A Systemwide Strategy to Embed Equity into Patient Safety Event Analysis.

BACKGROUND: There is a lack of framework to incorporate equity into event analysis. This quality improvement initiative involved the development of equity tools that were introduced in a two-hour interactive, case-based training across 11 acute care facilities at the largest municipal health care system in the United States. A pre and post survey (which included analysis of a clinical vignette) was also conducted to assess for knowledge and comfort embedding equity in patient safety event analysis, and to measure discomfort or distress during the training. A separate assessment was used to evaluate the tools. EQUITY TOOLS: A visual aid, the Patient Equity Wheel, was created to facilitate more comprehensive and robust health equity discussions by compiling a comprehensive list of equity categories, including internal, external, and organizational dimensions of equity. The Wheel was designed for use during each phase of event analysis. An Embedding Equity in Root Cause Analysis Worksheet was developed to aid in assessing considerations of equitable care in the investigation process and includes questions to ask staff to further assess bias or equitable care factors. INITIATIVE OUTCOME AND KEY INSIGHTS: Participant knowledge and level of comfort increased after training. The most commonly unrecognized categories of bias were Training/Competencies, Structural Workflow, and Culture/Norms. Most participants responded that they had no discomfort or distress during the training. Post-training feedback noted that the tools were being used across the system in various stages of event analysis and have been reported to improve health equity conversations.

Reducing Overuse of 3-Day Repeat Type and Screen Testing across an 11-Hospital Safety Net System.

BACKGROUND: According to the American Association of Blood Banks, a Type and Screen (T&S) is valid for up to three calendar days. Beyond a limited number of clinical indications such as a transfusion reaction, repeat T&S testing within 3 days is not warranted. Inappropriate repeat T&S testing is a costly medical waste and can lead to patient harm. OBJECTIVE: To reduce inappropriate duplicate T&S testing across a large, multihospital setting. SETTING: The largest urban safety net health system in the USA, with 11 acute care hospitals. INTERVENTIONS: Our first intervention involved adding the time elapsed since the last T&S order into the order and the process instructions that described when a T&S was indicated. The second intervention was a best practice advisory that triggered when T&S was ordered before the expiration of an active T&S. MAIN MEASURES: The primary outcome measure was the number of duplicate inpatient T&S per 1000 patient days. KEY RESULTS: Across all hospitals, the weekly average rate of duplicate T&S ordering decreased from 8.42 to 7.37 per 1000 patient days (12.5% reduction, p < 0.001) after the first intervention and to 4.32 per 1000 patient days (48.7% reduction, p < 0.001) after the second intervention. Using linear regression to compare pre-intervention to post-intervention 1, the level difference was - 2.46 (9.17 to 6.70, p < 0.001) and slope difference was 0.0001 (0.0282 to 0.0283, p = 1). For post-intervention 1 to post-intervention 2, the level difference was - 3.49 (8.06 to 4.58, p < 0.001) and slope difference was - 0.0428 (0.0283 to - 0.0145, p < 0.05). CONCLUSIONS: Our intervention successfully reduced duplicate T&S testing using a two-pronged electronic health record intervention. The success of this low effort intervention across a diverse health system provides a framework for similar interventions in various clinical settings.

Imaging Wisely Campaign: Initiative to Reduce Imaging for Low Back Pain Across a Large Safety Net System.

OBJECTIVES: Low back pain is a common clinical presentation that often results in expensive and unnecessary imaging that may lead to undue patient harm, including unnecessary procedures. We present an initiative in a safety net system to reduce imaging for low back pain. METHODS: This quality improvement study was conducted across 70 ambulatory clinics and 11 teaching hospitals. Three electronic health record changes, using the concept of a nudge, were introduced into orders for lumbar radiography (x-ray), lumbar CT, and lumbar MRI. The primary outcome was the number of orders per 1,000 patient-days or encounters for each imaging test in the inpatient, ambulatory, and emergency department (ED) settings. Variation across facilities was assessed, along with selected indications. RESULTS: Across all clinical environments, there were statistically significant decreases in level differences pre- and postintervention for lumbar x-ray (-52.9% for inpatient encounters, P < .001; -23.7% for ambulatory encounters, P < .001; and -17.3% for ED only encounters, P < .01). There was no decrease in ordering of lumbar CTs in the inpatient and ambulatory settings, although there was an increase in lumbar CTs in ED-only encounters. There was no difference in lumbar MRI ordering. Variation was seen across all hospitals and clinics. DISCUSSION: Our intervention successfully decreased lumbar radiography across all clinical settings, with a reduction in lumbar CTs in the inpatient and ambulatory settings. There were no changes for lumbar MRI orders.

