We included studies where the primary aim was to examine aspiration and mean age was greater or equal to 65 years. Studies were excluded if more than half the study population had acute stroke or Parkinson's disease; studies of diagnostic accuracy had to include a gold standard test. Two authors independently searched databases from 1966-July 2011 (Medline, CINAHL, Embase, Cochrane Library, Controlled Clinical Trials).
The search strategy identified 1962 reports for scrutiny. Fourteen studies were eligible for inclusion, ten of which were bedside screening studies. Heterogeneity of methods and outcomes precluded meta-analysis. Screening tests examined included water swallow tests, cough test, risk factor checklist, pulse oximetry, auscultation tests, and specific clinical features. Sensitivity of tests ranged from 0%to 100%and specificity 29%to 100%. Only one study was performed on an unselected population of hospitalised older patients; other studies were performed in highly selected patient cohorts.
Diagnostic accuracy of bedside swallow tests in this patient group remains unclear and the predictive value of bedside tests for pneumonia in this patient group remains unknown. Existing evidence is insufficient to support the use of these bedside tests in a general older population.