Non-collapsible marginal effect measures depend on joint distributions of effect modifiers and prognostic variables, so unadjusted anchored indirect comparisons can be biased even without individual-level treatment effect heterogeneity.
Estimands in heath technology assessment: a causal inference perspective.Statistics in medicine2022; 41(28): 5577–5585
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ME 1years
2022 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Transportability of model-based estimands in evidence synthesis
Non-collapsible marginal effect measures depend on joint distributions of effect modifiers and prognostic variables, so unadjusted anchored indirect comparisons can be biased even without individual-level treatment effect heterogeneity.