背景:尽管越来越重视循证医学,目前尚不清楚眼科临床实践指南背后的证据.此系统分析的目的是了解支持美国眼科学会(AAO)首选实践模式(PPP)指南的证据水平(LOE)。评估随时间的变化,并比较眼科亚专业的LOE。
方法:全面审查了所有当前的PPP指南及其前身,以确定提供LOE的所有建议(I[随机对照试验],II[病例对照或队列研究],和III[非分析研究])。
结果:24个当前的PPP中有23个是先前版本。在先前版本的PPP中,LOE的建议数量从以前的PPP中的1254个减少到当前的PPP中的94个。使用LOEI的建议数量从114减少到83,LOEII从147减少到2,LOEIII从993减少到9。然而,LOEI建议的比例从9%增加到88%,由低于LOEI的证据报告不成比例地减少所致。按亚专业划分的亚组分析显示出相似的趋势(先前PPPs与当前PPPs中的LOEI建议:视网膜:57[12%]vs19[100%];角膜:33[5%]vs24[100%];青光眼:9[23%]vs17[100%];白内障:13[17%]vs18[100%])。
结论:PPP指南中LOE报告的趋势表明,从2012年到2021年,随机对照试验的证据越来越受到重视。报告的LOE建议数量的下降表明未来指南需要改进的领域,因为LOE的存在对于促进临床实践指南的解释至关重要。
BACKGROUND: Despite the increased emphasis on evidence-based medicine, the current state of evidence behind ophthalmology clinical practice
guidelines is unknown. The purpose of this systematic analysis was to understand the levels of evidence (LOE) supporting American Academy of Ophthalmology (AAO) Preferred Practice Pattern (PPP)
guidelines, assess changes over time, and compare LOE across ophthalmology subspecialties.
METHODS: All current PPP
guidelines and their immediate predecessors were comprehensively reviewed to identify all recommendations with LOE provided (I [randomized controlled trials], II [case-control or cohort studies], and III [nonanalytic studies]).
RESULTS: Twenty-three out of 24 current PPPs had a prior edition. Among the PPPs with a prior edition, the number of recommendations with LOE decreased from 1254 in prior PPPs to 94 in current PPPs. The number of recommendations with LOE I decreased from 114 to 83, LOE II decreased from 147 to 2, and LOE III decreased from 993 to 9. However, the proportion of LOE I recommendations increased from 9 to 88%, driven by a disproportionate decrease in reporting of evidence lower than LOE I. Subgroup analysis by subspecialty showed similar trends (LOE I recommendations in prior PPPs vs current PPPs: retina: 57 [12%] vs 19 [100%]; cornea: 33 [5%] vs 24 [100%]; glaucoma: 9 [23%] vs 17 [100%]; cataract: 13 [17%] vs 18 [100%]).
CONCLUSIONS: Trends in LOE reporting in PPP
guidelines indicate an increasing emphasis on evidence from randomized controlled trials from 2012 to 2021. The decline in the number of recommendations with LOE reported suggests an area for improvement in future guidelines as the presence of LOE is crucial to facilitate interpretation of clinical practice
guidelines.