Telehealth Factors for Predicting Hospital Length of Stay.

Identifying older adults with heart failure at risk for hospital readmission is challenging, and risk prediction models may be improved with inclusion of telehealth factors. In the current study, demographic, clinical, telehealth, and use data for emergency department (ED) presentations, hospitalizations, and length of stay (LOS) were collected from the records of 187 Veterans with heart failure participating in a 90-day Care Coordination Home Telehealth program between September 2007 and September 2013. Heart failure-related ED visits were 17.6% and 18.2% required hospitalization with an average LOS of 7 days (range = 1 to 38 days). Binary logistic regression models failed to predict likelihood of an ED presentation or hospitalization. Poisson regression models significantly predicted hospital LOS on the factors of telehealth alerts, nurse response to alerts, advancing age, and chronic renal disease. Data collected from one telehealth program significantly contributed to heart failure-related risk prediction models and should be included in future models. [Journal of Gerontological Nursing, xx(x), xx-xx.].