You are here
Inactive Project: Nursing Home Screener
LHNCBC is no longer conducting active research on this project. Information is presented here for historical purposes.
CEB is developing a “Web 2.0” interface to information about American nursing homes. The goal is to offer the public a timely and easy to use site for the rapid location and comparison of nursing homes, thus identifying those most worth further review or a personal visit. Use by healthcare professionals involved in nursing home placement is also envisioned.
Currently in prototype form, our “Nursing Home Screener” locates homes on a Google Map (see figure). It allows nursing home quality, indicated by map icons, to be surveyed in any of four major categories, based on information provided by the Centers for Medicare and Medicaid Services (CMS):
Healthcare Deficiencies and Fire Safety Deficiencies, detected during CMS-mandated annual state inspections of nursing homes. Each specific deficiency is reported with levels of severity and extent of residents affected. NHScreener converts these to a score that can be percentile ranked.
Staffing Level of Nurses and Aides, that is, hours per day of nurses and certified nursing assistants per resident. NHScreener compares these values to minimum standards set by the states.
Care Quality Inferred from Residents’ Health. The staff of a home must fill out a quarterly resident assessment survey, from which CMS derives certain quality of care metrics for the home as a whole. NHScreener combines and ranks these further.
Within each category, options can be tailored to user preferences. Furthermore, home attributes (e.g., number of beds, type of home ownership) can be used to selectively hide home markers of less interest.
Technologically, the prototype currently employs a Microsoft IIS 6 web server with ASP 2 architecture. Monthly data from CMS is downloaded as an Access database, transformed (by data type conversion, data pivoting, statistics calculation, and ranking), augmented with latitude and longitude coordinates derived from street addresses, and delivered via a MySQL database.