ABSTRACT
The aim of the study was to establish factors with an impact on fracture risk and to develop an algorithm to predict osteoporotic fracture. A total of 978 postmenopausal women from the epidemiological, population-based RAC-OST-POL study with a mean age of 65.7 ± 7.3 years were enrolled. At baseline, bone mineral density at hip and clinical risk factors for fracture were collected. Afterward, each person was asked annually on fracture incidence in the 5-year follow-up. Finally, data for complete 5-year observation were gathered for the group of 802 patients. During the follow-up, 92 osteoporotic fractures occurred in 78 women. The most common fracture site was the forearm (n = 45). The following baseline factors were found as significant for fracture incidence: femoral neck bone mineral density, prior fractures, steroid use, falls within previous 12 months, and height. Fracture risk was predicted by the following formula: Riskoffractureincidence=11+e-(-9.899+1.077∗STEROIDS+0.681∗PRIORFALLS+0.611∗PRIORFRACTURES-0.483∗FNTscore+0.042∗HEIGHT). In our current longitudinal study, an algorithm predicting fracture occurrence over a period of 5 years was developed. It may find application in daily medical practice.
Subject(s)
Algorithms , Osteoporosis, Postmenopausal/complications , Osteoporotic Fractures/epidemiology , Risk Assessment/methods , Accidental Falls/statistics & numerical data , Adrenal Cortex Hormones/therapeutic use , Aged , Bayes Theorem , Body Height , Bone Density , Female , Femur Neck/physiology , Follow-Up Studies , Forearm Injuries/epidemiology , Humans , Incidence , Longitudinal Studies , Middle Aged , Osteoporosis, Postmenopausal/physiopathology , Poland/epidemiology , Recurrence , Risk FactorsABSTRACT
UNLABELLED: In 770 postmenopausal women, the fracture incidence during a 4-year follow-up was analyzed in relation to the fracture probability (FRAX risk assessment tool) and risk (Garvan risk calculator) predicted at baseline. Incident fractures occurred in 62 subjects with a higher prevalence in high-risk subgroups. Prior fracture, rheumatoid arthritis, femoral neck T-score and falls increased independent of fracture incidence. INTRODUCTION: The aim of the study was to analyze the incidence of fractures during a 4-year follow-up in relation to the baseline fracture probability and risk. METHODS: Enrolled in the study were 770 postmenopausal women with a mean age of 65.7 ± 7.3 years. Bone mineral density (BMD) at the proximal femur, clinical data, and fracture probability using the FRAX tool and risk using the Garvan calculator were determined. Each subject was asked yearly by phone call about the incidence of fracture during the follow-up period. RESULTS: Of the 770 women, 62 had a fracture during follow-up, and 46 had a major fracture. At baseline, BMD was significantly lower, and fracture probability and fracture risk were significantly higher in women who had a fracture. Among women with a major fracture, the percentage with a high baseline fracture probability (>10 %) was significantly higher than among those without a fracture (p < 0.01). Fracture incidence during follow-up was significantly higher among women with a high baseline fracture probability (12.7 % vs. 5.2 %) and a high fracture risk (9.2 vs. 5.3 %) so that the "fracture-free survival" curves were significantly different (p < 0.05). The number of clinical risk factors noted at baseline was significantly associated with fracture incidence (chi-squared = 20.82, p < 0.01). Prior fracture, rheumatoid arthritis, and femoral neck T-score were identified as significant risk factors for major fractures (for any fractures, the influence of falls was also significant). CONCLUSIONS: During follow-up, fracture incidence was predicted by baseline fracture probability (FRAX risk assessment tool) and risk (Garvan risk calculator). A number of clinical risk factors and a prior fracture, rheumatoid arthritis, femoral neck T-score, and falls were independently associated with an increased incidence of fractures. [Corrected]
