Tag Archive for: Philipp Dahm

Posts

Reaching for the stars – rating the quality of systematic reviews with the Assessment of Multiple Systematic Reviews (AMSTAR) 2

The number of published systematic reviews and meta‐analyses in the urological literature has dramatically increased in recent years [1]. This is good news given their importance in guiding clinical decision‐making, guideline development and health policy. However, many of these studies are of low quality, raising concerns about the trustworthiness of their results. As with other research studies, it is therefore important for readers to have a framework for determining the quality of a given systematic review. Therefore, in 2017 BJU International launched a scoring system for systematic reviews that provides readers with a summary assessment as to whether established methodological safeguards against bias for systematic reviews have been met [2]. This is based on the Assessment of Multiple Systematic Reviews (AMSTAR), a validated instrument that assesses methodological quality on an 11‐point scale (0–11), with higher scores reflecting greater methodological rigor and all criteria being given the same relative weight [3].

Recently, an updated version of this instrument has become available, offering a better assessment of systematic reviews [4]. The revised instrument (AMSTAR 2) includes 10 of the original domains; it has 16 items in total (compared with 11 in the original), simpler response categories to the original AMSTAR, and provides an overall rating that is largely based on seven critical domains that should all be met. These relate to: (i) documentation of an a priori registered protocol in Prospective Register of Systematic Reviews (PROSPERO) or through Cochrane, (ii) a comprehensive literature search, (iii) explicit justification for excluding studies, (iv) a risk of bias assessment of included studies, (v) appropriate use of meta‐analytical methods, (vi) consideration of risk of bias when interpreting the results of the review, and (vii) assessment of presence and likely impact of publication bias. Other, non‐critical domains include a clear description of the study question in Population, Intervention, Comparison, Outcome (PICO) format, study selection and data extraction in duplicate, and identification of sources of funding of the studies included in the review and the review itself. This results in a four‐tiered rating (high, moderate, low, and critically low) that reflects the confidence that a reader may place in the results. Notably, a high‐quality rating requires no critical weakness and allows for only one non‐critical weakness. More than one non‐critical weakness drops the rating down to moderate, and just one critical weakness (such as lack of an a priori protocol) drops the rating down to low. Any review that has more than one critical weakness will be rated as critically low.

BJU International editors will routinely apply this AMSTAR 2‐based scoring system to screen for methodological quality in order to raise the awareness of this issue and promote reviews of higher quality (Fig. 1)[1]. Needless to say, BJU International is not the place for systematic reviews of sub‐optimal methodological quality in which the readers cannot place their trust. Meanwhile, we also fully understand that methodological rigor is not everything but has to be paired with clinical relevance and newsworthiness. Much has been written about the dramatic redundancy of systematic reviews on the same topic; in certain areas of medicine, the number of systematic reviews exceeds that of eligible studies that these reviews included [5]. Therefore, when systematic reviews already exist, there needs to be a clear rationale for any ‘encore’ performance. BJU International also encourages the development of systematic reviews by author teams that are financially unconflicted and have thoughtfully managed any intellectual conflict of interest.

Figure 1: New BJUI rating system of systematic reviews based on AMSTAR 2. The number of coloured stars in the inner and outer layers of the system represents completeness of an individual critical domain and overall confidence rating of the systematic review, respectively. The number in the middle of the system refers to the summary AMSTAR 2 score based on the overall confidence rating of the systematic review (high: 4, moderate: 3, low: 2, critically low: 1).

Through this initiative, BJU International not only intends to become the premier journal for high‐quality systematic reviews as they relate to urology, but also to move the field forward, reducing redundancy and waste. As we embrace the higher standards of AMSTAR 2, we present the first review to be scored using this method in this issue [6] and we encourage all systematic review authors to accept this challenge and reach with us for the stars.

References

  1. Han JL, Gandhi S, Bockoven CG, Narayan VM, Dahm P. The landscape of systematic reviews in urology (1998 to 2015): an assessment of methodological quality. BJU Int 2017; 119: 638–49
  2. Dahm P. Raising the bar for systematic reviews with Assessment of Multiple Systematic Reviews (AMSTAR). BJU Int 2017; 119: 193
  3. Shea BJ, Grimshaw JM, Wells GA et al. Development of AMSTAR: a measurement tool to assess the methodological quality of systematic reviews. BMC Med Res Methodol 2007; 7: 10

 

About the authors:

Dr Philipp Dahm is Professor of Urology and Vice Chair of Veterans Affairs at the University of Minnesota. He also serves as Director of Research and Education for Surgical Services at the Minneapolis Veterans Administration Medical Center (@EBMUrology).

 

Dr Jae Hung Jung is from the Department of Urology, Wonju College of Medicine, Yonsei University, Korea.

