1
|
Angrist JD, Hull P. Instrumental variables methods reconcile intention-to-screen effects across pragmatic cancer screening trials. Proc Natl Acad Sci U S A 2023; 120:e2311556120. [PMID: 38100416 PMCID: PMC10742387 DOI: 10.1073/pnas.2311556120] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/07/2023] [Accepted: 11/03/2023] [Indexed: 12/17/2023] Open
Abstract
Pragmatic cancer screening trials mimic real-world scenarios in which patients and doctors are the ultimate arbiters of treatment. Intention-to-screen (ITS) analyses of such trials maintain randomization-based apples-to-apples comparisons, but differential adherence (the failure of subjects assigned to screening to get screened) makes ITS effects hard to compare across trials and sites. We show how instrumental variables (IV) methods address the nonadherence challenge in a comparison of estimates from 17 sites in five randomized trials measuring screening effects on colorectal cancer incidence. While adherence rates and ITS estimates vary widely across and within trials, IV estimates of per-protocol screening effects are remarkably consistent. An application of simple IV tools, including graphical analysis and formal statistical tests, shows how differential adherence explains variation in ITS impact. Screening compliers are also shown to have demographic characteristics similar to those of the full trial study sample. These findings argue for the clinical relevance of IV estimates of cancer screening effects.
Collapse
Affiliation(s)
- Joshua D. Angrist
- Department of Economics and National Bureau of Economic Research, Massachusetts Institute of Technology, Cambridge, MA02142
| | - Peter Hull
- Department of Economics and National Bureau of Economic Research, Brown University, Providence, RI02912
| |
Collapse
|
2
|
Diemer EW, Havdahl A, Andreassen OA, Munafò M, Njolstad PR, Tiemeier H, Zuccolo L, Swanson SA. Bounding the average causal effect in Mendelian randomisation studies with multiple proposed instruments: An application to prenatal alcohol exposure and attention deficit hyperactivity disorder. Paediatr Perinat Epidemiol 2023; 37:326-337. [PMID: 36722651 PMCID: PMC10946905 DOI: 10.1111/ppe.12951] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/01/2022] [Revised: 10/11/2022] [Accepted: 12/17/2022] [Indexed: 02/02/2023]
Abstract
BACKGROUND As large-scale observational data become more available, caution regarding causal assumptions remains critically important. This may be especially true for Mendelian randomisation (MR), an increasingly popular approach. Point estimation in MR usually requires strong, often implausible homogeneity assumptions beyond the core instrumental conditions. Bounding, which does not require homogeneity assumptions, is infrequently applied in MR. OBJECTIVES We aimed to demonstrate computing nonparametric bounds for the causal risk difference derived from multiple proposed instruments in an MR study where effect heterogeneity is expected. METHODS Using data from the Norwegian Mother, Father and Child Cohort Study (n = 2056) and Avon Longitudinal Study of Parents and Children (n = 6216) to study the average causal effect of maternal pregnancy alcohol use on offspring attention deficit hyperactivity disorder symptoms, we proposed 11 maternal SNPs as instruments. We computed bounds assuming subsets of SNPs were jointly valid instruments, for all combinations of SNPs where the MR model was not falsified. RESULTS The MR assumptions were violated for all sets with more than 4 SNPs in one cohort and for all sets with more than 2 SNPs in the other. Bounds assuming one SNP was an individually valid instrument barely improved on assumption-free bounds. Bounds tightened as more SNPs were assumed to be jointly valid instruments, and occasionally identified directions of effect, though bounds from different sets varied. CONCLUSIONS Our results suggest that, when proposing multiple instruments, bounds can contextualise plausible magnitudes and directions of effects. Computing bounds over multiple assumption sets, particularly in large, high-dimensional data, offers a means of triangulating results across different potential sources of bias within a study and may help researchers to better evaluate and emphasise which estimates are compatible with the most plausible assumptions for their specific setting.
