Stroke Rehabilitation at Home

Abstract

T he modern hospital evolved from a place to care for the sick staffed by members of religious orders to a symbol of rationality and progress fostering medical innovation, professional development, science, research, and training. 1 As hospitalization grew to be the norm for seriously ill patients, there was also the recognition that not all care had to be insti-tutionalized, and in the 1960s, a practice of offering hospital care at home 2 for the terminally or chronically ill emerged. By the 1990s, the hospital-at-home became an attractive option in response to demand for acute-care hospital beds. 3 In 1997, hospital-at-home was tested for acute stroke patients in Italy. 4 Published in 2004, 120 patients were randomized from the emergency department to be managed at home or to be managed in the hospital as usual. The results showed that functional and neurological outcomes improved similarly in both groups, but patients managed at home had lower depression scores, fewer complications, and were more likely still be at home at 6 months. With the development of effective therapies for acute stroke, hospital admission was considered best practice and stroke unit care was shown to be superior to other models of care. 5,6 The hospital-at-home approach changed from a focus on avoiding hospitalization to a focus on early discharge from acute care but with support for ongoing recovery by providing rehabilitation and other services in a community setting. 7 Home rehabilitation for stroke can now be considered under 3 broad rubrics: (1) rehabilitation at home to replace acute care-the early supported discharge (ESD) model; (2) rehabilitation at home to replace institutional rehabilitation; and (3) home exercise to prevent deterioration and promote health through physical activity. The aim of this review is to summarize what lessons have been learned from the many well-designed clinical trials evaluating the effect of providing ≥1 aspects of stroke rehabilitation in the home and identify promising avenues for implementation so that the greatest good can be achieved for the greatest number of people at the least cost. The studies that have been done are heterogeneous as to purpose, population, timing from stroke, nature of the interventions , and the type of control group. This heterogeneity provides rich learning material. Early Supported Discharge The evidence for ESD has been systematically reviewed. The meta-analysis of individual patients' data 7 from 11 trials involving 1597 patients summarized in the Figure found a reduced risk of death or dependency for the ESD group in comparison to the usual care group (summary odds ratio, 0.79; 95% confidence interval [CI], 0.64-0.97), shortened length of hospital stay by an average of 8 days (95% CI, −4 to −11 days), and showing strongly favorable effects on extended activities of daily living (odds ratio, 0.12; 95% CI, 0.0-0.25). Table 1 summarizes the results across the different models of ESD presented by Langhorne et al. 7 The effect was the greatest when the ESD was provided by a coordinated multidisciplinary team and for stroke patients with mild to moderate disability. One of the striking features of these trials is that less than half (median, 41%) of patients with stroke were eligible for ESD (range, 13%-68%) 8 because they were ill, discharge home was not realistic because of the lack of a caregiver, or the stroke was not disabling enough. The implication is that if ESD is implemented as a policy, similar eligibility criteria as the trials would need to be applied if the same benefit is to be observed. However, implementing an ESD program to a proportion of people with low disability would not necessarily be a bad thing as this group has many physical, emotional, cog-nitive, and participation consequences that have a negative effect on quality of life. 9 These difficulties are often unrecognized during hospitalization and may only become evident after returning home. Whether and what kind of intervention people with mild stroke need is not fully understood as the trials of ESD did not provide subgroup analyses. A recent trial providing telephone support post discharge for people with mild stroke 10 revealed that few availed themselves of this support service on their own and even when offered directly, there was no effect on outcomes. A more active ESD for people with low disability may be a way forward. One way of identifying the full effect of adopting a policy of ESD is to use an outcome measure that can be linked to costs. These measures fall under the rubric of utility measures, 11 which are designed to create a single value across different