Initiative to reduce unnecessary routine daily testing of complete blood counts across 11 safety net hospitals.

OBJECTIVES: National societies recommend against performing routine daily laboratory testing without a specific indication. Unnecessary testing can lead to patient harm, such as hospital-acquired anemia. The objective of this study was to reduce repeat complete blood counts (CBCs) after initial testing. METHODS: This was a quality improvement initiative implemented across 11 safety net hospitals in New York City. A best practice advisory (BPA) was implemented that asked the user to remove a CBC if the last 2 CBCs within 72 hours had normal white blood cell and platelet counts and unchanged hemoglobin levels. The outcome measure was the rate of CBCs per 1000 patient days preintervention (January 8, 2020, to December 22, 2020) to postintervention (December 23, 2020, to December 7, 2021). The process measure was the acceptance rate of the BPA, defined as the number of times the repeat CBC order was removed through the BPA divided by the total number of times the BPA triggered. RESULTS: Across 11 hospitals, repeat CBC testing decreased by 12.3% (73.05 to 64.04 per 1000 patient days, P < .001). Six of the 11 hospitals exhibited statistically significant decreases, ranging from a 10% to 48.9% decrease of repeat CBCs. The overall BPA action rate was 20.0% (24,029 of 119,944 repeat CBCs). CONCLUSIONS: This low-effort, electronic health record-based intervention can effectively reduce unnecessary laboratory testing.

Reducing low-value ED coags across 11 hospitals in a safety net setting.

BACKGROUND: Prothrombin/international normalized ratio and activated partial thromboplastin time (PT/INR and aPTT) are frequently ordered in emergency departments (EDs), but rarely affect management. They offer limited utility outside of select indications. Several quality improvement initiatives have shown reduction in ED use of PT/INR and aPTT using multifaceted interventions in well-resourced settings. Successful reduction of these low-value tests has not yet been shown using a single intervention across a large hospital system in a safety net setting. This study aims to determine if an intervention of two BPAs is associated with a reduction in PT/INR and aPTT usage across a large safety net system. METHODS: This initiative was set at a large safety net system in the United States with 11 acute care hospitals. Two Best Practice Advisories (BPAs) discouraging inappropriate PT/INR and aPTT use were implemented from March 16, 2022-August 30, 2022. Order rate per 100 ED patients during the pre-intervention period was compared to the post-intervention period on both the system and individual hospital level. Complete blood count (CBC) testing served as a control, and packed red blood cell transfusions served as a balancing measure. An interrupted time series regression analysis was performed to capture immediate and temporal changes in ordering for all tests in the pre and post-intervention periods. RESULTS: PT/INR tests exhibited an absolute decline of 4.11 tests per 100 ED encounters (95% confidence interval -5.17 to -3.05; relative reduction of 18.9%). aPTT tests exhibited absolute decline of 4.03 tests per 100 ED encounters (95% CI -5.10 to -2.97; relative reduction of 19.8%). The control measure, CBC, did not significantly change (-0.43, 95% CI -2.83 to 1.96). Individual hospitals showed variable response, with absolute reductions from 2.02 to 9.6 tests per 100 ED encounters for PT/INR (relative reduction 12.1%-30.5%) and 2.07 to 10.04 for aPTT (relative reduction 12.1%-31.4%). Regression analysis showed that the intervention caused an immediate 25.7% decline in PT/INR and 24.7% decline in aPTT tests compared to the control measure. The slope differences (rate of order increase pre vs post intervention) did not significantly decline compared to the control. CONCLUSIONS: This BPA intervention reduced PT/INR and aPTT use across 11 EDs in a large, urban, safety net system. Further study is needed in implementation to other non-safety net settings.

