by Paul Govern
This spring in the U.S., there were widespread delays in elective health care procedures and screenings. Hospitals, in observance of federal guidelines, were, for a time, conserving beds and protective equipment in preparation for a surge in COVID-19 admissions. And, perhaps on a more prolonged basis, patients in many areas of the country stayed away due to anxiety over catching COVID-19 from other patients or their health care team.
In a report in the Journal of Biomedical Informatics, Wei-Qi Wei, MD, PhD, and colleagues analyze system-wide delays in elective services at Vanderbilt University Medical Center associated with the arrival of the COVID-19 pandemic in Tennessee.
The analysis shows that delays in inpatient cardiovascular procedures at VUMC were associated with slightly longer hospital stays and slightly higher rates of unexpected hospital readmissions for these patients.
Delays in cancer screenings were associated with more advanced cancer at the time of diagnosis, translating to increased odds of five-year mortality for some cancers, on the order of 1% to 8%.
For other hospitals that might be interested, an electronic health record system paired with a research data warehouse provides all the means necessary for replicating the analysis performed at VUMC.
“In the course of this study we’ve developed a high-throughput pipeline that can be easily implemented by other hospitals to find any local patterns of delay in care, along with any adverse consequences,” said Wei, assistant professor of Biomedical Informatics. “Results of such analyses can be used by hospitals and health systems to prioritize procedures and screenings for rescheduling, hopefully minimizing the impact of delays on health outcomes.”
By comparing patient volumes from the period when VUMC was obligated to suspend elective procedures, that is, mid-March to mid-April 2020, to volumes for the same period from 2018 and 2019, the study produced a comprehensive list of procedures and screenings at VUMC for which COVID-19 appears to have precipitated delays: in all, more than 400 different types of procedures were delayed. Narrowing in on cardiovascular and cancer-related diagnoses and procedures, the study demonstrates methods for measuring the effects of delays.
Delays for surgical procedures are estimated based on elapsed time between diagnosis and procedure, compared to historical trends. Here the outcomes of interest in the report include hospital length of stay, subsequent unexpected hospital readmission or emergency room visits, and mortality during hospitalization. (No associations with in-hospital mortality were found.)
Delays in cancer screenings are estimated based on elapsed time between diagnosis and the most recent screening, compared to historical trends. Here the matter of interest is cancer stage at diagnosis and the consequent odds of five-year survival.
Automated analysis of electronic health records was followed by a review by three clinicians, working independently, to evaluate the plausibility of any associations. In all, the analysis found 24 plausible associations between delays and adverse outcomes, including associations involving 19 inpatient cardiovascular procedures and five cancer screenings.
The first author of the report is Neil Zheng, a former term employee at VUMC who’s now studying medicine at Yale University. Others joining Wei for the report include Jeremy Warner, MD, MS, Travis Osterman, DO, MS, Quinn Wells, MD, PharmD, MS, MSCI, Xiao-Ou Shu, MD, PhD, Stephen Deppen, PhD, MS, MA, Seth Karp, MD, MA, Shon Dwyer, MBA, RN, QiPing Feng, PhD, Nancy Cox, PhD, Josh Peterson, MD, MPH, C. Michael Stein, MBChB, Dan Roden, MD, and Kevin Johnson, MD, MS.
The study was supported by the National Institutes of Health (GM115305, HG011166).