2014 Volume 14 Issue 4 Pages 171-178
Falls in hospitals are common events that cause considerable morbidity. Though studies for prediction of the risk of falls have been reported, criteria which worked well in an original research study may not work in a hospital with different patient populations.
We studied the efficacy of risk-assessment tools that contain 41 items of questionnaires proposed by the National Hospital Organization by case-control study. From January 2010 until April 2010, 82 patients were reported as fall accidents and were defined as the cases. Three controls were selected in the source population for each case and matched by gender, age, and the ward in the hospital. Finally, 234 controls were selected. The suitability of 41 items was compared, and odds ratios were calculated. In 16 items, the lower limit of 95% confidential interval of odds ratio was greater than 1.0, and the lower limit of 99% confidential interval of odds ratio was greater than 1.0 in 9 items. Using multiple logistic regression analysis, 5 items were selected by forward selection as significant(p<0.05). These were “Assistance required in dressing”, “History of fall during hospitalization”, “Restlessness and agitation”, “Using wheel-chair, cane, walker or handrail”, and “Using narcotics”. Case-control study is a convenient method to evaluate the efficacy of assessment items.
