Severe Insulin Pump-Related Adverse Events: Potential Root Causes and Impact of the COVID-19 Pandemic.

OBJECTIVE: To explore insulin pump-associated severe adverse events (SAEs) involving intensive care unit (ICU) admissions and deaths and examine the impact of the COVID-19 pandemic on these SAEs. METHODS: Qualitative template analysis of narrative data in reported insulin pump-associated SAEs occurring between May 1, 2019, and January 31, 2021, involving MiniMed 670G, MiniMed 630G, Omnipod, Omnipod DASH, and t:slim X2 insulin pumps. RESULTS: Over the 21-month measurement period, 460 SAEs involving an ICU admission and 288 SAEs involving a death were reported to the Food and Drug Administration. Problems with the pump or pod reservoir/cartridge were among the most frequently cited potential root causes in SAEs involving ICU admissions and deaths overall. However, problems with the infusion set or site and the pump battery or power emerged in the top three potential root causes of SAEs involving an ICU admission, whereas the patient sleeping at the time of the event and the tasks of changing the pod/infusion set, including reservoir/cartridge and programming the pump emerged in the top three for SAEs involving a death. The median monthly number of reported SAEs involving ICU admissions and deaths decreased during the pandemic, but their potential root causes were unchanged. CONCLUSIONS: Although insulin pumps are generally safe, SAEs related to their components and external factors can and do occur. By learning from the potential root causes of insulin pump-associated SAEs, providers and patients can implement corrective actions to prevent future events, thereby reducing harm.

Insulin Pump Alarms During Adverse Events: A Qualitative Descriptive Study.

OBJECTIVE: Explore alarm signals cited in insulin pump-associated adverse events (AEs), describe the clinical consequences and other root cause informing remarks that cooccurred with the alarm signals, and identify opportunities for improvements to patient education, instructional materials, and alarm systems to prevent future AEs. RESEARCH DESIGN AND METHODS: We explored the type, frequency, and associated clinical consequences of alarm signals cited in a pre-coded data set of 2294 insulin pump-associated AEs involving the MiniMed 670G, MiniMed 630G, and t:slim X2. We also explored the clinical consequences and other root cause informing remarks that cooccurred with the top 10 most frequently cited alarm signals. RESULTS: Overall, 403 AEs narratives cited at least one alarm signal. Of the 40 unique alarm signals cited, 42.5% were "alarms," 25.0% were "alerts," and 32.5% were not referenced in the instructional materials packaged with the corresponding pump. The top 10 most frequently cited alarm signals included two obstruction of flow alarms, which accounted for 49.9% of all AEs citing at least one alarm, and two unreferenced alarms. The most frequent cooccurring root cause informing remark varied across the top 10 alarm signals and revealed valuable insight into why these alarms may have occurred. CONCLUSIONS: Our findings demonstrate the value of analyzing alarm signals cited in insulin pump-associated AEs and reveal multiple opportunities for providers to educate patients on how to respond to alarm signals and manage their pumps to avoid AEs, and for insulin pump manufacturers to update instructional materials and improve alarm systems to support appropriate patient response.

A Retrospective Cohort Study to Evaluate Adding Biomarkers to the Risk Analysis Index of Frailty.

INTRODUCTION: The Risk Analysis Index (RAI) is a frailty assessment tool associated with adverse postoperative outcomes including 180 and 365-d mortality. However, the RAI has been criticized for only containing subjective inputs rather than including more objective components such as biomarkers. METHODS: We conducted a retrospective cohort study to assess the benefit of adding common biomarkers to the RAI using the Veterans Affairs Surgical Quality Improvement Program (VASQIP) database. RAI plus body mass index (BMI), creatinine, hematocrit, and albumin were evaluated as individual and composite variables on 180-d postoperative mortality. RESULTS: Among 480,731 noncardiac cases in VASQIP from 2010 to 2014, 324,320 (67%) met our inclusion criteria. Frail patients (RAI ≥30) made up to 13.0% of the sample. RAI demonstrated strong discrimination for 180-d mortality (c = 0.839 [0.836-0.843]). Discrimination significantly improved with the addition of Hematocrit (c = 0.862 [0.859-0.865]) and albumin (c = 0.870 [0.866-0.873]), but not for body mass index (BMI) or creatinine. However, calibration plots demonstrate that the improvement was primarily at high RAI values where the model overpredicts observed mortality. CONCLUSIONS: While RAI's ability to predict the risk of 180-d postoperative mortality improves with the addition of certain biomarkers, this only observed in patients classified as very frail (RAI >49). Because very frail patients have significantly elevated observed and predicted mortality, the improved discrimination is likely of limited clinical utility for a frailty screening tool.

