952. Validating a Novel Framework to Classify Inpatient Infectious Diseases Consultation Requests
Abstract Background Consistent classification of consult requests may lead to more efficient and collegial conversations about patient care, which could improve work satisfaction and enhance the learning environment. The authors propose a framework of 7 consultation types (Table 1). We aimed to obtain validity evidence for this rubric to consistently classify consultation requests. Table 1. Framework for classifying consults into 7 types Methods A randomly selected sample of 100 de-identified infectious diseases (ID) consult requests from a single academic center were independently coded as 1 of the 7 consultation types by 3 ID specialists and 3 hospitalists. Perfect concordance (6/6 coders) and partial concordance (4/6 or 5/6 coders) was calculated. Total (3/3 coders) and partial (2/3 coders) concordance based on consult subtypes and provider specialty was also calculated. We compared proportions between groups using a chi square test. Results Perfect concordance was 30%, and partial concordance was 60% (Figure 1). Total concordance among ID specialists was 44% and among hospitalists was 54% (Table 2). In cases without perfect concordance (n=70), ID specialists had 20% total concordance and 70% partial concordance, while hospitalists had 34% total concordance and 59% partial concordance. ID specialists were less likely than hospitalists to have perfect concordance for ideal consults (52% vs 73%, P=0.01). ID specialists and hospitalists were similarly likely to classify a consult as ideal (65% vs 69%, P=0.34), but ID specialists were more likely to classify a consult as S.O.S. (25% vs. 17%, p=0.02), and less likely to classify a consult as confirmatory (3% vs 7%, P=0.02) (Table 3). Figure 1. Concordance overall by consensus consult type among hospitalists and infectious disease specialists Table 2. Concordance by consult type stratified among infectious disease specialists and hospitalists Table 3. Consult type by physician subspecialty, among infectious disease specialists and hospitalists Conclusion ID consults can be classified into a novel rubric of 7 subtypes. Overall, partial or perfect concordance between hospitalists and ID consultants was 90%. ID specialists were more likely to classify consult requests as S.O.S than hospitalists, and hospitalists were more likely to classify consults as confirmatory. Opportunities exist to utilize the rubric to improve provider communication and interprofessional education. Disclosures All Authors: No reported disclosures
