Uncertainty method improved on best–worst case analysis in a binary meta-analysis

Gamble, Carrol; Hollis, Sally

doi:10.1016/j.jclinepi.2004.09.013

Cited by 228 publications

(110 citation statements)

References 18 publications

(23 reference statements)

Supporting

Mentioning

107

Contrasting

Unclassified

Order By: Relevance

“…6,35 This method imputes worst case values (e.g., imputing death for a binary survival outcome variable for all experimental treatment subjects with a censored outcome) for those data that are missing in an effort to define the range of possible results under a ''worst case scenario.'' 36 This technique is not intended to result in a likely valid data set, but instead to demonstrate that censoring could not have qualitatively affected the study outcome, when this is indeed the case.…”

Section: Weighted Complete-case Analysismentioning

confidence: 99%

Advanced Statistics: Missing Data in Clinical Research—Part 1: An Introduction and Conceptual Framework

Haukoos

Newgard

2007

Academic Emergency Medicine

163

View full text Add to dashboard Cite

Missing data are commonly encountered in clinical research. Unfortunately, they are often neglected or not properly handled during analytic procedures, and this may substantially bias the results of the study, reduce study power, and lead to invalid conclusions. In this two-part series, the authors will introduce key concepts regarding missing data in clinical research, provide a conceptual framework for how to approach missing data in this setting, describe typical mechanisms and patterns of censoring of data and their relationships to specific methods of handling incomplete data, and describe in detail several simple and more complex methods of handling such data. In part 1, the authors will describe relatively simple approaches to handling missing data, including complete-case analysis, available-case analysis, and several forms of single imputation, including mean imputation, regression imputation, hot and cold deck imputation, last observation carried forward, and worst case analysis. In part 2, the authors will describe in detail multiple imputation, a more sophisticated and valid method for handling missing data.ACADEMIC EMERGENCY MEDICINE 2007; 14:662-668 ª

show abstract

Section: Weighted Complete-case Analysismentioning

confidence: 99%

Advanced Statistics: Missing Data in Clinical Research—Part 1: An Introduction and Conceptual Framework

Haukoos

Newgard

2007

Academic Emergency Medicine

163

View full text Add to dashboard Cite

show abstract

“…In the analysis using the extremes favoring air, the potential for great benefit is seen (typical RR 0.33, 95% CI 0.22–0.50), whereas the analysis using the extremes favoring 100% oxygen suggests the potential for great harm (typical RR 4.09, 95% CI 2.63–6.38). A more conservative and realistic estimate of the uncertainty can be derived using the methods of Gamble and Hollis [16]. This estimate suggests that we still have great uncertainty regarding the effect of using air for resuscitation, with the 95% CI for the risk of abnormal neurodevelopmental assessment ranging from 0.41 to 2.28 (table 4).…”

Section: Resultsmentioning

confidence: 99%

“…The interval for the best/worst analysis was calculated as the extremes from the best- and worst-case scenarios. Following the method of Gamble and Hollis [16], uncertainty intervals were calculated from the best/worst cases for each study. These were then pooled using the inverse variance weighting method to obtain a pooled uncertainty interval.…”

Section: Methods and Patientsmentioning

confidence: 99%

Neurodevelopmental Outcome of Infants Resuscitated with Air or 100% Oxygen: A Systematic Review and Meta-Analysis

et al. 2012

View full text Add to dashboard Cite

Background: The use of air for the initial resuscitation of newborn infants has been shown to reduce neonatal mortality. However, a precise estimate of the neurodevelopmental status upon follow-up of infants resuscitated in air is lacking. Objective: To perform a meta-analysis of all studies reporting resuscitation of newborn infants with air or 100% oxygen that included follow-up data. Methods: Bibliographic databases were searched. In addition, we estimated the effect of loss to follow-up on our analysis of abnormal neurodevelopmental outcome. Results: We identified 10 studies in which newborn infants had been randomly or quasi-randomly assigned to resuscitation with air or 100% oxygen. Three of these 10 studies had available follow-up data. A total of 678 infants were enrolled at centers that performed follow-up of these infants. Of these, 113 died, leaving 565 infants potentially eligible for follow-up. A total of 414 children were evaluated (73% of eligible children; 195 resuscitated with air and 219 with 100% oxygen). In the air group, 12.8% of infants had an abnormal neurodevelopmental outcome, compared with 10.5% in the 100% oxygen group [typical relative risk (RR) 1.24, 95% confidence interval 0.73–2.10]. This is consistent with an RR of abnormal development as low as 0.41 or as high as 2.28. Conclusions: Long-term follow-up did not detect any significant differences in these two groups regarding abnormal development. However, the results are imprecise and could be consistent with significant harm or benefit.

show abstract

“…We will provide reasons for the missing data in the narrative summary where these are available. If sufficient information is available, we plan to assess the extent to which the results of the review could be altered by the missing data by, for example, a sensitivity analysis based on consideration of 'best-case' and 'worst-case' scenarios (Gamble 2005). Here, the 'best-case' scenario is that where all participants with missing outcomes in the experimental condition had good outcomes and all those with missing outcomes in the control condition had poor outcomes; the 'worst-case' scenario is the converse (Higgins 2009; section 16.1.2).…”

Section: Cross-over Trialsmentioning

confidence: 99%

Pharmacological interventions for paranoid personality disorder

Völlm¹,

Farooq²,

Jones³

et al. 2014

Cochrane Database of Systematic Reviews

View full text Add to dashboard Cite

Uncertainty method improved on best–worst case analysis in a binary meta-analysis

Cited by 228 publications

References 18 publications

Advanced Statistics: Missing Data in Clinical Research—Part 1: An Introduction and Conceptual Framework

Advanced Statistics: Missing Data in Clinical Research—Part 1: An Introduction and Conceptual Framework

Neurodevelopmental Outcome of Infants Resuscitated with Air or 100% Oxygen: A Systematic Review and Meta-Analysis

Pharmacological interventions for paranoid personality disorder

Contact Info

Product

Resources

About