Understanding the Context for Incorporating Equity into Quality Improvement Throughout a National Health Care System.

Purpose: Although health care systems aspire to deliver equitable care, practical tools that empower the health care workforce to weave equity throughout quality improvement (QI) processes are lacking. In this article, we report findings from context of use interviews that informed the development of a user-centered tool to support equity-focused QI. Methods: Semistructured interviews were conducted from February to April of 2019. Participants included 14 medical center administrators, departmental or service line leaders, and clinical staff involved in direct patient care from three Veterans Affairs (VA) Medical Centers within a single region. Interviews covered existing practices for monitoring health care quality (i.e., priorities, tasks, workflow, and resources) and explored how equity data might fit into current processes. Themes extracted through rapid qualitative analysis were used to draft initial functional requirements for a tool to support equity-focused QI. Results: Although the potential value of examining disparities in health care quality was clearly recognized, the data necessary for examining disparities were lacking for most quality measures. Interviewees also desired guidance on how inequities could be addressed through QI. The ways in which QI initiatives were selected, carried out, and supported also had important design implications for tools to support equity-focused QI. Discussion: The themes identified in this work guided the development of a national VA Primary Care Equity Dashboard to support equity-focused QI within VA. Understanding the ways in which QI was carried out across multiple levels of the organization provided a successful foundation upon which to build functional tools to support thoughtful engagement around equity in clinical settings.

Interpreting the risk analysis index of frailty in the context of surgical oncology.

BACKGROUND AND OBJECTIVES: The Risk Analysis Index (RAI) accurately predicts adverse postoperative outcomes but the inclusion of cancer status in the RAI has raised two key concerns about its suitability for use in surgical oncology: (1) the potential over classification of cancer patients as frail, and (2) the potential overestimation of postoperative mortality for patients with surgically curable cancers. METHODS: We performed a retrospective cohort analysis to assess the RAI's power to appropriately identify frailty and predict postoperative mortality in cancer patients. We assessed discrimination for mortality and calibration across five RAI models-the complete RAI and four variants that removed different cancer-related variables. RESULTS: We found that the presence of disseminated cancer was a key variable driving the RAI's power to predict postoperative mortality. The model including only this variable [RAI (disseminated cancer)] was similar to the complete RAI in the overall sample (c = 0.842 vs. 0.840) and outperformed the complete RAI in the cancer subgroup (c = 0.736 vs 0.704, respectively, p < 0.0001, Max R2 = 19.3% vs. 15.1%, respectively). CONCLUSION: The RAI demonstrates somewhat less discrimination when applied exclusively to cancer patients, but remains a strong predictor of postoperative mortality, especially in the setting of disseminated cancer.

Insulin Pump-Associated Adverse Events: A Qualitative Descriptive Study of Clinical Consequences and Potential Root Causes.

Objective: To explore the clinical consequences and potential root causes of insulin pump-associated adverse events (AEs) reported in the Food and Drug Administration's Manufacturer and User Facility Device Experience (MAUDE) database. Research Design and Methods: Qualitative template analysis of narrative data in a 20% stratified random sample (n = 2429) of reported AEs that occurred during the first 6 months of 2020 involving five insulin pump models marketed at that time: (1) MiniMed™ 670G, (2) MiniMed™ 630G, (3) Omnipod®, (4) Omnipod DASH®, and (5) t:slim X2™. Results: Of the 2429 AEs, 92% included a clinical consequence in the narrative description, with critical hyperglycemia (i.e., blood glucose [BG] >400 mg/dL; 47%) and critical hypoglycemia (i.e., BG <54 mg/dL; 24%) being the most common consequence cited. Only 50% of the AE narratives included information to support the identification of a root cause. The most cited root cause informing remarks were issues with the pump or pod reservoir/cartridge (9%), the occurrence of an obstruction of flow alarm (8%), and problems with the infusion set or site (8%). Some clinical consequences and root cause informing remarks were cited more frequently in AE narratives involving specific insulin pump models, but manufacturer variability in the amount and type of information reported may have affected these findings. Conclusions: Our findings show general themes found in insulin pump-associated AE that providers can use to raise patient awareness of potential risks associated with insulin pump use and develop strategies to prevent future AEs. Improvements in AE investigation and reporting processes are still necessary.

