per 1000 patient days for the pre-, peri-, and post-EI were 0.33, 21.9, and 5.02 (p < 0.001). Compliance after an initial potential harm alert occurred more often post-EI (0.4 per 1000 patient days vs. 0 per 1000 patient days for pre- and peri-EI; p < 0.001), while alerts remaining in non-compliance were more prevalent if they initially occurred during the peri- and post-EI vs. pre-EI (6.1 and 0.6 per 1000 patient days vs. 0 per 1000 patient days; p < 0.001) period. Piperacillin-tazobactam infusions were administered faster than recommended during implementation (i.e., peri-EI) despite standardized orders.