A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization
Abstract Background Mendelian randomization is a popular method for causal inference with observational data that uses genetic variants as instrumental variables. Similarly to a randomized trial, a standard Mendelian randomization analysis estimates the population-averaged effect of an exposure on a...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2024-02-01
|
| Series: | BMC Medical Research Methodology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12874-024-02153-1 |
| _version_ | 1827327079632338944 |
|---|---|
| author | Haodong Tian Brian D. M. Tom Stephen Burgess |
| author_facet | Haodong Tian Brian D. M. Tom Stephen Burgess |
| author_sort | Haodong Tian |
| collection | DOAJ |
| description | Abstract Background Mendelian randomization is a popular method for causal inference with observational data that uses genetic variants as instrumental variables. Similarly to a randomized trial, a standard Mendelian randomization analysis estimates the population-averaged effect of an exposure on an outcome. Dividing the population into subgroups can reveal effect heterogeneity to inform who would most benefit from intervention on the exposure. However, as covariates are measured post-“randomization”, naive stratification typically induces collider bias in stratum-specific estimates. Method We extend a previously proposed stratification method (the “doubly-ranked method”) to form strata based on a single covariate, and introduce a data-adaptive random forest method to calculate stratum-specific estimates that are robust to collider bias based on a high-dimensional covariate set. We also propose measures based on the Q statistic to assess heterogeneity between stratum-specific estimates (to understand whether estimates are more variable than expected due to chance alone) and variable importance (to identify the key drivers of effect heterogeneity). Result We show that the effect of body mass index (BMI) on lung function is heterogeneous, depending most strongly on hip circumference and weight. While for most individuals, the predicted effect of increasing BMI on lung function is negative, it is positive for some individuals and strongly negative for others. Conclusion Our data-adaptive approach allows for the exploration of effect heterogeneity in the relationship between an exposure and an outcome within a Mendelian randomization framework. This can yield valuable insights into disease aetiology and help identify specific groups of individuals who would derive the greatest benefit from targeted interventions on the exposure. |
| first_indexed | 2024-03-07T14:55:50Z |
| format | Article |
| id | doaj.art-c9f473150dd5491baf5954549df3ff7b |
| institution | Directory Open Access Journal |
| issn | 1471-2288 |
| language | English |
| last_indexed | 2024-03-07T14:55:50Z |
| publishDate | 2024-02-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Medical Research Methodology |
| spelling | doaj.art-c9f473150dd5491baf5954549df3ff7b2024-03-05T19:28:33ZengBMCBMC Medical Research Methodology1471-22882024-02-0124111610.1186/s12874-024-02153-1A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomizationHaodong Tian0Brian D. M. Tom1Stephen Burgess2MRC Biostatistics Unit, School of Clinical Medicine, University of CambridgeMRC Biostatistics Unit, School of Clinical Medicine, University of CambridgeMRC Biostatistics Unit, School of Clinical Medicine, University of CambridgeAbstract Background Mendelian randomization is a popular method for causal inference with observational data that uses genetic variants as instrumental variables. Similarly to a randomized trial, a standard Mendelian randomization analysis estimates the population-averaged effect of an exposure on an outcome. Dividing the population into subgroups can reveal effect heterogeneity to inform who would most benefit from intervention on the exposure. However, as covariates are measured post-“randomization”, naive stratification typically induces collider bias in stratum-specific estimates. Method We extend a previously proposed stratification method (the “doubly-ranked method”) to form strata based on a single covariate, and introduce a data-adaptive random forest method to calculate stratum-specific estimates that are robust to collider bias based on a high-dimensional covariate set. We also propose measures based on the Q statistic to assess heterogeneity between stratum-specific estimates (to understand whether estimates are more variable than expected due to chance alone) and variable importance (to identify the key drivers of effect heterogeneity). Result We show that the effect of body mass index (BMI) on lung function is heterogeneous, depending most strongly on hip circumference and weight. While for most individuals, the predicted effect of increasing BMI on lung function is negative, it is positive for some individuals and strongly negative for others. Conclusion Our data-adaptive approach allows for the exploration of effect heterogeneity in the relationship between an exposure and an outcome within a Mendelian randomization framework. This can yield valuable insights into disease aetiology and help identify specific groups of individuals who would derive the greatest benefit from targeted interventions on the exposure.https://doi.org/10.1186/s12874-024-02153-1GeneticsInstrumental variableStratificationHeterogenous effectRandom forestVariable importance |
| spellingShingle | Haodong Tian Brian D. M. Tom Stephen Burgess A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization BMC Medical Research Methodology Genetics Instrumental variable Stratification Heterogenous effect Random forest Variable importance |
| title | A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization |
| title_full | A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization |
| title_fullStr | A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization |
| title_full_unstemmed | A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization |
| title_short | A data-adaptive method for investigating effect heterogeneity with high-dimensional covariates in Mendelian randomization |
| title_sort | data adaptive method for investigating effect heterogeneity with high dimensional covariates in mendelian randomization |
| topic | Genetics Instrumental variable Stratification Heterogenous effect Random forest Variable importance |
| url | https://doi.org/10.1186/s12874-024-02153-1 |
| work_keys_str_mv | AT haodongtian adataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization AT briandmtom adataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization AT stephenburgess adataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization AT haodongtian dataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization AT briandmtom dataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization AT stephenburgess dataadaptivemethodforinvestigatingeffectheterogeneitywithhighdimensionalcovariatesinmendelianrandomization |