State of the art in geriatric rehabilitation. Part II: clinical challenges

To examine common clinical problems in geriatric rehabilitation and to make recommendations for current practice based on evidence from the literature.

A CINAHL database and 2 MEDLINE searches were conducted for 1980 to 2001. A fourth search was completed by using the Cochrane database.

One author reviewed the references for relevance and another for quality. A total of 336 articles were considered relevant. Excluded articles were unrelated to geriatric rehabilitation or were anecdotal or descriptive reports on a small number of patients.

The following areas were the major geriatric rehabilitation subtopics identified in the search: frailty, comprehensive geriatric assessment, admission screening, assessment tools, interdisciplinary teams, hip fracture, stroke, nutrition, dementia, and depression. This article focuses on the latter 5 subtopics. The literature was reviewed by using a level-of-evidence framework. Level 1 evidence was a randomized controlled trial (RCT) or meta-analysis or systematic review of RCTs. Level 2 evidence included controlled trials without randomization, cohort, or case-control studies. Level 3 evidence involved consensus statements from experts, descriptive studies, or reports of expert committees.

Of the 336 articles evaluated, 108 were level 1, 39 were level 2, and 189 were level 3. Recommendations were made for each subtopic according to the level of evidence in the specific area. In cases in which several articles were written on a topic with similar conclusions, we selected the articles with the strongest level of evidence, thereby reducing the total number of references.

Frail older patients with hip fracture should receive geriatric rehabilitation. They should also be screened for nutrition, cognition, and depression. Older persons should receive nutritional supplementation when malnourished. If severe dysphagia occurs in stroke patients, gastrostomy tube feeding is superior to nasogastric tube feeding.