Timely antibiotic administration is crucial for improving sepsis outcomes, but delays are common and multifactorial. Fever, a key feature of sepsis, is often treated with paracetamol, which may mask the condition and contribute to delays in time to antibiotic. This study investigated the association between paracetamol and delayed antibiotic administration in hospitalized septic patients. This was a retrospective study using data from a prospectively collected sepsis registry. The study was conducted at Songklanagarind Hospital, a quaternary care university-affiliated hospital in southern Thailand. Adult patients (≥15 years) admitted for ≥24 h who developed suspected sepsis between March 2023 and March 2024. Suspected sepsis was defined by a high National Early Warning Score (NEWS). We recorded paracetamol administration within 8 hours prior to the sepsis trigger (time zero). The primary outcome was the delay of ≥180 min from time zero to antibiotic administration. Logistic regression was used to assess the associations, adjusting for confounders. Among the 1634 sepsis events, 12.5% involved prior paracetamol use. Paracetamol administration was not associated with increased odds of delayed antibiotic treatment (adjusted odds ratio 1.29, 95% confidence interval 0.94-1.78, p = .12). It also did not affect the temperature trajectories. Higher temperature was associated with lower in-hospital mortality in patients who did not receive paracetamol but this association was reversed in those receiving paracetamol. In this study, paracetamol administration before a sepsis alert was not associated with a delay in antibiotic administration and did not affect temperature changes.

Modern healthcare environments increasingly limit opportunities for physicians to build collegial relationships, contributing to feelings of isolation and disconnection. Commensality has been proposed as a way to restore relational infrastructure within clinical settings. While prior commensality programs have shown modest quantitative effects on burnout, less is known about how physicians experience these gatherings and the cultural functions they serve. To explore how physicians participating in a hospital-based commensality program perceived its impact on community, connection, and culture. We conducted a qualitative study of two cohorts of attending physicians participating in a 12-month hospital-based commensality program. Open-ended survey responses were analyzed using inductive thematic coding followed by an interpretive cultural analysis to situate emergent themes within broader frameworks of medical culture. Fifty-six physicians participated across two program years. Six themes reflected patterns of connection and constraint: creating community; shared experiences as support; learning new perspectives; bridging professional boundaries; professional growth and mentorship; and structural barriers to connection. Average attendance was 75.5% in 2023-2024 and 59.7% in 2024-2025. Physicians described commensality as restorative and humanizing while noting time pressure and workload as key barriers. Physician commensality facilitated meaningful connection, cross-departmental relationships, and informal mentorship within the clinical workplace. These findings suggest that shared meals can support relational aspects of professional life that traditional institutional structures may erode.

“It is surprising that a behavior as subtle and potentially meaningless as one's speech style should affect something as consequential as one's ability to get promoted or be respected.”1 - Deborah Tannen, Talking from 9 to 5: Women and Men at Work Most leaders are taught that effective communication requires confident, decisive (agentic) language, often accompanied by strong body language. This advice is particularly emphasized for women in leadership, who are often assumed to be skilled in displaying warmth and likability, and less so in projecting power. Does this conventional wisdom apply in all circumstances? In a landscape that has become increasingly team-based, healthcare leaders must shift from traditional hierarchical leadership models to collaborative approaches.2 Such a shift calls attention to the difference between power and status. Power is the ability to grant or withhold tangible resources. Many physician leaders, however, lack power in this sense and must influence those they lead via status—a term social scientists use to encompass the respect and regard an individual holds in the eyes of other group members. Unlike power, status cannot be demanded from others: one can have only as much status as others are willing to give. Power and status may coexist, though the holding of power does not automatically confer status, and a person may possess status without power. Despite business literature generally promoting agentic leadership, other research highlights the subtle utility of “power-less” (communal) speech.3 Power-less speech softens one's assertions and orients toward relationship and collaboration. It is characterized by using qualifiers (e.g. “kind of,” “sort of”), adding tag questions (e.g. “don't you think?”), and accompanying statements with disclaimers (“I'm not sure if this will work, but.”). This technique elevates the input of others by conveying that the speaker does not have all the answers. Power-less speech is most useful in settings where tasks and teams are interdependent since groups value traits that prioritize communal aims. Asking open-ended questions builds social capital by empowering the listeners in their own goals and can motivate a team around a shared vision. The complex challenges in healthcare require leaders to shift between communal and agentic language, with skilled leaders recognizing and adapting their communication methods to suit the needs at hand and foster positive change. For example, agentic language may help a division leader cast a leadership vision and strategy for their team. On the other hand, a medical director seeking to improve a quality measure will do well to use communal language to foster dialogue and solicit solutions from all team members. Language has power. It shapes our social realities and relationships. Regardless of gender, minor language adjustments allow leaders to build influence in dynamic, context-dependent work ecosystems. Agentic or powerful language is not inherently bad. In fact, for leaders with minoritized identities, powerful and agentic language is frequently necessary. But consider using power-less language as a status-building approach in situations of high interdependence. Leaders can use language in a manner that is fit for purpose rather than always promoting agentic and powerful language in every situation. The authors would like to thank Dr. Ben Kinnear for his contributions to this piece and his always wonderful sponsorship of others. The authors declare no conflicts of interest.

