Development of a comprehensive early monitoring platform will require a range of key data sets. Variables required include three general categories of data needed: prehospital/community, public health mandates that affect, demographic distribution (age, SES, ethnicity), vaccination rates, test positive rates, search term rates, case fatality rates, access to care, temporal behavioral changes (e.g. quarantine/isolation, masking, etc.); the number of hospitals, number of EDs, inpatient, and ICU beds per hospital, catchment area and out of area transfer rates, hospitalization rate, admission rates per ED volume, ICU admission rates, nursing staffing ratios, physician staffing ratios, staffing availability (attrition, burnout, quarantined/isolated), surgical volume and average length of stay (LOS), elective case volume, emergent case volume and LOS), emergent admission and mortality rates (stroke, heard attach, sepsis, liver failure, gastrointestinal bleeding, etc.); mortality and length of stay for disease of interest, ventilator and noninvasive respiratory support availability; background rates of hospitalizations for seasonal diseases (e.g., influenza), weather and climate related changes to utilization rates (e.g. extreme heat, drought, etc.); and post hospital- post acute hospitalization availability, length of stay, readmission rates, and utilization.
Developing A Community-Wide Early Disease Surveillance Monitoring System
Need
Early detection of disease outbreaks is critical to the effective management of system resources within any community, especially during pandemics. Due to the current fragmentation of timely informational resources to initially identify outbreaks, a comprehensive digital platform for early surveillance monitoring within communities is required to rapidly inform healthcare, community leaders, policy makers and EMS systems. The model will learn from experience and will be able to simulate the effect of a change in any variable (e.g., community behavior or hospital response such as cancelling elective surgeries, expanded ICU, or alternate staffing models) on the outcomes of interest. The simulation can be developed into unique digital platforms for early surveillance monitoring within communities to inform healthcare, community leaders, policymakers, and EMS systems.
Proposed Solution
The model will learn from experience and will be able to simulate the effect of a change in any variable (e.g, community behavior or hospital response such as cancelling elective surgeries, expanded ICU, or alternate staffing models) on the outcomes of interest. Further, this information can be developed into a dynamic, 24x7, on-line digital platform which can be utilized and monitored by key users of any local community health ecosystem to better manage early responses to disease outbreaks and pandemics.