老年患者多重用药处方精简干预临床效果的Meta分析

Impact of deprescribing interventions on clinical outcomes in the elderly patients:a Meta-analysis

摘要:

目的 系统评价处方精简干预对老年多重用药患者临床结局的影响。方法 检索数据库公开发表有关老年多重用药患者处方精简干预及相关结局的文献。根据纳入排除标准进行筛选,采用Review Manager 5.3软件,利用I²衡量文献研究内容的异质性,运用固定效应模型或随机效应模型合并分析处方精简干预对老年多重用药患者临床结局的影响。结果 本研究最终纳入18篇随机对照研究(RCT),文献总体质量较好。Meta分析显示,处方精简干预不能降低老年多重用药患者全因死亡率[OR=0.86, 95%CI(0.67~1.09)]。不同干预方式比较:特异性处方精简干预在一定程度上降低全因死亡率[OR=0.68, 95%CI(0.51~0.92)；P＜0.05]。随访时间长短比较:随访时间较长(＞6个月)相较于随访时间较短(≤6个月)的处方精简干预在降低全因死亡率方面具有一定优势[OR=0.58, 95%CI(0.39~0.86) vsOR=1.02, 95%CI(0.76~1.36)；P＜0.05]。不同年龄段比较:对不同年龄段患者进行精简干预,全因死亡率没有变化[OR=0.63,5%CI(0.40~1.02) vsOR=0.95, 95%CI(0.72~1.25)；P＞0.05]。认知状态比较:对不同认知状态患者进行精简干预,全因死亡率不发生改变[OR=0.63, 95%CI(0.37~1.07) vsOR=0.93, 95%CI (0.71~1.22)；P>0.05]。处方精简干预不能减少跌倒患者的数量[OR=0.98, 95%CI(0.74~1.27)],但可以显著降低年人均跌倒次数[MD=-0.11,5%CI(-0.21~-0.02)],缩短患者住院时长[MD=-0.49,5%CI(-0.76~-0.22)]。结论 现有数据分析表明,处方精简干预不能降低老年多重用药患者的全因死亡率,特异性或长时间随访的处方精简干预在降低患者全因死亡率方面有一定优势；处方精简干预不能减少跌倒患者人数但可以减少患者跌倒次数；处方精简干预有缩短住院时长的趋势。特异性处方精简干预在减少不适当的多重用药方面是安全可行的。

Abstract:

Objective To perform a systematic review to evaluate the impact of deprescribing interventions on the clinical outcomes in the elderly patients with polypharmacy. Methods A thorough search was made in the databases of the literature related to deprescribing interventions in the elderly patients with polypharmacy. The studies were screened according to the inclusion and the exclusion criteria. I² test for heterogeneity among studies was performed using Review Manager 5.3. Fixed effect model or random effect model was employed to analyze the impact of deprescribing interventions in the elderly patients with polypharmacy. Results A total of 18 studies were selected for the final randomized controlled trials (RCT), with overall quality of the literature being good. Meta-analysis showed that deprescribing interventions were not able to decrease all-cause mortality among elderly patients with polypharmacy [OR=0.86, 95%CI (0.67-1.09)]. The comparison between different intervention methods showed that patient-specific interventions decreased all-cause mortality [OR=0.68, 95%CI (0.51-0.92)]. The comparison of the follow-up length showed that long follow-up (>6 months) outperformed shorter follow-up (≤6 months) in decreasing all-cause mortality [OR=0.58,5%CI (0.39-0.86) vsOR=1.02, 95%CI (0.76-1.36); P＜0.05]. The comparison between different age groups showed no change in all-cause mortality [OR=0.63,5% CI (0.40-1.02) vs OR=0.95,5%CI (0.72-1.25); P＞0.05]. The comparison of cognitive status showed that deprescribing interventions in the groups of different cognitive statuses did not alter all-cause mortality [OR=0.63,5%CI(0.37-1.07) vs OR=0.93, 95%CI (0.71-1.22); P>0.05]. Deprescribing interventions did not decrease the number of patients with[JP+1]falls [OR=0.98, 95%CI (0.74-1.27)] but were able to significantly decrease the average falls they experienced [MD=-0.11,95%CI (-0.21--0.02)] and the length of hospital stay [MD=-0.49,5%CI (-0.76--0.22)]. Conclusion The analysis of data available suggested that deprescribing interventions did not decrease all-cause mortality in elderly patients with polypharmacy, but patient-specific or long-term follow-up seemed to have an advantage in decreasing it; that deprescribing interventions did not decrease the number of patients with falls but reduced the number of falls they experienced and the length of hospital stay. Patient-specific deprescribing interventions seem to be safe and feasible in decreasing inappropriate polypharmacy.