Home Hospital—toward a Tighter Definition

To the Editor: We read with interest the study by Ricauda et al.1 describing a home hospital (HH) model of care for patients with acute stroke because it highlights important issues related to defining a HH. A wide variety of models of care has been described in the international literature under the HH umbrella, engendering controversy over the definition of HH and the effectiveness of the model.2,3 These models include the outpatient infusion center where patients receive an intravenous infusion or other treatment;4 the physicians' office intravenous service where patients predominantly self-infuse their medications and are reviewed in physicians' offices;5 the delivery of home-based care by hospital, or hospital contracted, staff or in the United States by a home health agency to patients in their own homes;6 early discharge schemes, mostly for postsurgical patients in which patients early in the postoperative period are discharged home and receive postoperative nursing supervision and skilled therapies at home, again with little organized input from physicians;7 and a clinical unit model that delivers acute hospital-level medical care in the patient's home and substitutes entirely for an acute hospital admission. Physician and nursing care are provided in the home, as well as appropriate diagnostic and therapeutic care and technologies.2,8–10 The study of HH for acute stroke by Ricauda et al. is an excellent example of a clinical unit model of HH. This HH model possessed three elements we consider to be key features that define a HH. (1) It provides care that substitutes entirely for an inpatient acute hospital admission. (2) It provides an intensity of care, including medical and nursing care, similar to that provided in the hospital appropriate to the severity of the illness treated. (3) It provides care that cannot be provided by usual community-based home care services.8–10 Defining HH in this manner relates to the underlying rationale of the model. The main reasons for developing HH include reducing iatrogenic complications including functional decline, honoring patients' wishes for their care, and reducing expenses. Given this, substitution for an acute hospital inpatient stay is essential in satisfying the HH care model to its fullest extent. Postdischarge programs and nursing programs that do not substitute for a hospital admission entirely expose the patient to the acute hospital environment, may not satisfy patient preference to avoid the hospital, and probably are not as economically efficient as possible. To substitute entirely for an inpatient admission, HH should provide an intensity of care similar to that provided in the hospital to ensure proper evaluation and treatment of the acutely ill patient. This means that 24-hour access to hospital-level medical and nursing care and appropriate diagnostic and therapeutic modalities must be available to the care of the patient. This has generally not been available in the community setting11 or in many models claiming to be HH. Provision of HH needs to be care that typical community-based home care services cannot provide. HH programs that become blurred with other community-based programs run a high risk of subsequent failure, simply because the skills and organization are different from those required for non-HH community-based care provision. In terms of economics, labeling typical community-based home care services as HH care may simply undermine the economic rationale for the model if expensive HH care is provided to patients who do not require it. Finally, in terms of research and evaluation, blurred boundaries between HH and more typical home care services may result in substantial heterogeneity of patients and care requirements and ensure that outcome studies trend towards equivalence. We hope that focusing the definition of HH will facilitate further development and research on this emerging model of care.