Collapse
Affiliation(s)
- Elizabeth W. Diemer
- Department of Child and Adolescent PsychiatryErasmus MCRotterdamthe Netherlands
- CAUSALabHarvard T.H. Chan School of Public HealthBostonMassachusettsUSA
- Department of EpidemiologyHarvard T.H. Chan School of Public HealthBostonMassachusettsUSA
| | - Alexandra Havdahl
- MRC Integrative Epidemiology Unit at the University of BristolBristolUK
- Nic Waals InstituteLovisenberg Diaconal HospitalOsloNorway
- Department of Mental DisordersNorwegian Institute of Public Health
| | - Ole A. Andreassen
- NORMENT, KG Jebsen Centre for Psychosis Research, Department of Clinical ScienceUniversity of BergenBergenNorway
| | - Marcus R. Munafò
- MRC Integrative Epidemiology Unit at the University of BristolBristolUK
- School of Psychological ScienceUniversity of BristolBristolUK
- NIHR Biomedical Research CentreUniversity Hospitals Bristol NHS Foundation Trust and University of BristolBristolUK
| | - Pal R. Njolstad
- Department of Paediatric and Adolescent MedicineHaukeland University HospitalBergenNorway
- KG Jebsen Center for Diabetes Research, Department of Clinical ScienceUniversity of BergenBergenNorway
| | - Henning Tiemeier
- Department of Child and Adolescent PsychiatryErasmus MCRotterdamthe Netherlands
- Department of Social and Behavioral ScienceHarvard T.H. Chan School of Public HealthBostonMassachusettsUSA
| | - Luisa Zuccolo
- MRC Integrative Epidemiology Unit at the University of BristolBristolUK
- Department of Population Health Sciences, Bristol Medical SchoolUniversity of BristolBristolUK
- Centre for Health Data ScienceHuman Technopole FoundationMilanItaly
| | - Sonja A. Swanson
- CAUSALabHarvard T.H. Chan School of Public HealthBostonMassachusettsUSA
- Department of EpidemiologyHarvard T.H. Chan School of Public HealthBostonMassachusettsUSA
- Department of EpidemiologyErasmus MCRotterdamthe Netherlands
- Department of EpidemiologyUniversity of PittsburghPittsburghPennsylvaniaUSA
| |
Collapse
|
3
|
Diemer EW, Zuccolo L, Swanson SA. Partial Identification of the Average Causal Effect in Multiple Study Populations: The Challenge of Combining Mendelian Randomization Studies. Epidemiology 2023; 34:20-28. [PMID: 35944150 PMCID: PMC9719801 DOI: 10.1097/ede.0000000000001526] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/25/2021] [Accepted: 07/18/2022] [Indexed: 02/02/2023]
Abstract
BACKGROUND Researchers often use random-effects or fixed-effects meta-analysis to combine findings from multiple study populations. However, the causal interpretation of these models is not always clear, and they do not easily translate to settings where bounds, rather than point estimates, are computed. METHODS If bounds on an average causal effect of interest in a well-defined population are computed in multiple study populations under specified identifiability assumptions, then under those assumptions the average causal effect would lie within all study-specific bounds and thus the intersection of the study-specific bounds. We demonstrate this by pooling bounds on the average causal effect of prenatal alcohol exposure on attention deficit-hyperactivity disorder symptoms, computed in two European cohorts and under multiple sets of assumptions in Mendelian randomization (MR) analyses. RESULTS For all assumption sets considered, pooled bounds were wide and did not identify the direction of effect. The narrowest pooled bound computed implied the risk difference was between -4 and 34 percentage points. CONCLUSIONS All pooled bounds computed in our application covered the null, illustrating how strongly point estimates from prior MR studies of this effect rely on within-study homogeneity assumptions. We discuss how the interpretation of both pooled bounds and point estimation in MR is complicated by possible heterogeneity of effects across populations.
