Using a large, geographically diverse, hospital-based database in the United States (Premier PINC AI Healthcare Database), we aimed to describe the proportion and characteristics of patients receiving phenotype-desirable antimicrobial therapy (PDAT) among those hospitalized with Enterobacterales bloodstream infections.

Adult patients with an admission between January 1, 2017 and June 30, 2022 with ≥1 blood culture positive for Escherichia coli, Klebsiella oxytoca, Klebsiella pneumoniae, or Proteus mirabilis and receiving an empiric antibiotic therapy on blood culture collection (BCC) Days 0 or 1 were included. Receiving PDAT (defined as receipt of any antimicrobial categorized as “desirable” for the respective phenotype) on BCC Days 0−2 was defined as receiving early PDAT.

Among 35,880 eligible patients, the proportion of patients receiving PDAT increased (from 6.8% to 22.8%) from BCC Day 0−4. Patients who received PDAT (8,193, 22.8%) were more likely to visit large (500 + beds, 36% vs 31%), teaching (45% vs 39%), and urban (85% vs 82%) hospitals in the Northeast (22% vs 13%) compared to patients not receiving PDAT (all P <. 01). Among patients receiving PDAT, 61.4% (n = 5,033) received it early; they had a lower mean comorbidity score (3.2 vs 3.6), were less likely to have severe or extreme severity of illness (71% vs 79%), and were less likely to have a pathogen susceptible to narrow-spectrum β-lactams (31% vs 71%) compared to patients in the delayed PDAT group (all P < .01).

The proportion of patients receiving desirable therapy increased between BCC Day 0 and 4. Receipts of PDAT and early PDAT were associated with hospital, clinical, and pathogen characteristics.

Due to the unpredictable dementia trajectory, it is challenging to recognize illness progression and the appropriateness of a palliative approach. Further confusion occurs during hospitalization where the presence of comorbid conditions complicates prognostication. This research examined clinicians and families' perceptions of dementia as a terminal condition in relation to end-of-life admissions.

The study was based in the General Medicine units of one Australian public hospital. Medical, nursing, and social work clinicians were recruited to reflect multidisciplinary perspectives. Bereaved caregivers of deceased patients with dementia were interviewed 3 months following death.

Qualitative research underpinned by a social constructionist epistemology and framed through complex systems theory. Semi-structured interviews generated data that illuminated perceptions of deterioration observed toward the end of life.

Although participants anticipated general cognitive and physical deterioration associated with dementia, the emergence of comorbid illness made it difficult to predict the onset of the end of life. During a hospital admission, clinicians attributed the end of life to the advanced outcomes of dementia, whereas families described new medical crises. End-of-life admissions illuminated intersections between dementia and comorbidities rather than illness progression. In contrast with the perception that people with dementia lose awareness at the end of life, families drew attention to evidence that their loved one was present during the dying phase.

Our findings challenge the dominant understanding of dementia trajectories. Bifurcations between clinicians and families' views demonstrate the difficulties in recognizing end-of-life transitions. Implications for the integration of palliative care are considered.