French Dustin D, Campbell Robert, Spehar Andrea, Cunningham Francesca, Foulis Philip
VISN-8 Measurement and Evaluation Team, James A. Haley Hospital, Tampa, FL 33612, USA.
Drugs Aging. 2005;22(10):877-85. doi: 10.2165/00002512-200522100-00006.
Certain medications have been linked to falls. One of the most severe fall-related injuries in the elderly is a hip fracture.
The objective of this study was to examine the use of medications known to increase fall risk that were prescribed on an outpatient basis to veterans prior to hospital admission for hip fracture.
We identified and analysed the use of outpatient fall-related medications in 2212 unique patients with hip fractures admitted in fiscal year 2003 to Veterans Health Administration hospitals compared with that for matched controls (exact match for age and sex) admitted for acute myocardial infarction (MI) or pneumonia. We analysed the medications selected from the three drug categories most often linked with an increased risk for falls. These categories included medications that affected the cardiovascular (CVS), CNS or the musculoskeletal system (MSS). The unit of analysis was the hip fracture linked with outpatient medications in the study group compared with matched control groups of patients with hospitalisations for an acute MI or pneumonia.
Of the 2212 hip-fracture patients, 70% had fall-related medications prior to hospitalisation for hip fracture. The most notable differences in usage were seen in the drug classes antiepileptics/barbiturates, antidepressants (2-fold difference in use of selective serotonin reuptake inhibitors [SSRIs] and tricyclic antidepressants [TCAs]), and antiparkinson's drugs (nearly 4-fold difference between cases and controls). There were also notable differences in usage of antipsychotics (3-fold difference for hip-fracture cases compared with acute MI) and cholinesterase inhibitors (nearly 2-fold difference for hip-fracture cases compared with pneumonia or acute MI). The most notable differences in polypharmacy combinations were CVS and CNS categories with differences of 9.44% (absolute) and 43% (relative) for hip-fracture patients over acute MI, and 4.83% (absolute) and 18% (relative) for hip-fracture patients over pneumonia patients.
This is the first national Veterans Health Administration hip fracture hospitalisation study that temporally linked outpatient fall-related medications in hip-fracture patients with matched controls. We found that of veterans with hip-fracture hospitalisations, 70% were prescribed outpatient medications from selected major drug categories that may potentially increase fall risk. Moreover, over one-third of hip-fracture patients received concomitant prescriptions of drugs from multiple selected drug categories. Hip-fracture patients, compared with matched controls of acute MI and pneumonia, had the largest pronounced differences in prescribed medications in the antiepileptics, antidepressants, antipsychotics and antiparkinson's drug classes. Although a randomised clinical trial is the 'gold standard' for determining causation issues, exposing patients, particularly the elderly, to the potential risk of injurious falls would raise serious patient safety research approval issues. If the relationship between selected drugs and falls is indeed to some extent causative, future retrospective multivariate analyses could quantify the magnitude of these effects.
某些药物与跌倒有关。老年人与跌倒相关的最严重损伤之一是髋部骨折。
本研究的目的是检查在髋部骨折住院前门诊开具给退伍军人的已知会增加跌倒风险的药物的使用情况。
我们识别并分析了2003财年入住退伍军人健康管理局医院的2212例独特髋部骨折患者门诊使用的与跌倒相关药物，并与因急性心肌梗死（MI）或肺炎入院的匹配对照组（年龄和性别精确匹配）进行比较。我们分析了从最常与跌倒风险增加相关的三类药物中选择的药物。这些类别包括影响心血管（CVS）、中枢神经系统（CNS）或肌肉骨骼系统（MSS）的药物。分析单位是研究组中与门诊药物相关的髋部骨折，与因急性MI或肺炎住院的患者匹配对照组进行比较。
在2212例髋部骨折患者中，70%在髋部骨折住院前使用了与跌倒相关的药物。在抗癫痫药/巴比妥类药物、抗抑郁药（选择性5-羟色胺再摄取抑制剂[SSRIs]和三环类抗抑郁药[TCAs]的使用差异为2倍）和抗帕金森病药物（病例与对照组之间差异近4倍）等药物类别中，使用情况存在最显著差异。抗精神病药物的使用也存在显著差异（髋部骨折病例与急性MI相比差异为3倍），胆碱酯酶抑制剂的使用也存在显著差异（髋部骨折病例与肺炎或急性MI相比差异近2倍）。多药联合使用中最显著的差异在于CVS和CNS类别，髋部骨折患者与急性MI患者相比，绝对差异为9.44%，相对差异为43%；髋部骨折患者与肺炎患者相比，绝对差异为4.83%，相对差异为18%。
这是退伍军人健康管理局首次进行的全国性髋部骨折住院研究，该研究在时间上把髋部骨折患者门诊使用的与跌倒相关药物与匹配对照组联系起来。我们发现，在髋部骨折住院的退伍军人中，70%被开具了可能会增加跌倒风险的选定主要药物类别的门诊药物。此外，超过三分之一的髋部骨折患者接受了来自多个选定药物类别的联合处方。与急性MI和肺炎的匹配对照组相比，髋部骨折患者在抗癫痫药、抗抑郁药、抗精神病药和抗帕金森病药物类别中的处方药物差异最为明显。尽管随机临床试验是确定因果关系问题的“金标准”，但让患者，尤其是老年人面临跌倒受伤的潜在风险会引发严重的患者安全研究审批问题。如果选定药物与跌倒之间的关系确实在某种程度上是因果关系，未来的回顾性多变量分析可以量化这些影响的程度。
