2018
DOI: 10.1136/bmj.k182
|View full text |Cite
|
Sign up to set email alerts
|

How to estimate the effect of treatment duration on survival outcomes using observational data

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

2
127
0

Year Published

2018
2018
2024
2024

Publication Types

Select...
6
1

Relationship

0
7

Authors

Journals

citations
Cited by 119 publications
(129 citation statements)
references
References 17 publications
2
127
0
Order By: Relevance
“…In addition, because eligible individuals could not be uniquely assigned to a strategy at baseline when emulating target trials with a grace period, we duplicated ("cloned") eligible individuals and assigned 1 clone to each treatment strategy (eFigure 3 in the Supplement). 30 We then censored clones if they had no data compatible with their assigned treatment strategy by the end of the grace period. 6,30 Details on all possible censoring schemes are given in eAppendix 2 in the Supplement.…”
Section: Resultsmentioning
confidence: 99%
See 1 more Smart Citation
“…In addition, because eligible individuals could not be uniquely assigned to a strategy at baseline when emulating target trials with a grace period, we duplicated ("cloned") eligible individuals and assigned 1 clone to each treatment strategy (eFigure 3 in the Supplement). 30 We then censored clones if they had no data compatible with their assigned treatment strategy by the end of the grace period. 6,30 Details on all possible censoring schemes are given in eAppendix 2 in the Supplement.…”
Section: Resultsmentioning
confidence: 99%
“…30 We then censored clones if they had no data compatible with their assigned treatment strategy by the end of the grace period. 6,30 Details on all possible censoring schemes are given in eAppendix 2 in the Supplement.…”
Section: Resultsmentioning
confidence: 99%
“…An observational study design, as shown in this study, from a large clinical databases can be more representative to real patient population, may better reflect real-world practice, relatively inexpensive, the ethical issues appear less complex (no patients are randomized). In addition, the studies are in progress to use the big data and/or large observational study for causal analyses, besides correlation analyses [31][32][33] .…”
Section: Discussionmentioning
confidence: 99%
“…In recent years, many excellent methods to control these biases have been developed, such as propensity score analysis, inverse probability weighting, marginal structural modeling, bootstrapping, instrumental variable, etc. [5][6][7][8]. Given the low cost and easy accessibility of DBs and the necessity of bias control, observational studies are considered an excellent tool for new researchers.…”
Section: Modern Epidemiology and Clinical Researchmentioning
confidence: 99%
“…For analyses of these longitudinal data with time-varying confounders, Robins reported a g formula to measure the healthy worker survivor effect [7]. Although this formula is complicated, Hernan recently reported a simple three-step approach to estimate the effect of treatment duration on survival outcomes using observational data [8]. The first step is duplicating people to assign them to treatment duration strategies at time zero, eliminating immortal time bias [33][34][35].…”
Section: Flow Of a Propensity Score Analysismentioning
confidence: 99%