Page 55 - 2022-34-中国全科医学
P. 55

·4276· http://www.chinagp.net   E-mail:zgqkyx@chinagp.net.cn


           型,尽量做到将纵向数据与事件发生时间数据联合起来                                appraisal and data extraction for systematic reviews of prediction
           进行分析,进而使参数估计的偏差进一步减小,提高统                                modelling studies:the CHARMS checklist[J]. PLoS Med,
                                                                   2014,11(10):e1001744. DOI:10.1371/journal.pmed.1001744.
           计推断的效率。
                                                               [7]MOONS K G M,WOLFF R F,RILEY R D,et al. PROBAST:
           3.3.3 模型验证与应用 本研究发现,29.0%(9/31)                         a tool to assess risk of bias and applicability of prediction model
           的研究对模型进行了外部验证(其中 4 项研究采用内部                              studies:explanation and elaboration[J]. Ann Intern Med,2019,
           验证与外部验证相结合的方法对模型进行了验证),大                                170(1):w1-33. DOI:10.7326/M18-1377.
           部分模型仍处于内部验证阶段,这使得现有模型的转化                            [8]WOLFF R F,MOONS K G M,RILEY R D,et al. PROBAST:a tool
                                                                   to assess the risk of bias and applicability of prediction model studies[J].
           率较低且偏倚风险较高。外部验证和重新校准有助于增
                                                                   Ann Intern Med,2019,170(1):51. DOI:10.7326/M18-1376.
           强模型的泛化能力,与开发新模型相比,具有更高的成                            [9]AEKPLAKORN W,BUNNAG P,WOODWARD M,et al. A risk score
           本效益。医务工作者可基于本研究结果选取高质量的                                 for predicting incident diabetes in the Thai population[J]. Diabetes
           T2DM 发病风险预测模型并对其进行外部验证(大样本、                             Care,2006,29(8):1872-1877. DOI:10.2337/dc05-2141.
                                                               [10]CHIEN K,CAI T,HSU H,et al. A prediction model for type 2
           多中心数据)后,将其应用于临床;或通过开展多中心、
                                                                    diabetes risk among Chinese people[J]. Diabetologia,2009,52(3):
           跨国预测模型构建研究,提高 T2DM 发病风险预测模型
                                                                    443-450. DOI:10.1007/s00125-008-1232-4.
           的临床应用率、转化率。也可基于电子病历系统构建自                            [11]GAO W G,QIAO Q,PITKÄNIEMI J,et al. Risk prediction models for
           动化的 T2DM 发病风险预测工具;或将传统预测工具与                              the development of diabetes in Mauritian Indians [J]. Diabet Med,
           计算机技术相结合,如基于网络的交互式列线图,使模                                 2009,26(10):996-1002. DOI:10.1111/j.1464-5491.2009.02810.x.
                                                               [12]SUN F,TAO Q,ZHAN S. An accurate risk score for estimation 5-year
           型可以与电子病历系统相结合的同时,也可借由网页在
                                                                    risk of type 2 diabetes based on a health screening population in Taiwan
           移动设备、个人计算机上呈现,进而:(1)同时满足医
                                                                    [J]. Diabetes Res Clin Pract,2009,85(2):228-234.
           务人员和社区糖尿病患者对 T2DM 发病风险预测工具的                         [13]CHUANG S Y,YEH W T,WU Y L,et al. Prediction equations and
           使用需求;(2)使模型在得到复杂算法支撑的同时,减                                point system derived from large-scale health check-up data for estimating
           少人工计算的工作量;(3)使模型更易于理解,用户使                                diabetic risk in the Chinese population of Taiwan[J]. Diabetes Res
                                                                    Clin Pract,2011,92(1):128-136. DOI:10.1016/j.diabres.2010.12.022.
           用模型的积极性得以提高。
                                                               [14]BOZORGMANESH M,HADAEGH F,GHAFFARI S,et al. A
               本文存在一定的局限性。首先,受限于文献纳入与                               simple risk score effectively predicted type 2 diabetes in Iranian
           排除标准,本研究纳入的文献可能不够全面,存在一定                                 adult population:population-based cohort study[J]. Eur J Public
           的选择偏倚;其次,由于纳入文献在研究对象及方法上                                 Health,2011,21(5):554-559. DOI:10.1093/eurpub/ckq074.
