AAOS Evidence-Based Practice Committee
Recommendations in Clinical Practice Guidelines
• One MODERATE or 1 or more LOW quality studies
Please rate topic.
Average 3.8 of 21 Ratings
A physician is interested in using platelet-rich plasma (PRP) for treatment of osteochondral lesions of the talus. He is reviewing a prospective cohort study that compares 40 patients treated with PRP and cast immobilization for 6 weeks vs. 36 patients treated conservatively with cast immobilization for 6 weeks. All patients were treated at the same time and institution. The study was not randomized although treatment and control groups were matched appropriately to reduce selection bias. Follow-up in each group was >80% over 1 year. The paper reported significant improvement with use of PRP based on three standard foot and ankle outcome scores (AOFAS, SF-36, FOAS). What is the level of evidence for this study?
Select Answer to see Preferred Response
This is prospective cohort study with Level-II evidence.
Level of evidence provides guidance to the study quality. It is used to assess therapeutic studies (as with this question), prognostic studies, diagnostic studies and economic or decision models. When determining the level of evidence, readers must critically appraise the study question, treatment, intervention and outcomes of the study design. Level-II therapeutic studies consist of well-designed prospective cohort studies, poor-quality randomized controlled trials (follow-up less than 80%) and systematic review of Level-II studies or non-homogenous Level-I studies.
Wright et al. provided an excellent summary of clinical research study level of evidence. This has been provided as Illustration A.
Illustration A shows a chart of level of evidence. There is a column for each type of study which corresponds to a row that outlines the level of evidence based on study design.
Answer 1: Level-I evidence include randomized controlled studies with follow-up>80% and systematic review of Level-I RTC studies (homogenous studies)
Answer 3: Level-III evidence include case control studies, retrospective cohort studies and systematic review of Level-III studies
Answer 4: Level-IV evidence include case series with no control group (or compare to a historical control group)
Answer 5: Level V evidence include expert opinion
Wright JG, Swiontkowski MF, Heckman JD.
J Bone Joint Surg Am. 2003 Jan;85-A(1):1-3. PMID: 12533564 (Link to Abstract)
Please rate question.
Average 3.0 of 6 Ratings
A therapeutic study presents a systematic review of 15 high-quality randomized controlled trials with homogeneous results. What level of evidence is this considered?
A systematic review of high-quality clinical trials is considered a Level I study.
A systematic review is a powerful tool used to identify, evaluate and appraise all high-quality research related to a specific question. Systematic reviews, in contrast to most narrative review articles, adhere to strict scientific design by following eight steps; 1) defining a question and developing inclusion criteria, 2) searching for studies, 3) selecting studies and collecting data, 4) assessing risk of bias, 5) data analysis and meta-analyses, 6) addressing biases, 7) presenting results, and 8) interpreting results and drawing conclusions. When the results from the systematic review are homogeneous (less variability between studies than would be expected by chance), the data from a systematic review can be combined into a meta-analysis.
Wright et al. discuss the levels of evidence in orthopaedic journals as presented by the AAOS Evidence-Based Practice Committee. Based on levels of evidence, the AAOS provides grades of recommendation (A, B, C, I). Grade A recommendations are supported by Level 1 studies with consistent findings, whereas Grade I do not have enough evidence to support a recommendation.
Illustration A shows an "evidence pyramid." MA = meta-analysis, SR = systematic review, RCT = randomized controlled trial.
Answer 2: An example of a Level II study is prospective comparative study.
Answer 3: An example of a Level III study is a case control trial.
Answer 4: An example of a Level IV study is a case series.
Answer 5: An example of a Level V study is one based on expert opinion.
Average 1.0 of 9 Ratings
An orthopaedic resident wants to answer a focused research question of whether mobile bearing knee arthroplasty has superior functional outcomes compared to fixed bearing knee arthroplasty. The resident mathematically combines the results from multiple retrospective cohort studies following QUORUM (Quality of Reporting of Meta-analyses) guidelines. What is the highest level of evidence that this meta-analysis can achieve?
The level of evidence assigned to a meta-analysis is based on the lowest level of evidence of the included studies. In this case, the studies included in the meta-analysis were retrospective cohort (Level III) studies.
A meta-analysis is a systematic review that combines the results of multiple studies to answer a focused clinical question.
