A prospective study of symptoms, function, and medication use during acute illness in nursing home residents: design,...

Nursing home residents are at high risk for developing acute illnesses. Compared with community dwelling adults, nursing home residents are often more frail, prone to multiple medical problems and symptoms, and are at higher risk for adverse outcomes from acute illnesses. In addition, because of polypharmacy and the high burden of chronic disease, nursing home residents are particularly vulnerable to disruptions in transitions of care such as medication interruptions in the setting of acute illness. In order to better estimate the effect of acute illness on nursing home residents, we have initiated a prospective cohort which will allow us to observe patterns of acute illnesses and the consequence of acute illnesses, including symptoms and function, among nursing home residents. We also aim to examine the patterns of medication interruption, and identify patient, provider and environmental factors that influence continuity of medication prescribing at different points of care transition.

This is a prospective cohort of nursing home residents residing in two nursing homes in a metropolitan area. Baseline characteristics including age, gender, race, and comorbid conditions are recorded. Participants are followed longitudinally for a planned period of 3 years. We record acute illness incidence and characteristics, and measure symptoms including depression, pain, withdrawal symptoms, and function using standardized scales.

76 nursing home residents have been followed for a median of 666 days to date. At baseline, mean age of residents was 74.4 (+/- 11.9); 32% were female; 59% were white. The most common chronic conditions were dementia (41%), depression (38%), congestive heart failure (25%) and chronic obstructive lung disease (27%). Mean pain score was 4.7 (+/- 3.6) on a scale of 0 to 10; Geriatric Depression Scale (GDS-15) score was 5.2 (+/- 4.4). During follow up, 138 acute illness episodes were identified, for an incidence of 1.5 (SD 2.0) episodes per resident per year; 74% were managed in the nursing home and 26% managed in the acute care setting.

In this report, we describe the conceptual model and methods of designing a longitudinal cohort to measure acute illness patterns and symptoms among nursing home residents, and describe the characteristics of our cohort at baseline. In our planned analysis, we will further estimate the effect of the use and interruption of medications on withdrawal and relapse symptoms and illness outcomes.