Choosing Wisely and reducing the simultaneous ordering of erythrocyte sedimentation rate and C-reactive protein testing in a large safety net system.

OBJECTIVES: C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR) are commonly used inflammatory markers. C-reactive protein is more sensitive and specific for monitoring acute inflammation. However, it is commonly co-ordered with ESR despite recommendations against this. Our objective was to reduce unnecessary ESR orders and ESR/CRP co-ordering rates across a large safety net health system. METHODS: This was a quality improvement project that used a quasi-experimental pre- and postintervention interrupted time-series regression analysis. Patients with a positive COVID-19 test were excluded. We designed a nonintrusive, normative nudge within the ESR order that recommended against co-ordering ESR and CRP. In addition, a best practice advisory triggered when ESR and CRP were simultaneously ordered. The outcome measures were ESR order rates per 1000 patient days in the inpatient setting and per 1000 patient encounters in the outpatient setting, as well as ESR/CRP co-ordering rates. RESULTS: Inpatient ESR orders decreased from 12.02 preintervention to 5.61 per 1000 patient days (-53.3%, P < .001). Outpatient ESR orders decreased from 6.09 to 4.07 per 1000 patient encounters (-33.2%, P < .001). Relative co-ordering rates decreased by 50%. CONCLUSIONS: This electronic health record initiative successfully reduced ESR testing across 11 hospitals and 70 ambulatory centers in a safety net setting.

Reducing daily dosing in opioid prescriptions in 11 safety net emergency departments.

BACKGROUND: The United States continues to face a significant issue with opioid misuse, overprescribing, dependency, and overdose. Electronic health record (EHR) interventions have shown to be an effective tool to modify opioid prescribing behaviors. This quality improvement project describes an EHR intervention to reduce daily dosing in opioid prescriptions in 11 emergency departments (ED) across the largest safety net health system in the US. MEASURES: The primary outcome measure was the rates of oxycodone-acetaminophen 5-325 mg prescriptions exceeding 50 morphine milligram equivalents per day (MMED) pre- vs. post-intervention; and stratified by individual hospitals and provider type. INTERVENTION: The defaults for dose and frequency were uniformly changed to 'every 6 hours as needed' and '1 tablet', respectively, across 11 EDs. OUTCOMES: The percentage of prescriptions greater than or equal to 50 MMED decreased from 46.0% (1624 of 3530 prescriptions) to 1.6% (52 of 3165 prescriptions) (96.4% relative reduction; p < 0.001). All 11 hospitals had a significant reduction in prescriptions exceeding 50 MMED. Nurse practitioners had the highest relative reduction of prescriptions exceeding 50 MMED at 100% (p < 0.001), and the attendings/fellows had the lowest relative reduction at 95.6% (p < 0.001). CONCLUSIONS/LESSONS LEARNED: Default nudges are a simple yet powerful intervention that can strongly influence opioid prescribing patterns.

Reducing 25-Hydroxyvitamin D Testing in a Large, Urban Safety Net System.

BACKGROUND: 25-Hydroxyvitamin D testing is increasing despite national guidelines and Choosing Wisely recommendations against routine screening. Overuse can lead to misdiagnosis and unnecessary downstream testing and treatment. Repeat testing within 3 months is a unique area of overuse. OBJECTIVE: To reduce 25-hydroxyvitamin D testing in a large safety net system comprising 11 hospitals and 70 ambulatory centers. DESIGN: This was a quality improvement initiative with a quasi-experimental interrupted time series design with segmented regression. PARTICIPANTS: All patients in the inpatient and outpatient settings with at least one order for 25-hydroxyvitamin D were included in the analysis. INTERVENTIONS: An electronic health record clinical decision support tool was designed for inpatient and outpatient orders and involved two components: a mandatory prompt requiring appropriate indications and a best practice advisory (BPA) focused on repeat testing within 3 months. MAIN MEASURES: The pre-intervention period (6/17/2020-6/13/2021) was compared to the post-intervention period (6/14/2021-8/28/2022) for total 25-hydroxyvitamin D testing, as well as 3-month repeat testing. Hospital and clinic variation in testing was assessed. Additionally, best practice advisory action rates were analyzed, separated by clinician type and specialty. KEY RESULTS: There were 44% and 46% reductions in inpatient and outpatient orders, respectively (p < 0.001). Inpatient and outpatient 3-month repeat testing decreased by 61% and 48%, respectively (p < 0.001). The best practice advisory true accept rate was 13%. CONCLUSION: This initiative successfully reduced 25-hydroxyvitamin D testing through the use of mandatory appropriate indications and a best practice advisory focusing on a unique area of overuse: the repeat testing within a 3-month interval. There was wide variation among hospitals and clinics and variation among clinician types and specialties regarding actions to the best practice advisory.

