{"doi":"10.1111/biom.13415","title":"Two robust tools for inference about causal effects with invalid instruments","abstract":"<jats:title>Abstract</jats:title><jats:p>Instrumental variables have been widely used to estimate the causal effect of a treatment on an outcome. Existing confidence intervals for causal effects based on instrumental variables assume that all of the putative instrumental variables are valid; a valid instrumental variable is a variable that affects the outcome only by affecting the treatment and is not related to unmeasured confounders. However, in practice, some of the putative instrumental variables are likely to be invalid. This paper presents two tools to conduct valid inference and tests in the presence of invalid instruments. First, we propose a simple and general approach to construct confidence intervals based on taking unions of well‐known confidence intervals. Second, we propose a novel test for the null causal effect based on a collider bias. Our two proposals outperform traditional instrumental variable confidence intervals when invalid instruments are present and can also be used as a sensitivity analysis when there is concern that instrumental variables assumptions are violated. The new approach is applied to a Mendelian randomization study on the causal effect of low‐density lipoprotein on globulin levels.</jats:p>","journal":"Biometrics","year":2022,"id":637124,"datarank":0.3596842909197557,"base_score":2.3978952727983707,"endowment":2.3978952727983707,"self_citation_contribution":0.3596842909197557,"citation_network_contribution":0.0,"self_endowment_contribution":0.3596842909197557,"citer_contribution":0.0,"corpus_percentile":null,"corpus_rank":null,"citation_count":10,"citer_count":0,"citers_with_citation_signal":0,"citers_with_endowment":0,"datacite_reuse_total":0,"is_dataset":false,"is_dataset_confidence":null,"is_data_producer":false,"deposit_databanks":null,"is_oa":false,"file_count":0,"downloads":0,"has_version_chain":false,"published_date":null,"fair_score":null,"fair_percentile":null,"algorithm_id":"datarank_citation_only_1hop_v6","ranking_scope":"data_only","authors":[{"id":1154279,"name":"Youjin Lee","orcid":"0000-0001-7299-935X","position":1,"is_corresponding":false},{"id":835906,"name":"T. Tony Cai","orcid":null,"position":2,"is_corresponding":false},{"id":415406,"name":"Dylan S. Small","orcid":"0000-0003-4928-2646","position":3,"is_corresponding":false},{"id":556968,"name":"Hyunseung Kang","orcid":"0000-0001-5748-2707","position":0,"is_corresponding":false}],"reference_count":0,"raw_metadata":{"has_enrichment":true,"resolved":true,"title":"Two robust tools for inference about causal effects with invalid instruments","abstract":"<jats:title>Abstract</jats:title><jats:p>Instrumental variables have been widely used to estimate the causal effect of a treatment on an outcome. Existing confidence intervals for causal effects based on instrumental variables assume that all of the putative instrumental variables are valid; a valid instrumental variable is a variable that affects the outcome only by affecting the treatment and is not related to unmeasured confounders. However, in practice, some of the putative instrumental variables are likely to be invalid. This paper presents two tools to conduct valid inference and tests in the presence of invalid instruments. First, we propose a simple and general approach to construct confidence intervals based on taking unions of well‐known confidence intervals. Second, we propose a novel test for the null causal effect based on a collider bias. Our two proposals outperform traditional instrumental variable confidence intervals when invalid instruments are present and can also be used as a sensitivity analysis when there is concern that instrumental variables assumptions are violated. The new approach is applied to a Mendelian randomization study on the causal effect of low‐density lipoprotein on globulin levels.