Covariate adjustment in randomized controlled trials with dichotomous outcomes increases statistical power and reduces sample size requirements

Hernández, Adrian V.; Steyerberg, Ewout W.; Habbema, J. Dik F.

doi:10.1016/j.jclinepi.2003.09.014

Cited by 210 publications

(193 citation statements)

References 31 publications

Supporting

Mentioning

187

Contrasting

Order By: Relevance

“…Additional strategies for improving trial success might include use of prespecified covariate adjustment (37)(38)(39) (e.g., see Jansen and colleagues [28]), larger target sample sizes, and more realistic and conservative treatment effect expectations (40) (Table 3). Additionally, innovative trial designs, such as Bayesian adaptive trials, may be particularly valuable for assessing drug therapies (35,41).…”

Section: Discussionmentioning

confidence: 99%

Outcomes and Statistical Power in Adult Critical Care Randomized Trials

Harhay

Wagner²,

Ratcliffe³

et al. 2014

Am J Respir Crit Care Med

150

136

View full text Add to dashboard Cite

Rationale: Intensive care unit (ICU)-based randomized clinical trials (RCTs) among adult critically ill patients commonly fail to detect treatment benefits.Objectives: Appraise the rates of success, outcomes used, statistical power, and design characteristics of published trials.Methods: One hundred forty-six ICU-based RCTs of diagnostic, therapeutic, or process/systems interventions published from January 2007 to May 2013 in 16 high-impact general or critical care journals were studied.Measurement and Main Results: Of 146 RCTs, 54 (37%) were positive (i.e., the a priori hypothesis was found to be statistically significant). The most common primary outcomes were mortality (n = 40 trials), infection-related outcomes (n = 33), and ventilationrelated outcomes (n = 30), with positive results found in 10, 58, and 43%, respectively. Statistical power was discussed in 135 RCTs (92%); 92 cited a rationale for their power parameters. Twenty trials failed to achieve at least 95% of their reported target sample size, including 11 that were stopped early due to insufficient accrual/ logistical issues. Of 34 superiority RCTs comparing mortality between treatment arms, 13 (38%) accrued a sample size large enough to find an absolute mortality reduction of 10% or less. In 22 of these trials the observed control-arm mortality rate differed from the predicted rate by at least 7.5%.Conclusions: ICU-based RCTs are commonly negative and powered to identify what appear to be unrealistic treatment effects, particularly when using mortality as the primary outcome. Additional concerns include a lack of standardized methods for assessing common outcomes, unclear justifications for statistical power calculations, insufficient patient accrual, and incorrect predictions of baseline event rates.

show abstract

Section: Discussionmentioning

confidence: 99%

Outcomes and Statistical Power in Adult Critical Care Randomized Trials

Harhay

Wagner²,

Ratcliffe³

et al. 2014

Am J Respir Crit Care Med

150

136

View full text Add to dashboard Cite

show abstract

“…The gain in statistical power with the Cox proportional hazards models was expressed as the absolute difference in power and as the potential reduction in required sample size that can be obtained when the most powerful model would have the same power as the least powerful model. 11 Percentage reduction in required sample size was calculated as 100À100 (Z 2 /Z 1 ) 2 , where Z 1 and Z 2 are the Wald statistics of the most and least powerful model, respectively. These measures are independent of the effect estimate, the a-value and sample size.…”

Section: Discussionmentioning

confidence: 99%

“…These measures are independent of the effect estimate, the a-value and sample size. 11 Therefore, we calculated the mean percentage reduction in required sample size for each genotype frequency. All statistical analyses and simulations were performed using the R statistical package (version 2.5.1).…”

Section: Discussionmentioning

confidence: 99%

Cox proportional hazards models have more statistical power than logistic regression models in cross-sectional genetic association studies

et al. 2008

Self Cite

View full text Add to dashboard Cite

Cross-sectional genetic association studies can be analyzed using Cox proportional hazards models with age as time scale, if age at onset of disease is known for the cases and age at data collection is known for the controls. We assessed to what degree and under what conditions Cox proportional hazards models have more statistical power than logistic regression models in cross-sectional genetic association analyses. Analyses were conducted in an empirical study on the association of 65 polymorphisms and risk of coronary heart disease among 2400 familial hypercholesterolemia patients, and in a simulation study that considered various combinations of sample size, genotype frequency, and strength of association between the genotype and coronary heart disease. We applied Cox proportional hazards models and logistic regression models, and compared effect estimates (hazard ratios and odds ratios) and statistical power. In the empirical study, Cox proportional hazards models generally showed lower P-values for polymorphisms than logistic regression models. In the simulation study, Cox proportional hazards models had higher statistical power in all scenarios. Absolute differences in power did depend on the effect estimate, genotype frequency and sample size, and were most prominent for genotypes with minor effects. For example, when the genotype frequency was 30% in a sample with size n ¼ 2000 individuals, the absolute differences were the largest for effect estimates between 1.1 and 1.5. In conclusion, Cox proportional hazards models can increase statistical power in cross-sectional genetic association studies, especially in the range of effect estimates that are expected for genetic associations in common diseases.

show abstract

“…The primary outcome was analysed by logistic regression analysis. Multivariable logistic regression analysis was used to adjust for confounders12. A co‐variable was deemed a confounding variable when it showed a significant relationship with both the RAM distance at 1 month and the presence of incisional hernia in the univariable regression analysis.…”

Section: Methodsmentioning

confidence: 99%

Development of incisional herniation after midline laparotomy

Harlaar

Deerenberg

Dwarkasing³

et al. 2017

BJS Open

View full text Add to dashboard Cite

BackgroundIncisional herniation is a common complication after abdominal surgery associated with considerable morbidity. The aim of this study was to determine whether incisional hernia is an early complication, in order to understand better the aetiology of incisional hernia formation.MethodsThis study involved the secondary analysis of a subset of patients included in a large RCT comparing small and large tissue bites (5 mm every 5 mm, or 1 cm every 1 cm) in patients scheduled to undergo elective abdominal surgery by midline laparotomy. The distance between the rectus abdominis muscles (RAM distance) was measured by standardized ultrasound imaging 1 month and 1 year after surgery. The relationship between the 1‐year incidence of incisional hernia and the RAM distance at 1 month was investigated.ResultsSome 219 patients were investigated, 113 in the small‐bites and 106 in the large‐bites group. At 1 month after surgery the RAM distance was smaller for small bites than for large bites (mean(s.d.) 1·90(1·18) versus 2·39(1·34) cm respectively; P = 0·005). At 1 year, patients with incisional hernia had a longer RAM distance at 1 month than those with no incisional hernia (mean(s.d.) 2·43(1·48) versus 2·03(1·19) cm respectively; relative risk 1·14, 95 per cent c.i. 1·03 to 1·26, P = 0·015).ConclusionA RAM distance greater than 2 cm at 1 month after midline laparotomy is associated with incisional hernia. Closure with small bites results in a smaller distance between the muscles.

show abstract

Covariate adjustment in randomized controlled trials with dichotomous outcomes increases statistical power and reduces sample size requirements

Cited by 210 publications

References 31 publications

Outcomes and Statistical Power in Adult Critical Care Randomized Trials

Outcomes and Statistical Power in Adult Critical Care Randomized Trials

Cox proportional hazards models have more statistical power than logistic regression models in cross-sectional genetic association studies

Development of incisional herniation after midline laparotomy

Contact Info

Product

Resources

About