           存在明显的异质性,本研究未对纳入研究进行定量分析。                           [15]LIU M,PAN C,JIN M. A Chinese diabetes risk score for screening
                                                                    of undiagnosed diabetes and abnormal glucose tolerance[J].
               作者贡献:贺婷负责文章的构思、设计与撰写,并
                                                                    Diabetes Technol Ther,2011,13(5):501-507.
           对文章整体负责,监督管理;贺婷、叶子溦、李饶负责                            [16]ONAT A,CAN G,YÜKSEL H,et al. An algorithm to predict risk of
           文献、资料的收集与整理;贺婷、袁丽,杨小玲负责文                                 type 2 diabetes in Turkish adults:contribution of C-reactive protein[J].
           献偏倚风险和适用性评估、论文的修订;贺婷、古艳负                                 J Endocrinol Invest,2011,34(8):580-586. DOI:10.3275/7323.
           责文章的质量控制及审校。                                        [17]DOI Y,NINOMIYA T,HATA J,et al. Two risk score models for
                                                                    predicting incident Type 2 diabetes in Japan[J]. Diabet Med,
               本文无利益冲突。
                                                                    2012,29(1):107-114. DOI:10.1111/j.1464-5491.2011.03376.x.
           参考文献                                                [18]HEIANZA Y,ARASE Y,HSIEH S D,et al. Development of
           [1]JOHNSON E L,FELDMAN H,BUTTS A,et al. Standards of medical   a new scoring system for predicting the 5 year incidence of type 2
               care in diabetes:2020 abridged for primary care providers[J]. Clin   diabetes in Japan:the Toranomon Hospital Health Management
               Diabetes,2019,43(s1):1-212. DOI:10.2337/cd20-as01.   Center Study 6(TOPICS 6)[J]. Diabetologia,2012,55(12):
           [2]OGURTSOVA K,GUARIGUATA L,BARENGO N C,et al. IDF       3213-3223. DOI:10.1007/s00125-012-2712-0.
               diabetes Atlas:global estimates of undiagnosed diabetes in adults for   [19]LIM N K,PARK S H,CHOI S J,et al. A risk score for predicting
               2021[J]. Diabetes Res Clin Pract,2022,183:109118.    the incidence of type 2 diabetes in a middle-aged Korean cohort:
           [3]YANG J J,YU D,WEN W,et al. Association of diabetes with all-  the Korean Genome and Epidemiology Study[J]. Circ J,2012,
               cause and cause-specific mortality in Asia:a pooled analysis of more   76(8):1904-1910. DOI:10.1253/circj.cj-11-1236.
               than 1 million participants[J]. JAMA Netw Open,2019,2(4):  [20]YE X,ZONG G,LIU X,et al. Development of a new risk score for
               e192696. DOI:10.1001/jamanetworkopen.2019.2696.      incident type 2 diabetes using updated diagnostic criteria in middle-
           [4]CHARBONNEL B,SIMON D,DALLONGEVILLE J,et al. Direct    aged and older Chinese[J]. PLoS One,2014,9(5):e97042.
               medical costs of type 2 diabetes in France:an insurance claims database   [21]XU L,JIANG C Q,SCHOOLING C M,et al. Prediction of 4-year
               analysis[J]. Pharmacoecon Open,2018,2(2):209-219.    incident diabetes in older Chinese:recalibration of the Framingham
           [5]贺小宁,张雅雯,阮贞,等 . 中国 2 型糖尿病患者慢性并发症患                      diabetes score on Guangzhou Biobank Cohort Study[J]. Prev
               病率与次均医疗费用研究[J]. 中华内分泌代谢杂志,2019,                      Med,2014,69:63-68. DOI:10.1016/j.ypmed.2014.09.004.
               35(3):6. DOI:10.3760/cma.j.issn.1000-6699.2019.03.004.  [22]NANRI A,NAKAGAWA T,KUWAHARA K,et al. Development
           [6]MOONS K G,DE GROOT J A,BOUWMEESTER W,et al. Critical   of risk score for predicting 3-year incidence of type 2 diabetes:
   50   51   52   53   54   55   56   57   58   59   60