Reducing duplicate genetic testing in inpatient and outpatient settings across a large safety-net system.

OBJECTIVES: To develop a system-wide electronic health record (EHR) intervention at a large safety-net hospital system to reduce wasteful duplicate genetic testing. METHODS: This project was initiated at a large urban public health care system. An EHR alert was designed to be triggered when a clinician attempted to order any of 16 specified genetic tests for which a previous result existed within the EHR system. Measurements included the proportion of completed genetic tests that were duplicates and alerts per 1,000 tests. Data were stratified by clinician type, specialty, and inpatient vs ambulatory setting. RESULTS: Across all settings, the rate of duplicate genetic testing decreased from 2.35% (1,050 of 44,592 tests) to 0.09% (21 of 22,323 tests) (96% relative reduction, P < .001). The alert rate per 1,000 tests was 277 for inpatient orders and 64 for ambulatory orders. Among clinician types, residents had the highest alert rate per 1,000 tests at 166 and midwives the lowest at 51 (P < .01). Among clinician specialties, internal medicine had the highest alert rate per 1,000 tests at 245 and obstetrics and gynecology the lowest at 56 (P < .01). CONCLUSIONS: The EHR intervention successfully reduced duplicate genetic testing by 96% across a large safety-net setting.

Reducing Inappropriate Simultaneous Ordering of Heparin Antibody and Serotonin Release Assays.

INTRODUCTION: A 4T score with intermediate or high probability of heparin-induced thrombocytopenia prompts ordering of anti-platelet 4 heparin complex. If positive, a serotonin release assay (SRA) is recommended to confirm diagnosis. Despite these recommendations, overtesting of both anti-platelet 4 and SRA is highly prevalent. METHODS: This was a quality improvement initiative using two forms of clinical decision support across 11 acute care hospitals. First, a 4T calculator was incorporated into anti-platelet 4 orders. Second, a Best Practice Advisory fired when anti-platelet 4 and SRA were ordered simultaneously, prompting the provider to remove the SRA order. Data were analyzed by a quasi-experimental interrupted time series linear regression comparing weekly average laboratory tests per 1,000 patient-days pre- and postintervention. RESULTS: Average ordering frequency of anti-platelet 4 changed from 0.508 to 0.510 per 1,000 patient-days (0.5%, p = 0.42) without significant slope or level differences. Average ordering frequency of SRA decreased from 0.430 to 0.289 per 1,000 patient-days (32.8%, p < 0.001) with a significant level difference of -0.128 orders per 1,000 patient-days (-31.2%, p < 0.05). CONCLUSION: A simultaneous Best Practice Advisory was effective in reducing SRA orders, but not anti-platelet 4 orders.

Reduction of Intravenous Antihypertensives through Clinical Decision Support in a Large Safety Net System.

BACKGROUND: Asymptomatic severe hypertension (also known as hypertensive urgency) is frequently encountered in the hospital. Previous evidence suggests that management with one-time doses of intravenous (IV) antihypertensives may increase adverse events. Despite this, single-dose treatment remains common in the emergency department and inpatient settings. METHODS: This quality initiative was launched at New York City Health + Hospitals, the largest safety net hospital system in the United States. The initiative involved two changes to electronic orders for IV hydralazine and IV labetalol: a nonintrusive advisory statement within the order instructions and a mandatory requirement to document the indication for IV antihypertensive use. RESULTS: This initiative took place from November 2021 to October 2022. Of the indications selected for IV antihypertensive orders, 60.7% were for hypertensive emergency, 15.3% were for patients who were strictly NPO, 21.2% were for other, and 2.8% selected more than one indication. For ED-only encounters, aggregate IV hydralazine and IV labetalol orders per 1,000 patient encounters were 2.53 preintervention and 1.55 postintervention (38.7% reduction, p < 0.001). For inpatient encounters, aggregate IV hydralazine and IV labetalol orders per 1,000 patient-days were 18.25 preintervention and 15.81 postintervention (13.4% reduction, p < 0.001). Similar trends were observed for individual orders of IV hydralazine and IV labetalol. There were significant reductions in 7 of the 11 hospitals in inpatient administration of aggregate IV hydralazine and labetalol orders per 1,000 patient-days. CONCLUSION: This quality improvement initiative successfully reduced unnecessary IV antihypertensive use in an 11-hospital safety net system.