Prolonged boarding of older adults with neuropsychiatric illness has become a common, yet underrecognized, feature of contemporary mental health care delivery. I highlight the ways boarding functions as a de facto stage of care that impacts the system and hospitalists tasked with caring for these patients. Boarding reflects structural vulnerabilities around psychiatric capacity, long‑term care, and community services. Boarding time should be treated as a reportable safety metric, paired with unit‑level prevention and escalation pathways to guide hospital leaders, hospitalists, and regional planners.

Healthcare workers face escalating violence despite significant security investments, suggesting current approaches miss fundamental causes. We argue that most healthcare violence stems not from individual pathology but from systemic failures-financial barriers, insurance denials, access delays, and administrative complexity-that create volatile patient-provider interactions. Healthcare workers become targets for anger about institutional dysfunctions they cannot control. Current prevention strategies emphasizing individual risk assessment and physical security fail to address these root causes. We propose a research agenda examining connections between system failures and violence, potentially identifying upstream interventions that complement existing security measures while targeting the healthcare dysfunctions driving this epidemic.

Despite a decline in hepatitis A virus (HAV) incidence following vaccine introduction, HAV remains a public health concern in the United States. Recent multi-state outbreaks have demonstrated that HAV can re-emerge and highlight inequities in vaccine coverage or access, outbreak response, and in those with social or health risk factors. Continued investment in prevention, particularly vaccination and surveillance, is essential to prevent resurgence. Hospitalization offers a critical opportunity to address these gaps and provide equitable protection for at-risk populations.

The incarcerated population in the United States is underserved and aging rapidly; there is a dearth of information regarding their health, including hospital care. Epidemiological information is crucial to guide policymakers' planning. This analysis provides comparison data on the admissions of incarcerated and nonincarcerated patients. Identify differences in inpatient care between incarcerated and nonincarcerated patients by comparing 10 years of data at an academic medical center that serves as the largest single provider of hospital care for the state's Department of Corrections. Ten-year cohort study comparing all admissions of incarcerated males (n = 4525) to a matched cohort of nonincarcerated (n = 13,575) at an academic medical center serving as the largest provider of hospital care for the state's Department of Corrections. We included patients ≥18 years old and admitted to the hospital while in custody of the Department of Corrections between January 1, 2010, and December 31, 2019. Primary outcomes were patient characteristics, including demographics and comorbidities, and hospitalization characteristics, including length of stay and diagnosis. Compared with the nonincarcerated cohort, the incarcerated patients were more likely to be Black (33.4% vs. 6.6%, p < 0.05), Hispanic/Latine (6.1% vs. 2.1%, p < 0.001), admitted to a general medicine service (33.4% vs. 22.6%, p < 0.01), and less likely to be admitted to neurological/psychiatric services (5.2% vs. 9.1%, p < 0.01). The incarcerated cohort had a higher mean Charlson Comorbidity Index (3.83 vs. 3.65, p < 0.001), with statistically significant differences observed among patients aged 40-59. In this cohort study, there were differences in patient and hospitalization characteristics between incarcerated and nonincarcerated patients that may have clinical implications. Policymakers and researchers must work toward improving both the health and health care of this marginalized population.