Clarke discusses the QUORUM guidelines which are intended to address standards for improving the quality of reporting of meta-analyses of clinical randomized controlled trials. The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement and MOOSE (Meta-analysis Of Observational Studies in Epidemiology) guidelines are similar standards. A similar set of guidelines called CONSORT (Consolidated Standards of Reporting Trials) guidelines are available for randomized trials.
1: Level I studies include high-quality randomized controlled trials
2: Level II studies include lower-quality RCTs and prospective cohort studies.
4: Level IV studies include case-series or poor-quality cohort and case-control studies.
5: Level V studes are expert opinion articles.
Lancet. 2000 Feb 26;355(9205):756-7. PMID: 10703836 (Link to Abstract)
Average 4.0 of 5 Ratings
In the study by Moseley et al published in the New England Journal of Medicine, 180 patients with knee osteoarthritis were randomly assigned via sealed envelope to receive arthroscopic débridement, arthroscopic lavage, or placebo surgery. Outcomes were assessed by blinded evaluators at several points over a 2 year period with the use of five self-reported pain and function scores. There was a greater than 90% follow-up in the study. This study is best described as having which level of evidence?
Therapeutic study, evidence level I
Therapeutic study, evidence level II
Diagnostic study, evidence level I
Diagnostic study, evidence level II
Prognostic study, evidence level I
Therapeutic studies investigate the results of treatment. Level 1 evidence includes randomized controlled trial with statistically significant difference or no statistically significant difference but narrow confidence interval
A level II therapeutic study would include a lesser-quality randomized controlled trial (eg, <80% follow-up, no blinding, or improper randomization). A non-randomized prospective cohort study that compares the results of treatment or systematic reviews of level II studies or level I studies with heterogenous results would also be considered a level II study.
Prognostic studies investigate the effect of a patient characteristic on the outcome of disease. Diagnostic studies investigate the efficacy of a diagnostic test. Illustration A is a table detailing the levels of evidence.
Moseley JB, O'Malley K, Petersen NJ, Menke TJ, Brody BA, Kuykendall DH, Hollingsworth JC, Ashton CM, Wray NP
N. Engl. J. Med.. 2002 Jul;347(2):81-8. PMID: 12110735 (Link to Abstract)
Average 2.0 of 15 Ratings
Which of the following study designs represent a level III evidence study?
Prospective, randomized controlled trial
Retrospective case-control study
Retrospective case series
Prospective cohort study
The practice of evidence based medicine means integrating individual clinical expertise with the best available external clinical evidence from systematic research. Therapeutic study hierarchy of evidence has been established to better analyze studies in a reproducible fashion. Level I studies include well-designed randomized controlled prospective studies (RCT). Level II include lower quality designed prospective RCT as well as prospective cohort studies. Level III include retrospective cohort studies and case-control studies. Level IV include case series. Level V include case reports, expert opinion, and personal observation. This is summarized in illustration A. The referenced article by Brighton et al is a review of how the level of evidence has evolved and how the different levels can carry varied amounts of impact on clinical treatments and future research.
Brighton B, Bhandari M, Tornetta P 3rd, Felson DT.
Clin Orthop Relat Res. 2003 Aug;(413):19-24. PMID: 12897592 (Link to Abstract)
Average 3.0 of 11 Ratings
Using levels of evidence in research studies, which of the following represents a level II study?
Retrospective case control study
Case report of 3 patients with the same disease
High-quality randomized prospective clinical trial
The opinion of a review panel at the annual AAOS meeting
Level II Evidence includes prospective cohort studies. There are 5 levels of evidence in clinical research. Level I includes randomized clinical trials. Level II includes prospective cohort studies. Level III includes case control studies AND retrospective cohort studies. Level IV includes case series. Level V evidence includes expert opinion, case reports, and personal observation. The cited reference by Bernstein et al is an excellent review of Evidence Based Medicine (EBM) and describes the 5 step process of making medical decisions based on the available, and often limited, evidence. These steps include: 1) Formulate answerable questions. 2) Gather the evidence. 3) Appraise the evidence. 4) Implement the evidence. 5) Evaluate the process. The reference by Spindler et al is another review article that discusses the different levels of evidence for different study types.
J Am Acad Orthop Surg. 2004 Mar-Apr;12(2):80-8. PMID: 15089081 (Link to Abstract)
Spindler KP, Kuhn JE, Dunn W, Matthews CE, Harrell FE, Dittus RS
J Am Acad Orthop Surg. 2005 Jul-Aug;13(4):220-9. PMID: 16112979 (Link to Abstract)
Average 3.0 of 18 Ratings