 

 

 

The challenge with systematic reviews of non-randomised studies in urology

In this issue, BJU International has made the conscious decision to publish a systematic review (SR) and meta-analysis (MA) by Guo et al. [1]to inform the question of whether patients undergoing nephrouretectomy for upper tract urothelial carcinoma are at increased risk of worse oncological outcomes. This question was also the topic of a similar review by Marchioni et al. [2] published earlier this year in this journal. Both studies were submitted around the same time and underwent independent, parallel peer review that resulted in different editorial decisions. Given their similarity in methodological quality they both deserved similar consideration for publication, which the journal is hereby honouring.

At the same time, this provides the unique opportunity to reflect on methodological developments in the field and BJU International‘s efforts to raise the bar of the methodological quality of SRs, which include the provision of an Assessment of Multiple Systematic Reviews (AMSTAR) rating [3]. AMSTAR is a validated tool to assess the components of a SR on an 11-point scale (0–11), with higher scores reflecting higher methodological rigor. An updated version of this tool has recently been provided, which offers greater clarity in interpretation [4]. Another related instrument that has become available is the Risk of Bias in Systematic Reviews (ROBIS), which assesses the study limitations in SRs (i.e., the relevance of the review, concerns with the review process, and potential bias introduced during the review) [5]. Meanwhile, while it would be premature to claim success, it is our impression that BJU International’s initiative to provide AMSTAR ratings is making a valuable contribution in raising awareness for such methodological issues and improving the transparency of published reviews.

As BJU International takes a lead in promoting high-quality SRs in urology, the journal has seen a considerable increase in the number of submissions, including SRs of non-randomised studies (NRS). Whilst much of what we practice on a day-to-day basis is based on evidence from NRS, studies of those designs have infrequently been included in the Cochrane Library, which has pioneered much of the underlying methodology. This is for a few reasons: First, the ‘garbage in–garbage out’ phenomenon; if the underlying individual studies only provide very low-quality evidence, combining these studies will rarely enhance the confidence we place in their results. Second, the need for methodological advances in the assessment and analyses of NRS. Third, when high-quality evidence from randomised controlled trials (RCTs) is available, it may be inefficient to review the NRS literature.

However, progress is being made on the methodology front. Members of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Group are credited with having developed an approach for rating the quality or certainty of evidence from randomised and NRS to inform decision making [6]. While a body of evidence from RCTs, which starts as high-quality evidence, may be downgraded for study limitations, a body of evidence from NRS, which starts as low-quality evidence, may be upgraded for one of three reasons, most commonly for large magnitude of effect [7]. The underlying assumption is that, whilst we have to assume that bias is likely to be present in these studies, it is unlikely to explain the entire observed effect.

It nevertheless remains critical to assess the risk of bias of NRS. While the Newcastle-Ottawa scale (as used in both of these SRs) is a widely used instrument to evaluate risk of bias in NRS, it has critical limitations that the recently developed Risk Of Bias In Non-randomised Studies – of Interventions (ROBINS-I) seeks to overcome [8]. ROBINS-I evaluates NRS by using a standardised comparison to an RCT (i.e. target trial) [9]. In this way, ROBINS-I captures the bias inherent to studies without proper randomisation or allocation concealment, namely the lack of a balance of known and unknown confounders and selection of participants. ROBINS-I allows users to fundamentally start all studies at the same quality level, providing the transparency requested by some SR authors conducting SRs of NRS.

While ureterorenoscopy before nephroureterectomy may indeed increase the risk of intravesical recurrence as the authors suggest, additional exploration would be needed to make a statement about the causality of the relationship. Guo et al. [1] conducted sensitivity analyses to describe the potential for bias introduced by confounders of previous bladder tumour history and bladder-cuff management, thereby increasing our confidence that the observed effect may be closer to the truth. It seems equally important to note that Guo et al. found no increased risk in cancer-specific, recurrence-free or overall survival, which are other outcomes of potentially greater patient importance.

Understanding the inherent limitations of NRS, and placing their findings into appropriate clinical context are critical to the conduct of SRs. Moving forward, BJU International will continue to seek out the highest quality reviews that make use of the best, up-to-date methodology. We hope that these efforts will both serve as a beacon for the research community, but more importantly, result in improved evidence-based care for our patients.

