Background: Acute heart failure exacerbations remain a leading cause of morbidity and mortality and are associated with high rates of hospital admissions and healthcare expenditure. As hospital medicine (HM) primary service footprint expands, further information regarding patient care provided by HM compared to cardiology (CV) is needed. As part of a broader effort to improve heart failure care at our hospital, we aim to investigate the association between admission services and key process and outcome measures of interest at Thomas Jefferson University (TJUH).
Methods: This is a retrospective chart review of 73 patients discharged in August 2024 from a Department of Medicine service at TJUH with heart failure requiring intravenous diuretics. Inclusion verification and data extraction was performed by manual chart review. Patients were divided by primary service of CV or HM. Expected mortality was obtained from Vizient. Readmission rate was obtained through automated export. The remaining data were obtained through chart review. Adequacy of diuresis was assessed by proportion of days with intravenous (IV) diuretic dose on which the patient lost >2 kg or was > net negative 2L and rate of patients with weight loss by discharge. Guideline-directed medical therapy (GDMT) prescribing at discharge was assessed by heart failure type and percentage of doses at target. Follow-up was assessed by presence and timing of primary care physician (PCP) and cardiology follow-up in our system. Analysis was performed in Excel.
Results: Of the 73 patients, 48 (66%) were cared for by CV and 25 (34%) were cared for by HM. CV/HM mortality rates were 6.3/4.0%; CV/HM observed to expected mortality (O/E) was 1.2/0.58. CV/HM 30-day index-hospital readmission rate was 13/20%. Regarding diuresis, CV attained goal diuresis on 29% of days patients received IV diuretics; HM rate was 38%. 30 of 48 (63%) of CV patients lost weight by discharge compared to 11 of 25 (44%) of HM patients. Of HFrEF/HFimpEF patients, at discharge, 6 of 30 (20%) of CV patients were on all four pillars at discharge compared to 1 of 10 (10%) for HM with CV averaging 2.2 GDMT meds per patient compared to HM’s 1.6. Percent of CV/HM GDMT at target at discharge follows: ARNi/ACEi/ARB (3%/10%), MRA (37%/30%), BB (10%/20%). For patients with HFpEF/MFmrEF, 5 of 17 (29%) of CV patients were discharged on an SGLT2 compared to 1 of 15 (7%) for HM. For follow-up, 29 of 48 (60%) of CV patients saw an outpatient cardiologist in our system within 30 days compared to 10 of 25 (40%) for HM patients. Mean time to cardiology follow-up was 20.1 and 18.8 days, for CV, HM, respectively though the 20.1 was influenced by a 109 day outlier compared to 61-day maximum for HM. Regarding PCP follow-up, 3 of 48 (6%) of CV patients saw a PCP within our system within 30 days of discharge as compared to 7 of 25 (28%). Median time to follow-up was 33.4 to 10.6 days for CV and HM, respectively.
Conclusions: This study highlights differences in care of patients with heart failure between primary cardiology and hospital medicine services at TJUH. Patients under cardiology care demonstrated higher rates of weight loss and greater utilization of GDMT at discharge, particularly among those with HFrEF/HFimpEF, with lower readmission rates. HM outperformed CV in O/E mortality, and PCP follow-up. Both services have significant opportunities in process and outcomes measures, and we hope sharing best-practice across services through systems improvement will minimize deleterious variation and improve patient outcomes.