Fatores de risco e ferramentas clínico-funcionais de rastreio do risco de quedas em idosas com baixa densidade óssea: um estudo longitudinal

• Ano de defesa: 2014
• Autor(a) principal: Patricia Azevedo Garcia
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: por
• Instituição de defesa: Universidade Federal de Minas Gerais; UFMG
• Palavras-chave em Português: Análise de regressão; Coleta de dados; Idoso; Densidade mineral óssea; Acidentes por quedas; Osteoporose; Recordação mental; Medição de risco; Sensibilidade e especificidade; Fatores de risco; Força muscular; Fisioterapia para idosos; Idosos Saúde e higiene; Densidade óssea
• Link de acesso: http://hdl.handle.net/1843/BUOS-9LWPNH

Background: In older people with osteoporosis and osteopenia falls are noteworthy among factors that contribute to increased occurrence of fractures because they have a high incidence and are preventable. The identification of intrinsic determinants, the recognition of accuracy tools to predicting falls risk and the identification of validity of retrospective self-report of falls among these older women is an important step to implement preventive strategies and to reduce the incidence of falls and fractures. Objectives: to estimate the incidence of falls and recurrent falls in the present sample, to verify the associated falls risk factors, to determine the accuracy of six clinical-functional tools to predict falls and recurrent falls, to investigate the agreement between retrospective self-report and prospective monitoring methods and to determine the validity of retrospective self-reports of falls and recurrent falls in the period of 12 months in older women with low bone mineral density (BMD). Materials and Methods: A longitudinal study over one year with 116 community-dwelling older women with osteopenia or osteoporosis. The following risk factors for falls were assessed: previous falls, mobility and body balance, multidimensional risk, self-efficacy for falls, frailty and upper and lower limbs muscle performance. The following tolls and cutoff points were used for these assessments: self-report of previous falls ( 1 fall), self-report of previous recurrent falls ( 2 falls), Timed Up and Go test (> 10 seconds), Falls Risk (oscillation > 3.4 or 3.5), QuickScreen (4 risk factors), Falls Efficacy Scale International FES-I ( 23 or 31 points), Frailty Phenotype ( 1 criterion), hydraulic handgrip dynamometer and isokinetic dynamometer. The falls were monitored prospectively by monthly phone calls over a year. The outcomes prospectively investigated were the incidence of falls ( 1 fall) and of recurrent falls ( 2 falls) over one year. At the end of this follow-up older were asked about the recall of falls in the same 12 month period. Logistic regression analyses were performed to determine the association (OR) among the falls risk factors and the occurrence of falls and recurrent falls by stepwise method. Sensitivity, specificity, positive predictive value and negative predictive value were calculated and ROC curves were constructed for each study tool. It was analyzed the agreement between the prospective monitoring and retrospective self-report of falls methods and the sensitivity and the specificity of self-reported of previous falls were calculated. Results: 64 (55.2%) older women reported falls, of which 24 indicated that falls were recurrent during follow-up. The predictive model for falls was composed by muscle performance variable assessed by hamstrings muscle work per body weight (OR = 0.974 95% CI 0.955 to 0.995) and the predictive model for recurrent falls was composed by previous falls (OR = 1.336 95% CI 1.020 to 1.751) and by muscle performance variable assessed by hamstrings muscle work per body weight (OR = 0.966 95% CI 0.941 to 0.992). The areas under the ROC curves were significant to QuickScreen (AUC = 0.614, 95% CI 0.512 to 0.716, p=0.035) for falls prediction, and to self-report of previous falls (AUC = 0.635, 95% CI 0.501 -0.770, p=0.042) and to Falls Risk (AUC = 0.669, 95% CI 0.552 to 0.786, p=0.011) for recurrent falls prediction. Alternative cut points were proposed for QuickScreen ( 2 risk factors) to identify falls risk and for the Falls Risk ( 4.00) to identify risk of recurrent falls. There was moderate agreement between the prospective monitoring and the retrospective self-report of falls to classify fallers (Kappa = 0.595) and recurrent fallers (Kappa = 0.589), and the limits of agreement were 0.35 ± 1.66 fall. The self-report of prior falls had 67.2% sensitivity and 94.2% specificity to classify fallers older women and 50% sensitivity and 98.9% specificity to classify recurrent fallers. Conclusion: The low BMD older women showed high incidence of falls. The hamstring muscle performance measure was a predictor of falls and recurrent falls. The previous falls occurrence was a predictor of recurrent falls. The risk of falls and recurrent falls can be identified among low BMD older women using the following tools: Falls Risk, QuickScreen and self-report of previous falls. However the use of self-report of falls in the previous 12 months should be used with caution because underestimated 32.8% of falls and 50% of recurrent falls.