Skip navigation

Hospital at Home: is it an effective approach?

 

A model of 'Hospital at Home' originated in France in the early 1960s and has since been adopted in various ways in a number of countries. It has become a popular response to the pressures on hospital beds. In the UK, the schemes concentrate on providing personal nurse-led care. Some schemes are designed to care for specific conditions but more commonly they are support early discharge following, for example orthopaedic surgery. But do patients admitted to such schemes have better (or equivalent) health outcomes compared with patients receiving in-patient hospital care? This question prompted a systematic review [1] to determine the effectiveness of managing patients in hospital at home schemes compared with in-patient hospital care.

Reference

Sasha Shepperd, Steve Iliffe. The effectiveness of hospital at home compared with in-patient hospital care: a systematic review. Journal of Public Health Medicine. 20: 344-350.

How was the question tackled?

A search strategy ensured that all relevant sources including Medline, Embase etc were examined. Links were also established with the UK Collaborative R&D 'Hospital at Home' Group to ensure that the search was comprehensive and for example could draw on unpublished studies and identify other relevant work. Criteria to guide the review focused on randomized controlled trials of 'hospital at home' schemes that admitted patients aged 18 or over and involved comparison between such schemes and hospital in patient care. The quality of eligible trials was assessed using the criteria developed by Cochrane Effective Practice and the Organisation of Care Group (EPOC).

What did the review find?

A comprehensive search identified eighty-three studies that potentially met the inclusion criteria. Following an assessment of the quality of these studies, seven trials were accepted for inclusion in the review and the investigators contacted to check whether further unpublished information was available. Subsequently two (of these seven) trials were excluded because they did not include compare home versus in-patient care.

Five completed trials met all the review criteria. Two trials had evaluated the effectiveness of hospital at home for elderly medical patients. Two trials were concerned with patients discharged early from hospital after elective surgery (these two trials were conducted in the 1970s and have limited relevance in the 21st century). The other trial focused on care of the terminally ill.

All of the trials had methodological limitations and combining the studies was judged inappropriate because of variations in study populations and interventions. Multiple outcomes were measured in all the trials but there was little uniformity in the choice of measures. A direct comparison of costs was not possible because of the different methods used by each study to calculate costs.

The review did, however, note that there are several randomized trials currently underway (in the UK and elsewhere) and these should add substantially to the body of knowledge about the effectiveness of hospital at home.

What has the review told us?

The review was inconclusive - largely because of the paucity of relevant trials. This situation could change when the trials currently in hand are completed and more information become available. At this stage (2000) the review did not provide evidence to support the widespread adoption of hospital at home - nor conversely argue for the discontinuance of existing schemes.

There was simply insufficient evidence to determine the effect of hospital at home on patient outcomes or on the cost of such schemes to the health service (none of the trials included a full economic analysis). There is however some evidence to suggest that patients like the concept - although this needs to be countered by some reservations from carers of patients having elective surgery. Carers of terminally ill patients however seemed to like the support at home - but this support waned if the patient survived for a long time.