Assessment of proportional hazard assumption in aggregate data: a systematic review on statistical methodology in clinical trials using time-to-event endpoint

Rulli, Eliana; Ghilotti, Francesca; Biagioli, Elena; Porcu, Luca; Marabese, Mirko; D’Incalci, Maurizio; Bellocco, Rino; Torri, Valter

doi:10.1038/s41416-018-0302-8

Cited by 47 publications

(38 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figure 1 A presents the crude Kaplan–Meier BFS curves without adjustments for patients receiving robotic RP. The crude Kaplan–Meier BFS curves of the 1–50 and 51–100 hospital volume cross; there is a violation of the proportional-hazards assumption [ 37 ]. Thus, there were no statistically significant differences between 1–50 and 51–100 hospital-volume per year in robotic RP.…”

Section: Discussionmentioning

confidence: 99%

Effects of 1-Year Hospital Volume on Surgical Margin and Biochemical-Failure-Free Survival in Patients Undergoing Robotic versus Nonrobotic Radical Prostatectomy: A Nationwide Cohort Study from the National Taiwan Cancer Database

Chang

Hsu

Lin

et al. 2021

Cancers

View full text Add to dashboard Cite

Purpose: To examine the effect of hospital volume on positive surgical margin (PSM) and biochemical-failure-free survival (BFS) rates in patients with prostate cancer (PC) undergoing robotic-assisted or nonrobotic-assisted radical prostatectomy (RP). Patients and Methods: The patients were men collected in the National Taiwan Cancer Registry diagnosed as having PC without distant metastasis who received RP from 44 multi-institutes in Taiwan. The logistic regression method was used to analyze the risk from RP to PSM in included patients with hospital volume (i.e., number of patients with PC receiving robotic RP per year), and the Cox proportional hazards method was used to analyze the time from the index date to biochemical recurrence. Results: After propensity score adjustment, compared with hospitals with >100 patients/year, the adjusted odds ratios (aORs; 95% confidence intervals) of PSM in the robotic RP group in hospitals with 1–25, 26–50, and 51–100 patients/year were 2.25 (2.10–3.11), 1.42 (1.25–2.23), and 1.33 (1.13–2.04), respectively (type III p < 0.0001). Sensitivity analysis indicated that the aORs of PSM were 1.29 (1.07–1.81), 1.07 (0.70–1.19), and 0.61 (0.56–0.83), respectively, for patients receiving robotic RP compared with nonrobotic RP within hospitals with 1–25, 26–50, and 51–100 patients/year, respectively. Compared with hospitals with >100 patients/year, the adjusted hazard ratios (aHRs) of biochemical failure in the robotic RP group were 1.40 (1.04–1.67), 1.34 (1.06–1.96), and 1.31 (1.05–2.15) in hospitals with 1–25, 26–50, and 51–100 patients/year, respectively. Conclusions: Hospital volume significantly affected PSM and BFS in robotic RP, but not in nonrobotic RP. When patients with PC want to receive robotic RP, it should be performed in a relatively high-volume hospital (>100 patients/year).

show abstract

Section: Discussionmentioning

confidence: 99%

Effects of 1-Year Hospital Volume on Surgical Margin and Biochemical-Failure-Free Survival in Patients Undergoing Robotic versus Nonrobotic Radical Prostatectomy: A Nationwide Cohort Study from the National Taiwan Cancer Database

Chang

Hsu

Lin

et al. 2021

Cancers

View full text Add to dashboard Cite

show abstract

“…Inappropriate conduct and low quality of reporting SAMs have been identified previously and may lead to incorrect conclusions [ 1 , 16 – 18 ]. Previous published reviews of SAMs in medical research have found the quality of reporting SAMs inadequate [ 1 , 16 , 17 , 19 , 20 ].…”

Section: Introductionmentioning

confidence: 99%

A systematic review of the quality of conduct and reporting of survival analyses of tuberculosis outcomes in Africa

