Analysis of hospital readmissions during pandemics

Hospitals overwhelmed during the COVID-19 pandemic are only now recovering from early, severe lack of ICU beds and respirators. New York's health care system neared capacity, even with quickly-erected temporary facilities.

In states like Louisiana and Alabama (with so many rural hospitals and few extra ICUs and ventilators), equipment needs are spiking. Capacities are strained.

Further, there is evidence COVID-19 can cause return hospital visits with patients worse off. If this return visit pattern becomes widespread, an expected "second surge" of COVID-19 cases this fall could quickly put renewed pressure on hospital resources.

An emergency room doctor at Boston Medical Center says COVID-19 positive-testing patients are coming back in an even "sicker" condition. Many returning patients need ventilators and ICU beds.

Dr. Jon Santiago told Boston station WBZ such return hospital trips can occur as early as a week after the initial COVID-19 diagnoses. Answers to questions about how often patients will need to be readmitted to hospitals will have a profound effect on planning for a possible second wave.

Hospitals need the data collection and analytical procedures to assess the potential impact of return visits before any second wave occurs.

This need for accurate data and forecasting tools should include the capacity to predict the level, sincerity, and virus characteristics of return patients.

Some approaches to analyzing this problem can be learned from an analysis of return visits during the past H1N1 pandemic. Drs. Osaro Mgbere and Salma Khuwaja studied return visits to a Houston, Texas multispecialty clinic between April 2009 and March 2010.

Their study may indicate the kind of data and analytic tools necessary to assess the impact of return visits in the COVID-19 pandemic.

During the 2009 flu pandemic (pH1N1), say the authors, the proportion of outpatient visits to emergency rooms, clinics, and hospitals understandably climbed. Particularly in the pandemic's early months, the facilities were swamped with visits by the actually sick, people worried about H1N1 but who were really well (the so-called "worried well"), and returning patients seeking more care. Return visits data:  Houston Health Department Influenza Sentinel Surveillance Project.

It comprised 2,680 individuals accounting for a total of 2,960 clinic visits. Overall, 4.56 percent of the cohort had return visits. Significant variations among returnees were attributed to age group (76 percent), the type of vaccine patients received (18.4 percent), and pH1N1 test results (5.6 percent).

The authors write, "Understanding the dynamics in care-seeking behavior during pandemics would assist policymakers with appropriate resource allocation and in the design of initiatives aimed at mitigating surges and recurrent utilization of the healthcare system."

References:

Osaro Mgbere and Salma Khuwaja, Model-Based Recursive Partitioning of Patients’ Return Visits to Multispecialty Clinic During the 2009 H1N1 Pandemic Influenza Pandemic (Ph1n1), The Online Journal of Public Health Informatics, Vol 12 No. 1 (2020).