Lucy B. Schulson

Objectives: The COVID-19 pandemic challenged hospitals. Yet, little is known about the extent of COVID-19’s impact on patient safety. We assessed change in Patient Safety Indicators (PSIs), which identify potentially preventable adverse events, before and after the onset of COVID-19 and when a hospital had a high proportion of COVID-19 admissions. Methods: Using Medicare Provider Analysis and Review data, our sample included 51,960,029 non-COVID-19 admissions eligible for inclusion in at least 1 PSI. This included 23,262,714 patients and a total of 294,871 PSI events. We fit risk-adjusted trends from 2018 to 2022 using interrupted time series models of PSIs before and after the onset of COVID-19. We also tested for the difference of a hospital having a high proportion of COVID-19 admissions. PSIs were categorized into provider-resource, surgical/team-resource, or hospital-resource-dependent events. Results: Ten (of 14) PSIs were worse after the onset of COVID-19 and had substantial increases in event rates, ranging from 14.6% (postoperative sepsis) to 51.9% (central venous catheter-related bloodstream infections). Also, when a hospital had a high proportion of COVID-19 admissions, 8 PSIs significantly worsened and had substantial increases in event rates, ranging from 5.9% (postoperative pulmonary embolism/deep venous thrombosis) to 15.8% (postoperative sepsis). Conclusion: The COVID-19 pandemic placed immense, prolonged strain on hospitals, resulting in a worsening of inpatient patient safety events. Increased rates of safety events occurred whether events were provider-resource, surgical/team-resource, or hospital-resource-dependent. These findings are a call for hospitals to return to placing heightened emphasis on reducing preventable harm and minimizing hospital complications and adverse events.

Patient and caregiver perspectives are increasingly incorporated into health care research and policymaking, but their inclusion in the quality measure development process often is not robust. We describe a stakeholder panel model for incorporating patient/caregiver voices in the development of patient-reported measures, the Technical Expert/Clinical User/Patient Panel (TECUPP) model. This model is characterized by significant or equal representation of people with lived experience of the disease or condition (as patients or caregivers) to the clinicians and others with technical expertise who typically comprise technical expert panels. We report key design features of the TECUPP model and how we used this model to develop survey-based patient experience measures of timeliness of cancer diagnosis. We describe benefits and challenges of using the TECUPP model and considerations to guide others who might use it as part of developing patient-centered quality measures, based on our experience convening a TECUPP to inform development of a patient-reported measure on timeliness of cancer diagnosis. Benefits include creating space for significant contributions from patients/caregivers and development of a shared understanding of patient experiences and observability of measure domains between clinicians and patients/caregivers. Challenges include time management and managing conversations outside the project scope. Measure development efforts implementing this model should consider recruiting diverse individuals, scheduling short and frequent meetings, enabling participation from all TECUPP members, developing accessible pre-read materials, anchoring meetings with patient stories, and encouraging multiple communication modes. The TECUPP model promotes discussion and understanding by patients/caregivers and clinicians/measure experts helpful for development of survey-based patient-reported measures.

Introduction Studies on the impacts of COVID-19 on patient safety are emerging. However, few studies have elicited the perspectives of front-line clinicians. Methods We interviewed clinicians from 16 US hospitals who worked in the emergency department, intensive care unit or inpatient unit during the COVID-19 pandemic. We asked about their experiences with both clinician well-being and patient care throughout the pandemic. We used a rigorous thematic analysis to code the interview transcripts. This study was part of a larger randomised control trial of an intervention to improve healthcare worker well-being during the COVID-19 pandemic; the findings described here draw from clinicians who spontaneously raised issues related to patient safety. Results 11 physicians and 16 nurses in our sample raised issues related to patient safety. We identified two primary themes: (1) compromised access to healthcare and (2) impaired care delivery. First, clinicians discussed how changes in access to healthcare early in the pandemic–including a shift to telehealth and deferred care–led to delays in accurate diagnosis and patients presenting later in their disease course. Second, clinicians discussed the effects of COVID-19 on care delivery related to staffing, equipment shortages and space constraints and how they deviated from the standard of care to manage these constraints. Clinicians noted how these issues led to patient safety events such as central line infections, patient falls and serious medication administration errors. Conclusions Several well-intentioned interventions implemented in the early weeks of the pandemic created a unique context that affected patient safety throughout the pandemic. Future pandemic preparedness should consider planning that incorporates a patient safety lens to mitigate further harm from occurring during a public health crisis.

Background: Insurance status may influence quality of opioid analgesic (OA) prescribing among patients seen by the same clinician. Objective: To explore how high-risk OA prescribing varies by payer type among patients seeing the same prescriber and identify clinician characteristics associated with variable prescribing Design: Retrospective cohort study using the 2016–2018 IQVIA Real World Data – Longitudinal Prescription Participants: New OA treatment episodes for individuals ≥ 12 years, categorized by payer and prescriber. We created three dyads: prescribers with ≥ 10 commercial insurance episodes and ≥ 10 Medicaid episodes; ≥ 10 commercial insurance episodes and ≥ 10 self-pay episodes; and ≥ 10 Medicaid episodes and ≥ 10 self-pay episodes. Main Outcome(s) and Measure(s): Rates of high-risk episodes (initial opioid episodes with > 7-days’ supply or prescriptions with a morphine milliequivalent daily dose >90) and odds of being an unbalanced prescriber (prescribers with significantly higher percentage of high-risk episodes paid by one payer vs. the other payer) Key Results: There were 88,352 prescribers in the Medicaid/self-pay dyad, 172,392 in the Medicaid/commercial dyad, and 122,748 in the self-pay/commercial dyad. In the Medicaid/self-pay and the commercial-self-pay dyads, self-pay episodes had higher high-risk episode rates than Medicaid (16.1% and 18.4%) or commercial (22.7% vs. 22.4%). In the Medicaid/commercial dyad, Medicaid had higher high-risk episode rates (21.1% vs. 20.4%). The proportion of unbalanced prescribers was 11–12% across dyads. In adjusted analyses, surgeons and pain specialists were more likely to be unbalanced prescribers than adult primary care physicians (PCPs) in the Medicaid/self-paydyad (aOR 1.2, 95% CI 1.16–1.34 and aOR 1.2, 95% CI 1.03–1.34). For Medicaid/commercial and self-pay/commercial dyads, surgeons had lower odds of being unbalanced compared to PCPs (aOR 0.6, 95% CI 0.57–0.66 and aOR 0.6, 95% CI 0.61–0.68). Conclusions: Clinicians prescribe high-risk OAs differently based on insurance type. The relationship between insurance and opioid prescribing quality goes beyond where patients receive care.