Collapse
Affiliation(s)
- Elizabeth W. Diemer
- From the Department of Child and Adolescent Psychiatry, Erasmus MC, Rotterdam, the Netherlands
- CAUSALab, Harvard T.H. Chan School of Public Health, Boston, MA
- Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA
| | - Luisa Zuccolo
- MRC Integrative Epidemiology Unit at the University of Bristol, Bristol, United Kingdom
- Department of Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, United Kingdom
- Center for Health Data Science, Human Technopole Foundation, Milan, Italy
| | - Sonja A. Swanson
- CAUSALab, Harvard T.H. Chan School of Public Health, Boston, MA
- Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA
- Department of Epidemiology, Erasmus MC, Rotterdam, the Netherlands
- Department of Epidemiology, School of Public Health, University of Pittsburgh, Pittsburgh, PA
| |
Collapse
|
4
|
Rojas-Saunero LP, Labrecque JA, Swanson SA. Invited Commentary: Conducting and Emulating Trials to Study Effects of Social Interventions. Am J Epidemiol 2022; 191:1453-1456. [PMID: 35445692 PMCID: PMC9347019 DOI: 10.1093/aje/kwac066] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Download PDF] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/02/2021] [Revised: 01/24/2022] [Accepted: 03/15/2022] [Indexed: 01/28/2023] Open
Abstract
All else being equal, if we had 1 causal effect we wished to estimate, we would conduct a randomized trial with a protocol that mapped onto that causal question, or we would attempt to emulate that target trial with observational data. However, studying the social determinants of health often means there are not just 1 but several causal contrasts of simultaneous interest and importance, and each of these related but distinct causal questions may have varying degrees of feasibility in conducting trials. With this in mind, we discuss challenges and opportunities that arise when conducting and emulating such trials. We describe designing trials with the simultaneous goals of estimating the intention-to-treat effect, the per-protocol effect, effects of alternative protocols or joint interventions, effects within subgroups, and effects under interference, and we describe ways to make the most of all feasible randomized trials and emulated trials using observational data. Our comments are grounded in the study results of Courtin et al. (Am J Epidemiol. 2022;191(8):1444-1452).
Collapse
Affiliation(s)
| | | | - Sonja A Swanson
- Correspondence to Dr. Sonja A. Swanson, Department of Epidemiology, School of Public Health, University of Pittsburgh, 130 De Soto Street, Pittsburgh, PA 15261 (e-mail: )
| |
Collapse
|
5
|
Caniglia EC, Zash R, Swanson SA, Smith E, Sudfeld C, Finkelstein JL, Diseko M, Mayondi G, Mmalane M, Makhema J, Fawzi W, Lockman S, Shapiro RL. Iron, folic acid, and multiple micronutrient supplementation strategies during pregnancy and adverse birth outcomes in Botswana. Lancet Glob Health 2022; 10:e850-e861. [PMID: 35561720 PMCID: PMC9309424 DOI: 10.1016/s2214-109x(22)00126-7] [Citation(s) in RCA: 8] [Impact Index Per Article: 4.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/06/2021] [Revised: 02/10/2022] [Accepted: 03/16/2022] [Indexed: 12/18/2022]
Abstract
Background Methods Findings Interpretation Funding
Collapse
Affiliation(s)
- Ellen C Caniglia
- Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
| | - Rebecca Zash
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana; Beth Israel Deaconess Medical Center, Boston, MA, USA
| | | | - Emily Smith
- School of Public Health, George Washington University, Washington DC, USA
| | - Christopher Sudfeld
- Harvard T H Chan School of Public Health, Harvard University, Boston, MA, USA
| | - Julia L Finkelstein
- Division of Nutritional Sciences, Cornell College of Human Ecology, Cornell University, Ithaca, NY, USA
| | - Modiegi Diseko
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana
| | - Gloria Mayondi
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana
| | - Mompati Mmalane
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana
| | - Joseph Makhema
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana
| | - Wafaie Fawzi
- Harvard T H Chan School of Public Health, Harvard University, Boston, MA, USA
| | - Shahin Lockman
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana; Harvard T H Chan School of Public Health, Harvard University, Boston, MA, USA; Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA
| | - Roger L Shapiro
- Botswana-Harvard AIDS Institute Partnership, Gaborone, Botswana; Harvard T H Chan School of Public Health, Harvard University, Boston, MA, USA
| |
Collapse
|
6
|
Margolis KL, Crain AL, Bergdall AR, Beran M, Anderson JP, Solberg LI, O'Connor PJ, Sperl-Hillen JM, Pawloski PA, Ziegenfuss JY, Rehrauer D, Norton C, Haugen P, Green BB, McKinney Z, Kodet A, Appana D, Sharma R, Trower NK, Williams R, Crabtree BF. Design of a pragmatic cluster-randomized trial comparing telehealth care and best practice clinic-based care for uncontrolled high blood pressure. Contemp Clin Trials 2020; 92:105939. [PMID: 31981712 DOI: 10.1016/j.cct.2020.105939] [Citation(s) in RCA: 13] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/03/2019] [Revised: 01/16/2020] [Accepted: 01/20/2020] [Indexed: 11/30/2022]
