This study proposes to expand the traditional role of the home health aide (HHA) by training them to function as disease management coaches (DMCs) based on the model of community health worker (CHW) programs for the chronically ill. HHA DMCs will be trained to implement congestive heart failure (CHF) and diabetes disease management protocols via CHW curricula specially modified to reinforce protocol implementation. Two major hypotheses will be tested. First, home health care (HHC) patients with primary diagnoses of CHF or diabetes who are assigned to HHA DMCs will have fewer hospitalizations and ER visits, cheaper HHC episodes, and greater satisfaction with care than patients assigned to traditional HHAs or to no HHA. Second, HHA DMCs will have greater job satisfaction than traditional HHAs A 48-month randomized HHA DMC intervention trial with 2 arms, a 36-month patient accrual period, 60 day follow-up from the date of the HHC index admission, and random assignment of 14 of 28 HHAs to be trained as HHA DMCs, will be implemented in the Washington, DC, Bethesda and Baltimore MD sites of MedStar , Health VNA. In arm 1, all CHF (N=300) and diabetes (N=210) patients admitted who normally would get an HHA will be randomized equally to a traditional HHA or an HHA DMC.In arm 2, a randomly selected subset of Medicare patients (600 CHF, 600 diabetes) ordinarily not assigned HHAs will be randomized equally to an HHA DMC or no HHA. Patient baseline/outcome data will be obtained via OASIS, a standardized HHC clinical/demographic data set collected routinely for HHC patients, other agency clinical, administrative, and claims data, and a telephone survey to be crafted to collect satisfaction and self report hospital/ ER use data. HHA job satisfaction, and training/performance evaluation data will be captured by tools to be created This study will contribute to efforts to contain HHC costs, improve care quality and reduce HHC paraprofessional shortages. First, it will test a new HHC delivery model. If HHA DIVICS improve patient outcomes, decrease care costs, and are more satisfied with their jobs, the model holds promise if future studies support generalizability. Second, a unique curriculum for training HHAs as DMCs will be produced which incorporates for the first time curricula used to train CHWs. Evaluation of training and performance of the HHA DMCs will provide insight into how to cross-train paraprofessionals for effective and satisfying enhanced practice roles.