| 1. BackgroundPostmenopausal osteoporosis fracture is the most destructive ending of osteoporosis. It has now become the second threat to the health and quality of life after cardiovascular in the elderly. With the aging of population and the emphasis on quality of life, osteoporosis and Postmenopausal osteoporosis fractures is becoming a social problem plaguing the elderly. How to effectively predict the Postmenopausal osteoporosis fractures risks and take effective preventive measures against its relevant factors plays a vital role in preventing the occurrence of Postmenopausal osteoporosis fractures.As proved by modern medicine, many interacting risk factors play a role in its occurrence, such as age, sex, menopause, whether there is a family history of fractures, race, they can predict the occurrence of Postmenopausal osteoporosis fractures. After40years, the incidence of Postmenopausal osteoporosis fractures increases with age, as is said in plain questions that "after40years old, the yin qi is reduced by half and the daily life is weak". It means, with age-after40, the bodies function decline gradually. Symptoms like soreness of the waist, back pain, cramp of the lower limb and weakness appears. These are related to TCM syndrome factors like kidney deficiency, liver deficiency, spleen deficiency and blood stagnation. Symptoms can act as a warning of the occurrence of Postmenopausal osteoporosis fracture in the early stagesThe risk assessment tool of postmenopausal osteoporosis fracture is mostly developed on the basis of modern risk factors and thus lacks syndrome differentiation, which may limit its utilization. Due to differences in the populations of race, region, diet and socio-cultural aspects, its application often limited. Currently few postmenopausal osteoporosis fracture prediction tool was established based on characteristics of the population, and the risk factors and clinical symptoms. How to incorporate the modern medicine-related risk factors and establish postmenopausal osteoporosis fracture risk assessment tool in line with the characteristics of China’s population, will help to improve the accuracy of risk assessment tool for our high-risk population of postmenopausal osteoporosis fracture.2. Objective Initially establish the early Postmenopausal osteoporosis fracture risk prediction tool based on the demographic characteristics of40-65years old women in Beijing and Shanghai, and the combination of risk factors and TCM symptoms.3. Material and methods3.1Type of Study DesignThis is a registry study.3.2Researching populationFrom March2009to～July2009, in Dongcheng District in Beijing and Xuhui District in Shanghai, the on-site recruitment methods was used to record the habit and information of pregnancy and births of women aged between40to65years ".Secondary osteoporosis were excluded.3.3Survey"Questionnaires for risk factors and syndrome of osteoporosis in community woman aged40to65years" was developed from the basis of lot of work of the research group, it has good reliability and validity, and it includes:Content demographic, lifestyle, risk factors, and TCM symptoms, bone density testing.3.4Follow up visitFollow up visit to the participated women in the survey was done respectively in March2010to August and June2011to November, on-site visits and telephone interviews was used as a combination. The main content of the survey visit are:the investigation period, nearly one year if a fracture occurs, the reasons for fractures, fracture location and fracture time.3.5Quality controlAll data collected was through strict verification and management, and qualified questionnaire went through the independent dual-input and dual-check. The data was input in the "Osteoporosis Health Management System" which was developed in collaboration with Beijing University of Science and Technology, and a consistency check was done to ensure the authenticity and integrity.3.6Statistical software and methods3.6.1Statistical softwareSPSS19.0software was used on a general description of the follow-up populations, R2.15.3software imputers; survivalROC was used to conduct the Construction and analysis of forecasting models.3.6.2Analysis of influence Factors and the establishment of risk modelsInvestigate factors was regarded as covariates and prognostic factors were screened using Cox unilabiate, based on the influence factors screened by the single factor, Cox risk model was established to predict the risk of occurrence of Postmenopausal osteoporosis fractures in two years.3.6.3Evaluation on risk prediction modelReceiver Operating Characteristic curve (ROC) was used to evaluate the accuracy of risk prediction model. Through area under the ROC curve (AUC), the performance of the model was evaluated, comparison among the three models was conducted by Z test.4. Result4.1The completion of follow up visitThree investigation was carried from March2009to the end of November2011,1823questionnaire was disseminated and a total of1498people have completed the follow-up, follow-up response rate was82.17percent, falls and incomplete questionnaire accounted for17.83%.4.2General DescriptionAmong the1498people participated in the follow-up,50occurred Postmenopausal osteoporosis fractures. As to whether Postmenopausal osteoporosis fractures occurred, two populations were classified, differences occurred as to the "age, menopause, the production number of years since menopause, BMD values, evil heat, loose stools, lower limb Spasm, dizziness and so on" between the two groups (P<0.05), And the corresponding RR value>1.4.3Influencing factors screening of the Postmenopausal osteoporosis fracturesTime to Postmenopausal osteoporosis fractures was regarded as the outcome variable. By unilabiate Cox proportional hazards model selection, differences was significant as to age, menopause, menopause, production number, bone density, lower limb Spasm, dizziness, etc.(P<0.05)4.4The construction of multivariate Cox proportional hazards modelThe three types of "risk factors+bone density+TCM symptoms";"no BMD+with risk factors+with TCM symptoms";"with BMD+with risk factors+no TCM symptoms " were used to constructed the Cox proportional hazard assessment model: under the condition of "with BMD+with risk factors+with TCM symptoms ",the AUC1was0.750(0.684-0.815); under the condition of "with BMD+with risk factors+no TCM symptoms "the AUC2was0.697(0.628-0.767); under the condition of "no BMD+with risk factors+with TCM symptoms", the AUC3was0.726(0.654-0.798). The predictive accuracy of the existence of differences between Model1and Model2is statistically significant (P<0.05). In the absence of BMD testing, the predictive accuracy is no statistical difference between Model1and Model3(P>0.05)4.5The establishment of Postmenopausal osteoporosis fracture risk prediction toolsAccording to various risk factors when optimal prediction models at different levels of regression coefficients, the for scoring system was formulated, thereby two-year Postmenopausal osteoporosis fracture risk prediction tools was established.5. ConclusionBased on characteristics of the population, early Postmenopausal osteoporosis fracture risk prediction tool was established and on the basis of risk factors and with the integration of relevant TCM theory, satisfactory results were achieved, which can meet the practical needs of TCM clinics. |
PDF Full Text Request