中文摘要：

目的：观察＂出痧＂刮痧、＂不出痧＂刮痧与针刺治疗慢性非特异性下腰痛（NLBP）的临床疗效差异。方法：通过中央网络随机分组系统,将150例符合纳入标准的NLBP患者随机分为出痧组、不出痧组和针刺组,每组50例。出痧组在受试者腰部（肾俞、志室到次髎、秩边）、下肢（从承扶过委中至承山）进行刮痧,每个部位刮8-10次,并且要求＂刮致出痧＂,每次刮痧治疗间隔4d,总共治疗7次。不出痧组操作方法与出痧组相同,但要求＂不出痧＂,每次刮痧治疗间隔2d,总共治疗7次。针刺组针刺患者肾俞和委中,直刺进针,施平补平泻手法至局部出现酸麻肿重感,每次治疗30min,每次针刺间隔2d,总共治疗7次。比较3组受试者治疗前、治疗结束后及1个月与3个月的随访时视觉模拟量表（VAS）、自拟临床症状积分和Oswetery功能障碍指数（ODI）。根据M5P决策树算法,使用Weka Explorer 3.6软件计算刮痧治疗NLBP疗效预测方程。结果：1治疗结束时,出痧组脱落7例,不出痧组脱落6例,针刺组脱落8例,最终完成129例。3组受试者在治疗过程中未发生不良反应。2治疗结束后,3组NLBP患者VAS、临床症状积分、ODI指数均明显降低（P〈0.05）,且随访1个月、3个月随访疗效稳定;出痧组3个指标的改善均优于不出痧组和针刺组,针刺组3个指标的改善亦优于不出痧组（均P〈0.05）。3根据M5P决策树算法,提出刮痧治疗NLBP疗效预测方程,且使用ODI作为因变量方程拟合度较高。结论：刮痧治疗NLBP疗效明显,安全性高,且中短期疗效优于＂不出痧＂刮痧和针刺治疗,表明刮痧＂出痧＂是临床疗效的重要影响因素;同时ODI可以作为刮痧治疗腰痛的敏感指标之一。

英文摘要：

Objective： To observe the clinical efficacy differences between scrape therapy with sha symptom, scrape therapy without sha symptom and acupuncture for nonspecific low back pain （NLBP）. Methods： By using center network system for randomization, one hundred and fifty patients of NLBP who met the inclusive criteria were randomly divided into a sha symptom group, a non-sha symptom group and an acupuncture group, 50 cases in each group. The patients in the sha symptom group were treated with scrape therapy in the back and low extremity, 8-10 scrapings on each area, and sha symptom was mandatory; scrape therapy was given every 4 days for totally 7 sessions. The patients in the non-sha symptom group received identical scrape therapy as sha symptom group, but sha symptom was not mandatory; the scrape therapy was given every 2 days for totally 7 sessions. The patients in the acupuncture group were vertically needled at Shenshu （BL 23） and Weizhong （BL 40） with mild reinforcing-reducing method till the arrival of qi, the acupuncture was given 30rain per session, once every two days for totally 7 sessions. The VAS （visual analogue scale）, self-made symptom score and Oswestry disability index （ODI） were compared before and after treatment as well as in the 1-month and 3-month follow-up visit. Weka Explorer 3.6 software was applied to evaluate the predictive formula of Guasha for NLBP. Results： （1）When the treatment finished, 7 cases in the sha symptom group, 6 cases in the non-sha symptom group and 8 cases in the acupuncture dropped out; 129 cases finished the treatment. There was no adverse event during the treatment among three groups. （2）The VAS, self-made symptom score and ODI were all significantly reduced after treatment （P〈0.05）, which was also stable in the 1-month and 3-month follow-up. The improvement in the sha symptom group was superior to that in the non-sha symptom group and acupuncture group （P〈0.05）, and the improvement in the acupuncture group was superior to th