Philipp Dahm*, Jae Hung Jung† and Rebecca L. Morgan
*Department of Urology, Minneapolis VA Medical Center, University of Minnesota, Minneapolis, MN, USADepartment of Urology, Yonsei University Wonju College of Medicine,
Wonju, Korea and Department of Health Research MethodsEvidence, and Impact, McMaster University, Hamilton, ON, Canada

 

References

 

 

 

3 Dahm P. Raising the bar for systematic reviews with Assessment of Multiple Systematic Reviews (AMSTAR). BJU Int 2017; 119: 193

 

 

5 Whiting P, Savovic J, Higgins JP et al. ROBIS: a new tool to assess risk of bias in systematic reviews was developed. J Clin Epidemiol 2016; 69: 22534

 

6 Guyatt GH, Oxman AD, Vist GE et al. GRADE: what is quality of evidence and why is it important to clinicians? BMJ 2008; 336: 9958

 

7 Guyatt GH, Oxman AD, Sultan S et al. GRADE guidelines: 9. Rating up the quality of evidence. J Clin Epidemiol 2011; 64: 13116

 

8 Deeks JJ, Dinnes J, DAmico R et al. Evaluating non-randomised intervention studies. Health Technol Assess 2003; 7: iiix, 1173

 

9 Sterne JA, Hernan MA, Reeves BC et al. ROBINS-I: a tool for assessing risk of bias in non-randomised studies of interventions. BMJ 2016; 355: i4919

 

February Editorial: Raising the bar for systematic reviews with Assessment of Multiple Systematic Reviews (AMSTAR)

The BJUI has a longstanding track record in promoting the dissemination of high-quality unbiased evidence and helping their readership to understand why the principles of evidence-based medicine matter. This devotion is witnessed by the work that goes into every issue of the journal, as well as past initiatives such as providing a level of evidence rating for clinical research articles or publishing educational articles such as the ‘Evidence-Based Urology in Practice’ series [1, 2].

Major foci for clinically oriented specialty journals are systematic reviews and meta-analyses. Systematic reviews have a preeminent role in guiding the practice of evidence medicine by addressing focused clinical questions in a systematic, transparent and reproducible manner. Defining criteria of a high-quality systematic review include: an a priori registered protocol, a comprehensive search of multiple sources including unpublished studies (to avoid publication bias), an assessment of the quality of evidence that goes beyond study design alone, and a thoughtful interpretation of the findings. Systematic reviews inform clinicians and patients at the point of care, form the foundation of evidence-based clinical practice guidelines, and help shape health policy [3]. They also find frequent citation and can raise a journal’s impact factor. There is therefore more than one good reason for journals to care about the quality of systematic reviews.

Meanwhile, a study in this issue of the BJUI [4] shows that the methodological quality of systematic reviews published in the urological literature is modest, varies substantially, and has failed to improve over time. This contrasts to randomised controlled trials’ reporting quality that appears to have improved substantially over time, probably due to increased awareness among clinical researchers, urology readers and journal reviewers [4, 5]. The study [4] used the Assessment of Multiple Systematic Reviews (AMSTAR), a validated 11-item instrument, to measure the methodological quality of systematic reviews with higher scores reflecting better quality.

The authors [4] surveyed four major urological journals and compared the periods 2013–2015 to 2009–2012 and 1998–2008. Despite a dramatic increase in the number of systematic reviews published each year, methodological quality has stagnated with mean AMSTAR scores ± standard deviations of 4.8 ± 2.4 (2013–2015; = 125), 5.4 ± 2.3 (2009–2012; = 113) and 4.8 ± 2.0 (1998–2008; = 57). The average systematic review therefore has deficits in over half the 11 AMSTAR criteria and is of only modest quality thereby undermining our confidence in their results. Although the mean AMSTAR score of 5.6 ± 2.9 for 25 systematic reviews published in the BJUI in 2013–2015 compared favourably to similar studies in other leading urology journals, the difference was not statistically significant.

What are we going to do about it? Inspired by these findings, the BJUI is launching a new initiative to raise awareness for the issue of methodological quality of systematic reviews among its readership and raise the bars for its contributors. Future systematic review authors will be asked to submit an AMSTAR-based checklist to provide enhanced transparency about its methods that will be reviewed as part of the editorial review process. These include documentation of an a priori written protocol and ideally, registration of the systematic review through the Cochrane Collaboration or the Prospective Register of Systematic Reviews (PROSPERO). Such a protocol should outline all important steps of the review process including the definition of outcomes, study inclusion and exclusion criteria, details about the literature search, study selection and data abstraction process, analytical approach including planned sensitivity and subgroup analyses. Authors should also rate the quality of evidence looking beyond study limitation alone by using an approach such as the Grading of Recommendations Assessment, Development, and Evaluation (GRADE), which recognises such additional domains such as imprecision, inconsistency, indirectness and publication bias [6]. Critical steps of the systematic review process should be completed in duplicate to guard against random and systematic error and authors should provide readers with the information about who funded the studies included in the review, as well as their own potential conflicts of interests. To guard against publication bias, systematic review authors should also search for ongoing trials and unpublished studies through registries and abstract proceedings.

It is understood that the methodological handiwork that goes into the planning, execution and reporting of a systematic review do not assure clinical relevance or newsworthiness, nor does it address any issues surrounding the limited quality of studies that the review may be summarising. However, it is nevertheless a sine quae no to assure readers that they can be confident of the results. The new BJUI initiative will raise awareness for the issue of systematic review quality by providing a summary AMSTAR score to accompany each article. We hope that with this initiative we will provide a beacon for other specialty journals to follow, with the goal of raising the bar for all published systematic reviews and ultimately leading to improved patient care.