Preferred Terms and Icons for Labels on Electrosurgical Units: Survey of VA Nurses.

Electrosurgical units (ESUs) developed by different manufacturers use varying terminology and icons to label the same components, which can result in confusion among users and the potential for erroneous ESU configuration. The objective of the current study was to identify nurse-preferred terms and icons for labeling ESU components. A total of 165 operating room (OR) nurses from Veterans Health Administration facilities across the United States were surveyed regarding terms and icons found on 25 ESU models. The results showed that 81% of OR nurses preferred ESUs that included both a term and an icon for labeling each component. In addition, greater consensus existed among OR nurses regarding preferred terms, rather than preferred icons, for representing each component. These findings on OR nurses' preferred terms and icons can be leveraged to improve ESU labeling practices and inform the development of a standardized, user-centered set of labels for ESU components.

Effects of Cross-Training on Medical Teams' Teamwork and Collaboration: Use of Simulation.

Previous research in the US Navy demonstrated that cross-training enhances teamwork and interpersonal collaboration. Limited data exists on cross-training effectiveness in medical education. This research aimed to assess whether cross-training would have similar effects on medical teams. A multidisciplinary pair of resident participants-consisting of one physician and one pharmacist-was randomly assigned to cross-training or current training condition. The training experience involved one video-based content module (training a pharmacist's task of pharmacokinetic dosing and a physician's task of intubation) and one simulation-based practice scenario (collaborative treatment of an unstable critically ill simulated patient). Interprofessional pairs randomized to cross-training condition participated in both the content module and practice scenario in the alternative professional role whereas pairs randomized to current training condition participated in their own professional role. Pairs also participated in pre- and post- training assessment scenarios in their own professional role. Teamwork and interprofessionalism were measured immediately following assessment scenarios. Knowledge assessments were conducted at the start and end of the scenario sequence. Multidisciplinary pairs experiencing cross-training showed a significant improvement in teamwork (increased by 6.11% vs. 3.24%, p < 0.05). All participants demonstrated significant improvement in knowledge scores (increase of 14% cross-training, p < 0.05, and increase of 13.9% control, p < 0.05). Our project suggests that cross-training can improve teamwork in interprofessional medical teams.

Reducing Treatment Errors Through Point-of-Care Glucometer Configuration.

BACKGROUND: Blood glucose (BG) testing is the most widely performed point-of-care (POC) test in a hospital setting. Multiple adverse events reported to the Food and Drug Administration (FDA) revealed that treatment decisions may be affected by information displayed on the POC glucometer's results screen. A randomized, crossover simulation study was conducted to compare two results screen configurations for ACCU-CHEK Inform II, a POC glucometer. METHODS: Prior to the study, a heuristic evaluation of the results screen configurations and a pilot study were conducted to select the two results screen configurations for comparison. At two multicampus medical centers, 66 nurse participants experienced two computer-based simulation scenarios that asked them to interpret glucometer readings and make treatment decisions for simulated patients with 32 mg/dL BG levels and subtle symptoms of hypoglycemia. One scenario displayed a numeric value ("32 mg/dL"), and the other displayed a range abbreviation, such as "RR LO" (out of reportable range; low). Treatment errors were recorded when the participant did not treat the hypoglycemic patient with glucose or when they administered insulin. RESULTS: When ACCU-CHEK Inform II displayed an "RR LO" reading, 10.6% of participants made a treatment error, including 6.7% of participants with prior training on the meaning of an "RR LO" reading. None of the participants made a treatment error when ACCU-CHEK Inform II displayed a "32 mg/dL" reading. CONCLUSION: Displaying a numeric BG reading eliminated potentially life-threating treatment errors caused by confusing range abbreviations. Manufacturers should consider these findings during future research and development of POC glucometers.