Ngari

Schmitz

Maronga

et al. 2021

BMC Med Res Methodol

View full text Add to dashboard Cite

Background Survival analyses methods (SAMs) are central to analysing time-to-event outcomes. Appropriate application and reporting of such methods are important to ensure correct interpretation of the data. In this study, we systematically review the application and reporting of SAMs in studies of tuberculosis (TB) patients in Africa. It is the first review to assess the application and reporting of SAMs in this context. Methods Systematic review of studies involving TB patients from Africa published between January 2010 and April 2020 in English language. Studies were eligible if they reported use of SAMs. Application and reporting of SAMs were evaluated based on seven author-defined criteria. Results Seventy-six studies were included with patient numbers ranging from 56 to 182,890. Forty-three (57%) studies involved a statistician/epidemiologist. The number of published papers per year applying SAMs increased from two in 2010 to 18 in 2019 (P = 0.004). Sample size estimation was not reported by 67 (88%) studies. A total of 22 (29%) studies did not report summary follow-up time. The survival function was commonly presented using Kaplan-Meier survival curves (n = 51, (67%) studies) and group comparisons were performed using log-rank tests (n = 44, (58%) studies). Sixty seven (91%), 3 (4.1%) and 4 (5.4%) studies reported Cox proportional hazard, competing risk and parametric survival regression models, respectively. A total of 37 (49%) studies had hierarchical clustering, of which 28 (76%) did not adjust for the clustering in the analysis. Reporting was adequate among 4.0, 1.3 and 6.6% studies for sample size estimation, plotting of survival curves and test of survival regression underlying assumptions, respectively. Forty-five (59%), 52 (68%) and 73 (96%) studies adequately reported comparison of survival curves, follow-up time and measures of effect, respectively. Conclusion The quality of reporting survival analyses remains inadequate despite its increasing application. Because similar reporting deficiencies may be common in other diseases in low- and middle-income countries, reporting guidelines, additional training, and more capacity building are needed along with more vigilance by reviewers and journal editors.

show abstract

“…The logrank test 1,2 or, equivalently, the score test from the Cox proportional hazards (PH) model 3 with only a treatment arm indicator, remains a popular option for testing H 0 in randomized clinical trials. [4][5][6][7] Two well-known issues emerge when the PH assumption is non-trivially violated, as often seen [8][9][10][11][12][13] in practice: the power of the logrank test can be substantially diminished and the hazard ratio estimate from the Cox PH model can be hard to interpret.…”

mentioning

confidence: 99%

Survival analysis using a 5‐step stratified testing and amalgamation routine (5‐STAR) in randomized clinical trials

Mehrotra

West

2020

Statistics in Medicine

View full text Add to dashboard Cite

Randomized clinical trials are often designed to assess whether a test treatment prolongs survival relative to a control treatment. Increased patient heterogeneity, while desirable for generalizability of results, can weaken the ability of common statistical approaches to detect treatment differences, potentially hampering the regulatory approval of safe and efficacious therapies. A novel solution to this problem is proposed. A list of baseline covariates that have the potential to be prognostic for survival under either treatment is pre‐specified in the analysis plan. At the analysis stage, using all observed survival times but blinded to patient‐level treatment assignment, “noise” covariates are removed with elastic net Cox regression. The shortened covariate list is used by a conditional inference tree algorithm to segment the heterogeneous trial population into subpopulations of prognostically homogeneous patients (risk strata). After patient‐level treatment unblinding, a treatment comparison is done within each formed risk stratum and stratum‐level results are combined for overall statistical inference. The impressive power‐boosting performance of our proposed 5‐step stratified testing and amalgamation routine (5‐STAR), relative to that of the logrank test and other common approaches that do not leverage inherently structured patient heterogeneity, is illustrated using a hypothetical and two real datasets along with simulation results. Furthermore, the importance of reporting stratum‐level comparative treatment effects (time ratios from accelerated failure time model fits in conjunction with model averaging and, as needed, hazard ratios from Cox proportional hazard model fits) is highlighted as a potential enabler of personalized medicine. An R package is available at https://github.com/rmarceauwest/fiveSTAR.

show abstract

Assessment of proportional hazard assumption in aggregate data: a systematic review on statistical methodology in clinical trials using time-to-event endpoint

Cited by 47 publications

References 87 publications

Effects of 1-Year Hospital Volume on Surgical Margin and Biochemical-Failure-Free Survival in Patients Undergoing Robotic versus Nonrobotic Radical Prostatectomy: A Nationwide Cohort Study from the National Taiwan Cancer Database

Effects of 1-Year Hospital Volume on Surgical Margin and Biochemical-Failure-Free Survival in Patients Undergoing Robotic versus Nonrobotic Radical Prostatectomy: A Nationwide Cohort Study from the National Taiwan Cancer Database

A systematic review of the quality of conduct and reporting of survival analyses of tuberculosis outcomes in Africa

Survival analysis using a 5‐step stratified testing and amalgamation routine (5‐STAR) in randomized clinical trials

Contact Info

Product

Resources

About