</jats:p>","is_dataset_classified":null,"base_score":2.3978952727983707,"endowment":2.3978952727983707,"datacite_reuse_total":0,"file_count":0,"downloads":0,"views":0,"has_version_chain":false,"is_dataset":false,"is_oa":false,"pmid":"33616910","pmcid":null,"openalex_id":"https://openalex.org/W3112532512","authors":[],"funders":[{"funder_name":"National Science Foundation","grant_id":"DMS‐1712735","title":null},{"funder_name":"National Science Foundation","grant_id":"DMS‐2015259","title":null},{"funder_name":"NHLBI NIH HHS","grant_id":"HHSN268201500001I","title":null},{"funder_name":"NHLBI NIH HHS","grant_id":"N01HC25195","title":null},{"funder_name":"National Institute of Health (NIH)","grant_id":"R01-GM123056","title":null},{"funder_name":"National Institute of Health (NIH)","grant_id":"R01-GM129781","title":null},{"funder_name":"National Institutes of Health","grant_id":"N01HC025195-005","title":"THE FRAMINGHAM HEART STUDY-268025195"},{"funder_name":"National Science Foundation","grant_id":"1811414","title":"Theory and Methods for Inferring Causal Effects with Mendelian Randomization"}],"total_grants":8,"fwci":1.1214,"citation_percentile":0.79601515,"influential_citations":0,"citation_trend":[{"year":2021,"count":1},{"year":2022,"count":2},{"year":2023,"count":4},{"year":2024,"count":2},{"year":2026,"count":1}],"oa_status":"closed","license":"Wiley Online Library User Agreement","oa_locations":[{"url":"https://onlinelibrary.wiley.com/doi/pdf/10.1111/biom.13415","host_type":"publisher"},{"url":"https://onlinelibrary.wiley.com/doi/full-xml/10.1111/biom.13415","host_type":"publisher"},{"url":"https://onlinelibrary.wiley.com/doi/am-pdf/10.1111/biom.13415","host_type":"publisher"},{"url":"https://doi.org/10.1111/biom.13415","host_type":"journal"},{"url":"https://pubmed.ncbi.nlm.nih.gov/33616910","host_type":"repository"},{"url":"https://rss.onlinelibrary.wiley.com/doi/am-pdf/10.1111/biom.13415","host_type":""},{"url":"https://dx.doi.org/10.48550/arxiv.2006.01393","host_type":""},{"url":"http://arxiv.org/abs/2006.01393","host_type":""},{"url":"https://zbmath.org/7756159","host_type":""},{"url":"https://dx.doi.org/10.1111/biom.13415","host_type":""}],"fields_of_study":["Genetic Associations and Epidemiology","Advanced Causal Inference Techniques","Statistical Methods in Clinical Trials","05 social sciences","01 natural sciences","0502 economics and business","0101 mathematics"],"mesh_terms":["Bias","Causality","Mendelian Randomization Analysis"],"keywords":["Instrumental variable","Causal inference","Mendelian randomization","Inference","Econometrics","Outcome (game theory)","Confidence interval","Confounding","Statistics","Variable (mathematics)","Causal model","Computer science","Mathematics","Artificial intelligence","Likelihood Ratio Test","Anderson-rubin Test","Invalid Instrument","Weak Instrument","FOS: Computer and information sciences","Mendelian Randomization Analysis","Statistics - Applications","Applications of statistics to biology and medical sciences; meta analysis","Causality","Methodology (stat.ME)","Bias","Applications (stat.AP)","Statistics - Methodology"],"sdg_mappings":[],"linked_datasets":[],"clinical_trials":[],"software_tools":[],"database_accessions":[],"source":"live","citation_network_status":"fetched"},"created_at":"2026-08-06T18:21:25.653901Z","pmid":null,"pmcid":null,"fwci":null,"citation_percentile":null,"influential_citations":0,"oa_status":null,"license":null,"views":0,"total_file_size_bytes":0,"version_count":0,"fair_f":null,"fair_a":null,"fair_i":null,"fair_r":null,"fair_zscore":null,"fair_rationale":null,"fair_model":null,"fair_agent_version":null,"fair_fulltext_source":null,"fair_has_llm":null,"fair_computed_at":null,"clinical_trials":[],"software_tools":[],"db_accessions":[],"linked_datasets":[],"topics":[]}