Examining Heterogeneity in Meta-Analysis

Furlan, Andrea D.; Tomlinson, George; Jadad, Alejandro; Bombardier, Claire

doi:10.1097/brs.0b013e31816233b5

Cited by 29 publications

(4 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The study design is not the most important factor, even the NRCTs which should be seemed as a supplement for RCTs can provide a same significance of RCTs. [ 30 , 31 ] The reason why the result is presented like this may be the RCTs which have a too small number, the NRCTs can provide a low evidence in our opinion. But whether the rate of morbidity in no drainage group is lower than the routine drainage group is still debated and it needs more high-quality RCTs, which compare the without drainage with routine drainage in the future.…”

Section: Discussionmentioning

confidence: 88%

Is peritoneal drainage essential after pancreatic surgery?

et al. 2017

View full text Add to dashboard Cite

Aim:Our objective is to assess the function of peritoneal drainage, which is placed after pancreatic surgery.Background:With the medical advancement some study put forward that peritoneal drainage is not the necessary after pancreatic surgery; it cannot improve the complications of postoperation even leading to more infection and so on. However, there is no one study can clear and definite whether omitting the drainage after surgery or not.Method:Searching databases consist of all kinds of searching tools, such as Medline, The Cochrane Library, Embase, PubMed, etc. All the included studies should meet our demand of this meta-analysis. In the all interest outcomes blow we take the full advantage of RevMan5 to assess, the main measure is odds ratio (OR) with 95% confidence, the publication bias are assessed by Egger test and Begg test.Result:The rate of postoperative pancreatic fistula (POPF) in no drainage group is much lower than that in routine drainage group (OR = 0.47, I2 = 43%, P < .00001). The result of the 2 randomized controlled trials (RCTs) in this pool are almost accord with the former (OR = 0.57, I2 = 0%, P = .05). In subgroup the result suggest that the peritoneal drainage can increase the morbidity (OR = 0.71, I2 = 15%, P = .0002) after pancreaticoduodenectomy (PD), but reduce the mortality (OR = 1.92, I2 = 8%, P = .03) after PD. In distal pancreatectomy (DP) the rate of POPF and clinically relevant pancreatic fistula (CR-PF) is lower without drainage; there is no significant difference in the CR-PF, hospital stay, intra-abdominal abscess, radiologic invention, and the reoperation.Conclusion:In the current meta-analysis, we cannot make a clear conclusion whether to abandon the routine drainage or not, but from the subgroup we can see something is safer than nothing to routine peritoneal drainage. And the patients who underwent DP can attempt to omit the drainage. But it still needs more RCTs to assess the necessity of drainage.

show abstract

Section: Discussionmentioning

confidence: 88%

Is peritoneal drainage essential after pancreatic surgery?

et al. 2017

View full text Add to dashboard Cite

show abstract

“…(5) Of the 3 studies that were described as finding a difference, 2 indicated lower estimates from observational studies(17,18) and 1 indicated higher estimates from observational studies. (19) None of the studies included in the review appeared to employ the selective inversion rule used by Hemkens et al, but their use of the ROR to quantify bias is just as flawed and dependent on the direction of comparison chosen for each study by investigators. In addition, several of the earliest papers comparing observational studies and RCTs used simple graphical displays without any quantitative assessment of differences.…”

Section: Discussionmentioning

confidence: 99%

A Bias in the Evaluation of Bias Comparing Randomized Trials with Nonexperimental Studies

Franklin

Dejene

Huybrechts

et al. 2017

Epidemiologic Methods

View full text Add to dashboard Cite

In a recent BMJ article, the authors conducted a meta-analysis to compare estimated treatment effects from randomized trials with those derived from observational studies based on routinely collected data (RCD). They calculated a pooled relative odds ratio (ROR) of 1.31 (95% confidence interval [CI]: 1.03–1.65) and concluded that RCD studies systematically over-estimated protective effects. However, their meta-analysis inverted results for some clinical questions to force all estimates from RCD to be below 1. We evaluated the statistical properties of this pooled ROR, and found that the selective inversion rule employed in the original meta-analysis can positively bias the estimate of the ROR. We then repeated the random effects meta-analysis using a different inversion rule and found an estimated ROR of 0.98 (0.78–1.23), indicating the ROR is highly dependent on the direction of comparisons. As an alternative to the ROR, we calculated the observed proportion of clinical questions where the RCD and trial CIs overlap, as well as the expected proportion assuming no systematic difference between the studies. Out of 16 clinical questions, 50% CIs overlapped for 8 (50%; 25 to 75%) compared with an expected overlap of 60% assuming no systematic difference between RCD studies and trials. Thus, there was little evidence of a systematic difference in effect estimates between RCD and RCTs. Estimates of pooled RORs across distinct clinical questions are generally not interpretable and may be misleading.

show abstract

“…The risk of bias of the included studies will be assessed according to the Furlan checklist, [ 27 , 28 ] which includes 7 items: (A) Was the method of randomization adequate? (B) As the treatment allocation concealed?…”

Section: Methodsmentioning

confidence: 99%