Differences in length of stay by teaching team status in an academic medical center in the Midwestern United States

The landscape of inpatient admissions for alcoholic hepatitis in the era of early liver transplantation

INTRODUCTION AND OBJECTIVES

Liver transplant (LT) is a recent option available in the United States (US) to treat those with severe, refractory alcoholic hepatitis (AH). We examined changes in clinical characteristics of patients admitted with AH and tracked hospital outcomes as practice changes involving LT have shifted.

MATERIALS AND METHODS

Using the National Inpatient Sample, we performed a cross-sectional analysis of patients admitted with AH during the years 2016-2020 in the US. Differences in clinical characteristics over time were assessed. To compare outcomes between 2016-2017 (when LT was less common) and 2018-2020 (when LT was more common), we conducted linear and logistic regression. Propensity-score matching was used to compare outcomes between patients with and without LT.

RESULTS

From 2016-2017 to 2018-2020, patients admitted with AH tended to have a higher frequency of infection (p = 0.006), hepatorenal syndrome (<0.001), and ascites (<0.001). Hospital costs and length of stay (LOS) were highest in transplant hospitals, and costs rose over time in both non-transplant (NT) teaching and non-teaching hospitals (p < 0.001). Mortality decreased in NT teaching hospitals [aOR 0.7 (95% CI: 0.6-0.8)] and slightly decreased in NT non-teaching hospitals [aOR 0.7 (95% CI: 0.5-1.0)]. In the propensity-matched cohort involving LT versus non-LT patients, there was a 10% absolute reduction in-hospital mortality, but this came at a higher cost (p < 0.001) and length of stay (p < 0.001).

CONCLUSIONS

The severity of AH has been increasing over time, yet mortality has declined after adjusting for severity of disease. Patients who underwent LT survived; however, the healthcare burden of LT is substantial.

Physical Restraint Usage in Hospitals Across the United States: 2011-2019

Objective

To determine the change in rates of physical restraint (PR) use and associated outcomes among hospitalized adults.

Patients and Methods

Using national inpatient sample databases, we analyzed years 2011-2014 and 2016-2019 to determine trends of PR usage. We also compared the years 2011-2012 and 2018-2019 to investigate rates of PR use, in-hospital mortality, length of stay, and total hospital charges.

Results

There were 242,994,110 hospitalizations during the study period. 1,538,791 (0.63%) had coding to signify PRs, compared with 241,455,319 (99.3%), which did not. From 2011 to 2014, there was a significant increase in PR use (p-trend<.01) and a nonsignificant increase in PR rates from 2016-2019 (p-trend=.07). Over time, PR use increased (2011-2012: 0.52% vs 2018-2019: 0.73%; p<.01). Patients with PRs reported a higher adjusted odds for in-hospital mortality in 2011-2012 (adjusted odds ratio [aOR], 3.9; 95% CI, 3.7-4.2; p<.01) and 2018-2019 (aOR, 3.5; 95% CI, 3.4-3.7; p<.01). Length of stay was prolonged for patients with PRs in 2011-2012 (adjusted mean difference [aMD], 4.3 days; 95% CI, 4.1-4.5; p<.01) and even longer in 2018-2019 (aMD, 5.8 days; 95% CI, 5.6-6.0; p<.01). Total hospital charges were higher for patients with PRs in 2011-2012 (aMD, +$55,003; 95% CI, $49,309-$60,679; p<.01). Following adjustment for inflation, total charges remained higher for patients with PRs compared with those without PRs in 2018-2019 (aMD, +$70,018; 95% CI, $65,355-$74,680; p<.01).

Conclusion

Overall, PR rates did not decrease across the study period, suggesting that messaging and promulgating best practice guidelines have yet to translate into a substantive change in practice patterns.

Comparison of hospitalist service staffing models at Baylor University Medical Center

Background

Baylor University Medical Center benefits from being a quaternary care center with 900+ licensed beds and multiple different models to staff patients on the hospitalist service. These models include hospitalist only, resident teaching teams, and two different advanced practice practitioner teams. The primary goal of this study was to assess these different staffing models and to ascertain which model, if any, has better outcomes related to length of stay, total hospital charges, 30-day readmission rates, patient satisfaction, hospital-acquired infections, mortality, and early discharges.

Methods

The study was an observational retrospective chart review of all discharges from the hospitalist service at Baylor University Medical Center from October 1, 2021, to February 28, 2022. Patients were included if the hospitalist team was the primary team of record at the time of discharge. A total of 7803 patients were included.

Results

There was no difference in patient satisfaction, hospital-acquired infections, and mortality between the groups. The teaching teams had a shorter length of stay before the removal of outliers. Independent advanced practice practitioners reliably had more patients discharged before 11:30 am. Results support the concept of continuity of care, as well as lower patient-to-provider ratios.

Conclusions

These results have actionable implications that support the use of different advanced practice practitioner teams for the safe care of hospitalized patients as well as the safe integration of residents into patient care.