Purpose

Examine the effectiveness of interventions to approach guideline-adherent surgical referrals for low back pain assessed via systematic review and meta-analysis.

Methods

Five databases (10 September 2021), Google Scholar, reference lists of relevant systematic reviews were searched and forward and backward citation tracking of included studies were implemented. Randomised controlled/clinical trials in adults with low back pain of interventions to optimise surgery rates or referrals to surgery or secondary referral were included. Bias was assessed using the Cochrane ROB2 tool and evidence certainty via Grading of Recommendations Assessment, Development and Evaluation (GRADE). A random effects meta-analysis with a Paule Mandel estimator plus Hartung–Knapp–Sidik–Jonkman method was used to calculate the odds ratio and 95% confidence interval, respectively.

Results

Of 886 records, 6 studies were included (N = 258,329) participants; cluster sizes ranged from 4 to 54. Five studies were rated as low risk of bias and one as having some concerns. Two studies reporting spine surgery referral or rates could only be pooled via combination of p values and gave evidence for a reduction (p = 0.021, Fisher’s method, risk of bias: low). This did not persist with sensitivity analysis (p = 0.053). For secondary referral, meta-analysis revealed a non-significant odds ratio of 1.07 (95% CI [0.55, 2.06], I² = 73.0%, n = 4 studies, Grading of Recommendations Assessment, Development and Evaluation [GRADE] evidence certainty: very low).

Conclusion

Few RCTs exist for interventions to improve guideline-adherent spine surgery rates or referral. Clinician education in isolation may not be effective. Future RCTs should consider organisational and/or policy level interventions.

PROSPERO registration

CRD42020215137.

中文翻译：

---

促进腰痛循证指南一致手术的干预措施：随机对照试验的系统评价和荟萃分析

目的

检查干预措施的有效性，以通过系统回顾和荟萃分析评估符合指南的腰痛手术转诊。

方法

检索了五个数据库（2021 年 9 月 10 日）、谷歌学术、相关系统评价的参考文献列表，并对纳入的研究实施了前向和后向引用跟踪。纳入了针对患有腰痛的成人进行干预以优化手术率或转诊至手术或二次转诊的随机对照/临床试验。使用 Cochrane ROB2 工具评估偏倚，并通过建议评估、制定和评估分级 (GRADE) 评估证据的确定性。使用 Paule Mandel 估计量和 Hartung–Knapp–Sidik–Jonkman 方法进行的随机效应荟萃分析分别用于计算比值比和 95% 置信区间。

结果

在 886 条记录中，有 6 项研究被纳入 ( N  = 258,329) 名参与者；集群大小从 4 到 54 不等。五项研究被评为低偏倚风险，一项被评为存在一些问题。两项报告脊柱手术转诊或比率的研究只能通过p值的组合进行汇总，并提供减少的证据（p  = 0.021，Fisher 方法，偏倚风险：低）。这并不适用于敏感性分析 ( p  = 0.053)。对于二次转诊，荟萃分析显示非显着比值比为 1.07（95% CI [0.55, 2.06]，I ²  = 73.0%，n  = 4 项研究，推荐评估分级，开发和评估 [GRADE] 证据确定性： 非常低）。

结论

很少有 RCT 用于干预以提高符合指南的脊柱手术率或转诊率。孤立的临床医生教育可能不会有效。未来的随机对照试验应考虑组织和/或政策层面的干预措施。

PROSPERO注册

CRD42020215137。