STUDIO PROSPETTICO PER LA VALUTAZIONE DEI FATTORI DI RISCHIO DELLE CADUTE NELLE RESIDENZE SANITARIE ASSISTENZIALI (RSA) PER ANZIANI. FALL RISK ASSESSMENT IN INSTITUTIONALIZED ELDERLY LONGITUDINAL STUDY - FRAILS

Keywords: falls; risk factors; older; nursing homes Introduction Falls are common in older persons and sometimes lead to unfavorable outcomes, as traumas and hip fracture. The elderly living in nursing homes are at greater risk of injury resulting from a fall, due to a higher degree of frailty. Some factors are more associated to falls and especially to traumatic falls: functional deficiency, use of aids for walking, osteoporosis, postural instability, deficit of balance and gait; chronic disease, cognitive impairment, and polytherapy use of psychotropic drugs, new environments, previous falls (1,2). The guidelines on the prevention of falls in older people recommend the evaluation of risk factors, as a key element to prevent further falls and, especially, trauma secondary to falls (3,4). Aim of the study and Methods In order to know the risk factors associated with falls, the incidence of falls and traumatic falls in people cared for in nursing home, it was performed a prospective observational study with 12 months of follow up (October 2013 - September 2014). For every resident we detect the functional, cognitive and nutritional status, comorbidity, number and type of drugs, risk of falling, number and outcomes of falls. It was used univariate and multivariate logistic regression to estimate the predictive role of clinical variables on people who had fallen. Results were considered significant at a level of p value <0.05 and a confidence interval of 95%, with two-tailed test. Results The study, carried out in two nursing homes in Milan, involved 409 residents of which 331 in ordinary units, and in 78 specialized units for dementia, predominantly women (82%) with a mean age of 83 years (± 9.4). Most of the residents had a total dependence (43.8%) or severe (32.8%) in the activities of daily living (Barthel index median: 26) and mild (11%), moderate (20%) or severe (42%) cognitive impairment, detected by the Mini Mental State Examination. Residents showed an average 6.30 (± 2.2) disease with moderate severity (2.24 ± 0.47), derived by the average of 13 categories of the Cumulative Illness Rating Scale. In 117 residents it was detected a risk of malnutrition: moderate (11.5%) or high (17.1%), using the Malnutrition Universal Screening Tool. However only 67 residents were underweight (body mass index: 0 to 18.49). In one of the nursing home 83% (185/223) had a risk of falling, with a median value of 5, detected by the Conley Scale (≥2 values are indicative of risk). In the second RSA, based on the scores of Tinetti Balance Assessment Tool, 106/186 residents were at risk of falling, including 31.2% lower and 25.8% higher, while 80 residents were not considered at risk, because they did not walk. About 40% of the residents were restrained with one or more restraint and at 286/409 residents (70%) were applied to bilateral bed rails. The residents took an average of 8.5 (± 3.4) active ingredients per day (median 8, range 0-19), of which an average of 4.8 (± 2.2) drugs at risk (median 5, range 0-14 ) which include cardiovascular drugs, agents on metabolism/alimentary tract and acting on the nervous system. In the observation period 111/409 (27%) residents fell, of which 54 had a lesion from mild to severe. It was detected an average of 1.26 (± 0.48) falls/resident, with a range from 1 to 10 and a higher percentage (and statistically significant) of falls among males (28/75) compared with women (83/334 ) and the residents cared for in special unit for dementia (29/78) than residents in ordinary units (83/331). Multivariate logistic regression confirmed statistically significant risk factors: type of nursing home unit, level of dependence in activities of daily living, previous falls, psycholeptics. Residents with a lower dependence in activities of daily living (Barthel> 27), compared to those with a total dependence, had a risk more than four times significantly higher (OR: 4.75; 95% CI 2.50 to 9.02 p: 0.000). Previous falls bring a three-fold risk of falling (OR: 3.34; 95% CI 1.5 to 7.44; P 0.003), instead the residents cared for in specialized units presented a two-fold risk (OR: 2.22; 95% CI 1.16 to 4.27; P 0.016). Although it was less significant, there was a greater risk of falling in people taking antipsychotic drugs (OR: 1.66; 95% CI 0.99 to 2.76; p: 0.052). Conclusion The prevalence of falls is quite in line with results described in literature. A lesser dependency in ADL seems to have a high predictive role, although in some studies was observed an opposite direction. The results confirmed that a previous fall should be taken as alert of further falls, although it is considered a predictor of more than one risk factor, and as such it cannot be eliminated or preventable, but can direct a first level of screening. The risk assessment should be the first step to predict recurrent falls, although it should accept some degree of risk, because it is not possible to eliminate all of the falls. The results of this study will allow to direct the development of new research hypotheses, including the evaluation of multidimensional programs for the prevention of traumatic falls in nursing home, specifically targeted to major risk factors found in the elderly, which include the improvement / maintenance of functional ability and medication review.