An interdisciplinary student-led multifaceted intervention addressing overuse of broad-spectrum antibiotics for patients with penicillin allergies.

BACKGROUND: Though 15% of hospitalized patients have a documented penicillin (PCN) allergy, fewer than 1% have an IgE-mediated reaction that necessitates avoidance of ß-lactam antibiotics. OBJECTIVE: Our interdisciplinary team of medical and nursing students led and executed a two-pronged quality improvement intervention to reduce prescribing of non-ß-lactam antibiotics (NBLs) for patients with reported PCN allergies. To the best of our knowledge, this is the first multidisciplinary student-led intervention aimed at educating providers on low-risk penicillin allergy and encouraging best antibiotic prescribing practices. DESIGN AND PARTICIPANTS: The intervention took place from June 2021 to February 2022. We developed and provided clinician education modules, including peer-to-peer information sharing and in-person small group discussions, as well as clinical decision support (CDS) strategies through the electronic medical record (EMR). The target population was attendings, residents, nurse practitioners, and physician assistants on the hospital medicine service at a large urban academic tertiary care center. We followed the SQUIRE 2.0 guidelines for reporting on quality improvement. MAIN MEASURES: Primary outcome measures included number of NBL prescriptions and use of nonspecific descriptors (e.g., "other" or "unknown") for PCN allergy reaction type, and were compared with a pre-intervention period. KEY RESULTS: The percent of ß-lactam prescriptions for patients with a PCN allergy after the intervention increased from 19 to 23% (p = 0.006). For patients with a low severity PCN allergy, the percent of ß-lactam prescriptions increased from 20 to 28% (p = 0.001). There was a significant decrease in nonspecific PCN allergy reaction type from 23% in the pre-intervention period to 20% post-intervention (p = 0.012). CONCLUSIONS: An intervention focused on educating prescribers and CDS strategies delivered through the EMR increased appropriate ß-lactam prescribing for patients with a documented low-risk PCN allergy and reduced the use of nonspecific PCN allergy reaction type in EMR documentation.

Choosing wisely initiative for reducing urine cultures for asymptomatic bacteriuria and catheter-associated asymptomatic bacteriuria in an 11-hospital safety net system.

BACKGROUND: Treatment of asymptomatic bacteriuria (ASB) is common. Overtreatment of ASB leads to harm, including adverse effects from antibiotics, antibiotic resistance, and increased length of stay. METHODS: This quality improvement initiative targeted inappropriate urine cultures across 11 hospitals in a safety-net setting. A mandatory prompt for appropriate indications for urine culture orders and a best practice advisory (BPA) for urine culture on patients with urinary catheters were created. Urine culture ordering was compared pre-intervention (6/2020 to 10/2021) to post-intervention (12/2021 to 8/2022). Catheter associated urinary tract infections (CAUTI) was compared pre- and post-intervention. Variation in urine culture ordering and CAUTI rates in hospitals were assessed. RESULTS: Inpatient urine cultures decreased by 20.9% (p<0.001). Inpatient urine cultures on patients with urinary catheters decreased by 21.6% (p<0.001). CAUTI rates remained unchanged post-intervention. High variation in urine culture ordering and CAUTI rates was seen among hospitals. CONCLUSIONS: This initiative successfully decreased urine cultures in a large, safety-net system. Further study is needed in assessing variation among hospitals.

Nudging to select single-lumen over multiple-lumen peripherally inserted central catheters (PICCs) in a large safety net system.

