Thomson Reuters, 4301 Connecticut Ave. NW, Washington, DC 20008, USA.
Diabetes Obes Metab. 2012 Jul;14(7):634-43. doi: 10.1111/j.1463-1326.2012.01583.x. Epub 2012 Mar 14.
This retrospective observational study examined the association between International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM)-coded outpatient hypoglycaemic events and fall-related fractures in Medicare-covered patients with type 2 diabetes.
Data were derived from healthcare claims for individuals with employer-sponsored Medicare supplemental insurance. The study period consisted of two consecutive 1-year periods; the baseline period (1 April 2008 to 31 March 2009) and the evaluation period (1 April 2009 to 31 March 2010). Patients selected for study were at least 65 years of age with evidence of type 2 diabetes during the baseline period, as identified using a Healthcare Effectiveness Data and Information Set algorithm or by at least two prescription claims for oral antidiabetic drugs. The baseline period was used to collect information on the patients' demographics and clinical characteristics. The evaluation period was used to identify the presence of hypoglycaemic events and fall-related fractures. Logistic regression was employed to examine the association between hypoglycaemic events and fall-related fractures occurring during the evaluation period, adjusting for patients' demographics and clinical characteristics.
Of 361 210 included patients, 16 936 had hypoglycaemic events during the evaluation period. Patients with hypoglycaemic events had 70% higher regression-adjusted odds (hypoglycaemic events odds ratio = 1.70; 95% confidence interval = 1.58-1.83) of fall-related fractures than patients without hypoglycaemic events. Multiple sensitivity analyses also yielded results suggesting increased odds of fall-related fractures in patients with hypoglycaemic events.
ICD-9-CM-coded outpatient hypoglycaemic events were independently associated with an increased risk of fall-related fractures. Further studies of the relationship between hypoglycaemia and the risk of fall-related fractures are warranted.
本回顾性观察研究旨在探讨医疗保险覆盖的 2 型糖尿病患者中，国际疾病分类第 9 版临床修订版（ICD-9-CM）编码的门诊低血糖事件与跌倒相关骨折之间的关联。
数据来自具有雇主赞助的医疗保险补充保险的个人的医疗保健索赔。研究期间包括两个连续的 1 年期间；基线期（2008 年 4 月 1 日至 2009 年 3 月 31 日）和评估期（2009 年 4 月 1 日至 2010 年 3 月 31 日）。选择参加研究的患者年龄至少为 65 岁，并且在基线期有 2 型糖尿病的证据，这是通过医疗保健效果数据和信息集算法或至少两种口服抗糖尿病药物的处方来确定的。基线期用于收集患者人口统计学和临床特征的信息。评估期用于确定低血糖事件和跌倒相关骨折的发生情况。采用逻辑回归分析评估评估期发生的低血糖事件与跌倒相关骨折之间的关联，同时调整患者的人口统计学和临床特征。
在 361210 名纳入患者中，有 16936 名患者在评估期内发生低血糖事件。与无低血糖事件的患者相比，发生低血糖事件的患者发生跌倒相关骨折的调整后优势比（低血糖事件优势比 = 1.70；95%置信区间 = 1.58-1.83）高 70%。多项敏感性分析也表明，发生低血糖事件的患者跌倒相关骨折的几率增加。
ICD-9-CM 编码的门诊低血糖事件与跌倒相关骨折的风险增加独立相关。有必要进一步研究低血糖与跌倒相关骨折风险之间的关系。
