Time‐dependent mediators in survival analysis: Modeling direct and indirect effects with the additive hazards model

Aalen, Odd O.; Stensrud, Mats Julius; Didelez, Vanessa; Daniel, Rhian; Røysland, Kjetil; Strohmaier, Susanne

doi:10.1002/bimj.201800263

Cited by 45 publications

(62 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…• The facts that 0 ≤ Ω 2 ≤ 1 and 0 ≤ R 2 ≤ 1 follow immediately from their definitions 7and (13), assuming that the latter is estimated using (11).…”

Section: Properties Of 2 and Rmentioning

confidence: 99%

“…[4][5][6][7][8] The collapsibility as well as other behaviors similar to a linear model, has also enabled the additive hazards model to be used in mediation analysis of survival data. [9][10][11][12][13] In addition, doubly robust methods have been developed for estimating treatment effects and applied in practice under the additive hazards model including for optimal treatment regimes, [14][15][16] while the noncollapsibility of the Cox model presents an obstacle in the development of doubly robust method when confounders are present. 17 Estimation and inference procedures have been well developed and implemented under the additive hazards model (eg, R package "timereg"), and diagnostic methods have also been proposed.…”

Section: Introductionmentioning

confidence: 99%

“…For this reason it has been used in the development of instrumental variable approaches for survival data including competing risks 4‐8 . The collapsibility as well as other behaviors similar to a linear model, has also enabled the additive hazards model to be used in mediation analysis of survival data 9‐13 . In addition, doubly robust methods have been developed for estimating treatment effects and applied in practice under the additive hazards model including for optimal treatment regimes, 14‐16 while the noncollapsibility of the Cox model presents an obstacle in the development of doubly robust method when confounders are present 17 …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Explained variation under the additive hazards model

Rava

2020

Statistics in Medicine

View full text Add to dashboard Cite

We study explained variation under the additive hazards regression model for right-censored data. We consider different approaches for developing such a measure, and focus on one that estimates the proportion of variation in the failure time explained by the covariates. We study the properties of the measure both analytically, and through extensive simulations. We apply the measure to a well-known survival dataset as well as the linked surveillance, epidemiology, and end results-Medicare database for prediction of mortality in early stage prostate cancer patients using high-dimensional claims codes.

show abstract

“…• The facts that 0 ≤ Ω 2 ≤ 1 and 0 ≤ R 2 ≤ 1 follow immediately from their definitions 7and (13), assuming that the latter is estimated using (11).…”

Section: Properties Of 2 and Rmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Explained variation under the additive hazards model

Rava

2020

Statistics in Medicine

View full text Add to dashboard Cite

show abstract

“… 18 , 22 , 49 , 50 This difference is due to interventional effects being population-level quantities, like the total causal effect, whilst natural effects are individual-level effects. 22 An exception for natural effects is when the exposure is separable into components acting through distinct pathways, 6 , 51 , 52 with the resulting separable effects emulating hypothetical trials of intervention regimes on the exposure components.…”

Section: Discussionmentioning

confidence: 99%

Mediation effects that emulate a target randomised trial: Simulation-based evaluation of ill-defined interventions on multiple mediators

Moreno‐Betancur

Moran

Becker

et al. 2021

Stat Methods Med Res

View full text Add to dashboard Cite

Many epidemiological questions concern potential interventions to alter the pathways presumed to mediate an association. For example, we consider a study that investigates the benefit of interventions in young adulthood for ameliorating the poorer mid-life psychosocial outcomes of adolescent self-harmers relative to their healthy peers. Two methodological challenges arise. First, mediation methods have hitherto mostly focused on the elusive task of discovering pathways, rather than on the evaluation of mediator interventions. Second, the complexity of such questions is invariably such that there are no well-defined mediator interventions (i.e. actual treatments, programs, etc.) for which data exist on the relevant populations, outcomes and time-spans of interest. Instead, researchers must rely on exposure (non-intervention) data, that is, on mediator measures such as depression symptoms for which the actual interventions that one might implement to alter them are not well defined. We propose a novel framework that addresses these challenges by defining mediation effects that map to a target trial of hypothetical interventions targeting multiple mediators for which we simulate the effects. Specifically, we specify a target trial addressing three policy-relevant questions, regarding the impacts of hypothetical interventions that would shift the mediators’ distributions (separately under various interdependence assumptions, jointly or sequentially) to user-specified distributions that can be emulated with the observed data. We then define novel interventional effects that map to this trial, simulating shifts by setting mediators to random draws from those distributions. We show that estimation using a g-computation method is possible under an expanded set of causal assumptions relative to inference with well-defined interventions, which reflects the lower level of evidence that is expected with ill-defined interventions. Application to the self-harm example in the Victorian Adolescent Health Cohort Study illustrates the value of our proposal for informing the design and evaluation of actual interventions in the future.

show abstract

“…An extensive discussion of the dynamic viewpoint is given in Aalen et al (2012). Local independence can be formulated causally (Didelez 2019, in this special issue), and dynamic path analysis has now been formulated in a causal setting using Pearl's do-operator (Aalen et al 2019).…”

mentioning

confidence: 99%