Investigating Nursing Home Emergency Preparedness for Environmental and Climatological Determinants of Resident Health

PROJECT SUMMARY The medical and functional needs of nursing home residents impose a differential risk of morbidity and mortality due to severe weather events, including hurricanes. A potential explanation for the increases observed in adverse resident outcomes after hurricane exposure is inadequate compliance by nursing home

facilities to federal emergency preparedness standards; however, prior research has not rigorously evaluated this association. Furthermore, nursing home characteristics, such as private versus not-for-profit or public ownership, are known to affect resident outcomes, but the impact of ownership on emergency preparedness

and post-disaster outcomes is unknown. To address these knowledge gaps, we propose a retrospective cohort study using data from the Centers for Medicare and Medicaid Services (CMS) and National Oceanic and Atmospheric Institute (NOAA) to evaluate facility-level and resident outcomes as a function of exposure to

Hurricane Michael (2018) and administrative emergency preparedness. Within our analytic sample of 1,423 nursing homes, we have classified 317 nursing homes as exposed to Hurricane Michael. Evaluating the post- disaster outcomes of 30-day all-cause mortality (primary), 30-day hospitalization, and 120-day functional

decline, we will use predictive modeling to identify which of the 250 federal emergency preparedness deficiencies identified in CMS Life Safety Code inspections are most predictive of the facility-level rate of adverse post-disaster outcomes (Aim 1). We hypothesize that deficiencies pertaining to evacuation and

sheltering preparedness will be included within this subset of “critical deficiencies.” In Aims 2 and 3, we will construct a dichotomous measure of noncompliance from the subset of critical deficiencies identified in Aim 1 (i.e., a nursing home with ≥1 critical deficiencies will be classified as noncompliant). In Aim 2, we will evaluate

whether the impact of hurricane exposure on the resident-level likelihood of adverse post-disaster outcomes is modified by noncompliance. We hypothesize that noncompliance will confer an increased risk of adverse post- disaster resident outcomes. Lastly, we will assess whether private nursing home ownership is associated with

the resident-level likelihood of adverse post-disaster outcomes, and whether this potential association is mediated by noncompliance (Aim 3). We hypothesize that residents from privately owned nursing homes will have worse post-disaster outcomes, and that these associations will be mediated by noncompliance. Our

integration of geospatial data with administrative health and compliance data positions us to provide the first empirical assessment of the relative importance of individual federal emergency preparedness standards and the mechanisms by which they affect nursing home resident outcomes. The proposed research responds to

federal priorities regarding the need to improve emergency preparedness for an aging population, and will identify modifiable factors that can be intervened upon by nursing home staff and regulators to improve resident outcomes after exposure to severe weather events.