Philipp Dahm

 

Department of Urology, Minneapolis Veterans Administration Health Care System and University of Minnesota , MinneapolisMN, USA


References

 

1 Dahm P, Preminger GM. Introducing levels of evidence to publications in urology. BJU Int 2007; 100: 2467

 

 

 

4 HanJL, Gandhi S, Bockoven CG, Narayan VM, Dahm PThe landscape osystematic reviews in urology (1998 to 2015): an assessment of methodological quality. BJU Int 2016 [Epub ahead of print]. doi: 10.1111/bju.13653.

 

5 Narayan VM, Cone EB, Smith D, Scales CD Jr, Dahm P. Improved reporting of randomized controlled trials in the urologic literature. Eur Urol 2016; 70: 10449

 

6 Guyatt GH, Oxman AD, Vist GE et al. What is quality of evidence and why is it important to clinicians? BMJ 2008; 336: 9958

 

Editorial: The Jury on Posterior Muscolofascial Reconstruction is still out

In their systematic review and meta-analysis, Grasso et al. [1] address the question of whether posterior muscolofascial reconstruction (PMR), the so-called Rocco stitch, positively affects urinary continence after radical prostatectomy. The relevance of the question to this structured form of inquiry is that individual studies to date have been inconclusive. We recognize Sir Archie Cochrane, who gave his name to the Cochrane Collaboration that pioneered the methods for conducting systematic reviews, for emphasizing the critical importance of looking at the entire body of evidence in a structured manner when seeking to answer a clinical question [2]. In the present study, which included both randomized controlled trials (RCTs) and observational studies of variable methodological quality, a favourable impact of PMR across all postoperative time points (3–7 days, 30 days, 3 and 6 months) was observed. The effect was most pronounced early on at the time of catheter removal, when the patients undergoing PMR were nearly twice as likely as the control group (risk ratio 1.9; 95% CI 1.3–2.9) to be continent, thereby suggesting a major benefit of this approach. It should be noted, however, that this analysis was dominated by the observational studies, particularly retrospective observational studies, which offer the least degree of methodological rigor.

Even more important, therefore, than the act of pooling across studies is the rating of the quality of evidence for the body of evidence on an outcome-specific basis. Based on the GRADE approach, which has become the most widely endorsed framework for rating the quality of evidence, we would initially place a high and low level of confidence in a body of evidence drawn from RCTs and observational studies, respectively [3]. As a result, one might plan a separate analysis of those two groups of studies first, and only move to pool them if their results were similar. In this case, the results from the RCTs and observational studies were different, with prospective and retrospective studies reporting larger, probably exaggerated effect sizes; however, it is also understood that other aspects such as study limitation (risk of bias), inconsistency, impression, indirectness and risk publication bias may lower our confidence in the effect estimates from RCTs [4]. Focusing on the body of evidence from RCTs alone (Table 1) we have ‘moderate’ confidence that PMR may not improve early continence at the time of catheter removal. Similarly, the few RCTs that contributed to the assessment of continence at later timepoints do not provide evidence that continence is affected favourably, although our confidence for those outcomes is only ‘low’ or ‘very low’, suggesting that future trials may change these estimates of effect. Meanwhile, it should be noted that none of the RCTs appeared to provide information on the potential downsides of PMR, such as rates of urinary retention or bladder neck contracture. As a result, enough uncertainty remains to state that the jury on PMR is still out; this is consistent with the authors’ call for a future high-quality trial, which is reportedly ongoing. While PMR is already widely used by open and robot-assisted prostatectomy surgeons around the globe, this example sheds light on current evidentiary standards of surgical innovation. Following the IDEAL recommendations, it would be much preferred if the urological community committed to well designed trials for novel surgical approaches and device-dependent interventions up front, before moving to widespread dissemination [5].

JulEOTW2

Read the full article

Philipp Dahm
Department of Urology, Minneapolis VA Health Care System, Urology Section 112D and University of Minnesota, Minneapolis, MN, USA

 

References

 

 

2 Hajebrahimi S, Dahm P, Buckingham J. Evidence-based urology in practice: the cochrane library. BJU Int 2009; 104: 10489

 

3 Caneld SE, Dahm P. Rating the quality of evidence and the strength of recommendations using GRADE. World J Urol 2011; 29: 3117

 

4 Guyatt GH, Oxman AD, Vist GE et al. GRADE: what is quality of evidence and why is it important to clinicians? BMJ 2008; 336: 9958

 

5 McCulloch P. The IDEAL recommendations and urological innovation. World J Urol 2011; 29: 3316

 

© 2024 BJU International. All Rights Reserved.