Abstract
BACKGROUND Uncontrolled hypertension is the largest single contributor to all-cause and cardiovascular mortality in the U.S. POPULATION Nurse- and pharmacist-led team-based care and telehealth care interventions have been shown to result in large and lasting improvements in blood pressure (BP); however, it is unclear how successfully these can be implemented at scale in real-world settings. It is also uncertain how telehealth interventions impact patient experience compared to traditional clinic-based care. AIMS/OBJECTIVES To compare the effects of two evidence-based blood pressure care strategies in the primary care setting: (1) best-practice clinic-based care and (2) telehealth care with home BP telemonitoring and management by a clinical pharmacist. To evaluate implementation using mixed-methods supported by the RE-AIM framework and Consolidated Framework for Implementation Research. METHODS The design is a cluster-randomized comparative effectiveness pragmatic trial in 21 primary care clinics (9 clinic-based care, 12 telehealth care). Adult patients (age 18-85) with hypertension are enrolled via automated electronic health record (EHR) tools during primary care encounters if BP is elevated to ≥150/95 mmHg at two consecutive visits. The primary outcome is change in systolic BP over 12 months as extracted from the EHR. Secondary outcomes are change in key patient-reported outcomes over 6 months as measured by surveys. Qualitative data are collected at various time points to investigate implementation barriers and help explain intervention effects. CONCLUSION This pragmatic trial aims to inform health systems about the benefits, strengths, and limitations of implementing home BP telemonitoring with pharmacist management for uncontrolled hypertension in real-world primary care settings.
Collapse
Affiliation(s)
- Karen L Margolis
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America.
| | - A Lauren Crain
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Anna R Bergdall
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - MarySue Beran
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Jeffrey P Anderson
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Leif I Solberg
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Patrick J O'Connor
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - JoAnn M Sperl-Hillen
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Pamala A Pawloski
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Jeanette Y Ziegenfuss
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Dan Rehrauer
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Christine Norton
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Patricia Haugen
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Beverly B Green
- Kaiser Permanente Washington Health Research Institute, 1730 Minor Av, Seattle, WA 98101, United States of America
| | - Zeke McKinney
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Amy Kodet
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Deepika Appana
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Rashmi Sharma
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Nicole K Trower
- HealthPartners Institute, Mailstop 23301A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - RaeAnn Williams
- HealthPartners, Mailstop 31100A, PO Box 1524, Minneapolis, MN 55440-1524, United States of America
| | - Benjamin F Crabtree
- Rutgers Robert Wood Johnson Medical School, Department of Family Medicine and Community Health, New Brunswick, NJ 08901, United States of America
| |
Collapse
|
7
|
Breskin A, Westreich D, Cole SR, Edwards JK. Using Bounds to Compare the Strength of Exchangeability Assumptions for Internal and External Validity. Am J Epidemiol 2019; 188:1355-1360. [PMID: 30834430 DOI: 10.1093/aje/kwz060] [Citation(s) in RCA: 9] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 08/21/2018] [Revised: 02/18/2019] [Accepted: 02/27/2019] [Indexed: 11/13/2022] Open
Abstract
In the absence of strong assumptions (e.g., exchangeability), only bounds for causal effects can be identified. Here we describe bounds for the risk difference for an effect of a binary exposure on a binary outcome in 4 common study settings: observational studies and randomized studies, each with and without simple random selection from the target population. Through these scenarios, we introduce randomizations for selection and treatment, and the widths of the bounds are narrowed from 2 (the width of the range of the risk difference) to 0 (point identification). We then assess the strength of the assumptions of exchangeability for internal and external validity by comparing their contributions to the widths of the bounds in the setting of an observational study without random selection from the target population. We find that when less than two-thirds of the target population is selected into the study, the assumption of exchangeability for external validity of the risk difference is stronger than that for internal validity. The relative strength of these assumptions should be considered when designing, analyzing, and interpreting observational studies and will aid in determining the best methods for estimating the causal effects of interest.