Assessing Use Errors Related to the Interface Design of Electrosurgical Units.

Medical device use errors, such as instrument connection errors made with electrosurgical units (ESUs), can lead to adverse events. Current device acquisition processes at health care facilities do not typically include a proactive evaluation of use-error risk before device purchase. We conducted an evaluation to identify ESU user interface design features that can help prevent or mitigate instrument connection errors during clinical care. Thirty-six current ESU users participated in the evaluation. We used a randomized crossover design in which each participant used two ESU models in a simulated OR scenario. We compared participants' instrument connection accuracy, efficiency, and subjective feedback regarding the user interface design across the two ESU models. Overall, we found that the ESU model that incorporated more user interface design principles resulted in better performance and increased acceptance from users. Based on the results, we designed a decision-support tool to assess the risk of instrument connection errors before ESU purchase.

Label Design Affects Medication Safety in an Operating Room Crisis: A Controlled Simulation Study.

OBJECTIVE: Several factors contribute to medication errors in clinical practice settings, including the design of medication labels. The objective of this study was to quantify the impact of label design on medication safety in a realistic, high-stress clinical situation. METHODS: Ninety-six anesthesia trainee participants were randomly assigned to either the redesigned or the current label condition. Participants were blinded to the study's focus on medication label design and their assigned label condition. Each participant was the sole anesthesia provider in a simulated operating room scenario involving an unexpected vascular injury. The surgeon asked the participant to administer hetastarch to the simulated patient because of hemodynamic instability. The fluid drawer of the anesthesia cart contained three 500-ml intravenous bags of hetastarch and one 500-ml intravenous bag of lidocaine. We hypothesized that redesigned labels would help participants correctly select hetastarch from the cart. If the participants incorrectly selected lidocaine from the cart, we hypothesized that the redesigned labels would help participants detect the lidocaine before administration. RESULTS: The percentage of participants who correctly selected hetastarch from the cart was significantly higher for the redesigned labels than the current labels (63% versus 40%; odds ratio, 2.61 [95% confidence interval, 1.1-6.1]; P = 0.03). Of the participants who incorrectly selected lidocaine from the cart, the percentage who detected the lidocaine before administration did not differ by label condition. CONCLUSIONS: The redesigned labels helped participants correctly select hetastarch from the cart, thus preventing some potentially catastrophic medication errors from reaching the simulated patient.

Comparison of chest compression interruption times across 2 automated devices: a randomized, crossover simulation study.

OBJECTIVE: The goal of this study was to compare chest compression interruption times required to apply, adjust, and remove 2 different automated chest compression (ACC) devices using the same evaluation protocol. METHODS: Twenty-nine registered nurses and respiratory therapists used 2 ACC devices in separate resuscitation scenarios involving a patient manikin simulating a 45-year-old man in cardiac arrest in his intensive care unit room. Device presentation was randomized, with half of the participants using LUCAS 2 in the first scenario and the other half using AutoPulse in the first scenario. RESULTS: The mean chest compression interruption time to apply the ACC device to the patient was significantly shorter for AutoPulse (mean [M] = 31.6 ± 8.44) than for LUCAS 2 (M = 39.1 ± 11.20; t(28) = 3.65, P = .001). The mean chest compression interruption time to remove the ACC device from the patient and resume manual compressions was also significantly shorter for AutoPulse (M = 6.5 ± 3.65) than for LUCAS 2 (M = 10.1 ± 3.97; t(26) = 3.36, P = .002). There was no difference in the mean chest compression interruption time to adjust the position of the ACC device on the patient between AutoPulse (M = 14.3 ± 5.24) and LUCAS 2 (M = 12.5 ± 3.89; t(23) = -1.45, P = .162). CONCLUSIONS: The results of this study trended in favor of AutoPulse. However, the interruption in chest compression to apply either device to the patient was notably longer than the maximum interruption time recommended by the American Heart Association.