Methods
The University of Missouri, Columbia, is a tertiary care academic health system with 5 hospitals and more than 60 clinic locations across central Missouri. In the spring of 2018, an MDT was developed, with support from administrative leaders, to improve the care of patients with IE at our institution. The work group prioritized one leverage point to improve IE outcomes, which was improving the number of surgeries performed on those IE patients who had guideline indications for surgery. A clinical pathway was developed around this leverage point (Figure 1). The pathway leveraged the 6 T’s (Table 1) to guide providers through the evaluation and management of IE.17 The pathway focused on improving adherence to standards of care and reduction in practice variation by defining indications for referrals and diagnostic interventions, helping to reduce delays in consultation and diagnosis. The pathway also clearly outlined the surgical indications and timing for patients with IE and provided the basis for decisions to proceed with surgery.
Starting in late 2018, in collaboration with cardiology and CTS teams, ID specialists socialized the clinical pathway to inpatient services that cared for patients with IE. Infectious diseases physicians also provided recurring conferences on the effectiveness of MDTs in IE management and participated in heart-valve team case discussions. Finally, in May 2019, an electronic version of the pathway was embedded in the EHR system using a Cerner PowerChart feature known as Care Pathways. The feature presents the user with algorithm questions in the EHR and provides recommendations, relevant orders, timelines, and other decision support in the clinical pathway. The feature is available to all providers in the health system.
To evaluate the effectiveness of our intervention, we recorded outcomes for patients with IE with surgical indications between December 2018 and June 2020 and compared them with our prior published data from January to December 2016. Cases of IE for the current study period were identified via retrospective chart review. Records from December 2018 to June 2020 were searched using International Statistical Classification of Diseases, Tenth Revision (ICD-10) discharge codes for IE (I33, I33.0, I33.9, I38, I39, M32.11). To select those patients with definitive IE and indications for surgery, the following criteria were applied: age ≥ 18 years; fulfilled modified Duke criteria for definite IE18; and met ≥ 1 American Heart Association (AHA)/Infection Diseases Society of America criteria for recommendation for surgery. Indications for surgery were ≥ 1 of the following: left-sided endocarditis caused by S aureus, fungal, or highly resistant organism; new heart block; annular or aortic abscess; persistent bacteremia or fever despite 5 days of appropriate antimicrobials; vegetation size ≥ 10 mm and evidence of embolic phenomena; recurrence of prosthetic valve infection; recurrent emboli and persistent vegetation despite antimicrobials; and increase in vegetation size despite antimicrobials.16
Age was treated as a categorical variable, using the age groups 18 to 44 years, 45 to 64 years, and 65 years and older. Gender was self-reported. Primary outcomes were surgery or transfer to a higher center for surgery and in-hospital death. Secondary outcomes included consults to teams involved in multidisciplinary care of patients with IE, including ID, cardiology, and CTS. Bivariate analyses were performed using Pearson χ2 tests. Odds ratios for surgery and death were calculated using a multivariate logistic regression model including age and gender covariates. Statistical significance was defined at α = 0.05, and statistical analysis was performed using Stata/IC v16.1 (StataCorp LLC). Our university institutional review board (IRB) reviewed the project (#2010858-QI) and determined that the project was quality-improvement activity, not human subject research, and therefore did not require additional IRB review.