BACKGROUND: Peripherally inserted central catheters (PICCs) are increasingly used for vascular access in inpatient settings. Compared to multilumen PICCs, single-lumen PICCs carry a lower rate of complications, including central-line-associated bloodstream infection and thrombosis. Despite this, multilumen PICCs are still overused. METHODS: This quality improvement initiative was implemented across 11 hospitals at New York City Health + Hospitals safety net system. The electronic health record (EHR) interventional radiology or vascular access team consultation orders were modified to allow for lumen choice, with default selection to a single-lumen PICC. RESULTS: Average single-lumen PICC utilization increased by 25.5%, from 44.4% to 69.9% (P < .001). CLABSI rates had a nonsignificant reduction by 26.7% from 2.44 to 1.79 infections per month (P = .255). Among provider types in the postintervention period, single-lumen PICC utilization ranged from 67.7% for advanced practice providers to 82.4%-94.6% for physicians. Among provider specialties, utilization ranged from 31.8% for neurology to 97.7% for orthopedics. Additionally, there was large variation in pre- and postintervention differences in utilization by hospital. CONCLUSIONS: We successfully increased single-lumen PICC utilization across all 11 safety net hospitals. This expands on previous work on improving single-lumen PICC use and use of default nudges in large, resource-limited settings. Further study is needed to examine variation among provider types, specialties, and hospitals.

Diarrhea don'ts: Reducing inappropriate stool cultures and ova and parasite testing for nosocomial diarrhea.

BACKGROUND: Diarrhea that develops in patients after 72 hours of hospitalization is likely to have a nosocomial or iatrogenic etiology. Testing with stool cultures and stool ova and parasites (O&P) is not recommended. Our goal was to reduce this inappropriate testing within a large, urban safety-net hospital system. METHODS: This was a quality improvement project. We created a best practice advisory (BPA) within the electronic medical record that fires when a stool culture or O&P order is placed 72 hours after admission for any immunocompetent patient. It states that stool testing is low yield and offers the option to remove the order. We measured weekly counts of stool culture and stool O&P orders pre- and postintervention. We also measured the BPA acceptance rate, the 24-hour stool testing reorder rate, and Clostridioides difficile infection rates. Data were analyzed using Welch tests as well as a quasi-experimental pre- and postintervention interrupted time series regression analysis. RESULTS: Stool culture orders decreased by 24.4% (P < .001). There was a significant level difference and slope difference with linear regression. Five of the 11 hospitals had a significant reduction in stool culture orders. Stool O&P orders decreased by 18.2% (P < .01). Three of the 11 hospitals had a significant reduction in stool O&P orders. CONCLUSIONS: Our intervention successfully reduced inappropriate stool testing within a large safety-net hospital system.

Interventions to reduce repetitive ordering of low-value inpatient laboratory tests: a systematic review.

BACKGROUND: Over-ordering of daily laboratory tests adversely affects patient care through hospital-acquired anaemia, patient discomfort, burden on front-line staff and unnecessary downstream testing. This remains a prevalent issue despite the 2013 Choosing Wisely recommendation to minimise unnecessary daily labs. We conducted a systematic review of the literature to identify interventions targeting unnecessary laboratory testing. METHODS: We systematically searched MEDLINE, EMBASE, Cochrane Central and SCOPUS databases to identify interventions focused on reducing daily complete blood count, complete metabolic panel and basic metabolic panel labs. We defined interventions as 'effective' if a statistically significant reduction was attained and 'highly effective' if a reduction of ≥25% was attained. RESULTS: The search yielded 5646 studies with 41 articles that met inclusion criteria. We grouped interventions into one or more categories: audit and feedback, cost display, education, electronic medical record (EMR) change, and policy change. Most interventions lasted less than a year and used a multipronged approach. All five strategies were effective in most studies with EMR change being the most commonly used independent strategy. EMR change and policy change were the strategies most frequently reported as effective. EMR change was the strategy most frequently reported as highly effective. CONCLUSION: Our analysis identified five categories of interventions targeting daily laboratory testing. All categories were effective in most studies, with EMR change being most frequently highly effective. PROSPERO REGISTRATION NUMBER: CRD42021254076.