Collapse
Affiliation(s)
- Alexander Breskin
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| | - Daniel Westreich
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| | - Stephen R Cole
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| | - Jessie K Edwards
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina
| |
Collapse
|
8
|
Swanson SA, Hernán MA. The challenging interpretation of instrumental variable estimates under monotonicity. Int J Epidemiol 2019; 47:1289-1297. [PMID: 28379526 DOI: 10.1093/ije/dyx038] [Citation(s) in RCA: 39] [Impact Index Per Article: 7.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 02/24/2017] [Indexed: 11/12/2022] Open
Abstract
Background Instrumental variable (IV) methods are often used to identify 'local' causal effects in a subgroup of the population of interest. Such 'local' effects may not be ideal for informing clinical or policy decision making. When the instrument is non-causal, additional difficulties arise for interpreting 'local' effects. Little attention has been paid to these difficulties, even though commonly proposed instruments in epidemiology are non-causal (e.g. proxies for physician's preference; genetic variants in some Mendelian randomization studies). Methods For IV estimates obtained from both causal and non-causal instruments under monotonicity, we present results to help investigators pose four questions about the local effect estimates obtained in their studies. (1) To what subgroup of the population does the effect pertain? Can we (2) estimate the size of or (3) describe the characteristics of this subgroup relative to the study population? (4) Can the sensitivity of the effect estimate to deviations from monotonicity be quantified? Results We show that the common interpretations and approaches for answering these four questions are generally only appropriate in the case of causal instruments. Conclusions Appropriate interpretation of an IV estimate under monotonicity as a 'local' effect critically depends on whether the proposed instrument is causal or non-causal. The results and formal proofs presented here can help in the transparent reporting of IV results and in enhancing the use of IV estimates in informing decision-making efforts.
Collapse
Affiliation(s)
- Sonja A Swanson
- Department of Epidemiology, Erasmus Medical Center, Rotterdam, The Netherlands.,Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA, USA
| | - Miguel A Hernán
- Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA, USA.,Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA, USA.,Harvard-MIT Division of Health Sciences and Technology, Boston, MA, USA
| |
Collapse
|
9
|
Cole SR, Hudgens MG, Edwards JK, Brookhart MA, Richardson DB, Westreich D, Adimora AA. Nonparametric Bounds for the Risk Function. Am J Epidemiol 2019; 188:632-636. [PMID: 30698633 PMCID: PMC6438811 DOI: 10.1093/aje/kwz013] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/31/2018] [Revised: 01/10/2019] [Accepted: 01/11/2019] [Indexed: 01/08/2023] Open
Abstract
Nonparametric bounds for the risk difference are straightforward to calculate and make no untestable assumptions about unmeasured confounding or selection bias due to missing data (e.g., dropout). These bounds are often wide and communicate uncertainty due to possible systemic errors. An illustrative example is provided.
Collapse
Affiliation(s)
- Stephen R Cole
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
| | - Michael G Hudgens
- Department of Biostatistics, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
| | - Jessie K Edwards
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
| | - M Alan Brookhart
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
- NoviSci LLC, Durham, North Carolina
| | - David B Richardson
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
| | - Daniel Westreich
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
| | - Adaora A Adimora
- Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina
- Department of Medicine, School of Medicine, University of North Carolina, Chapel Hill, North Carolina
| |
Collapse
|
10
|
Swanson SA, Hernán MA, Miller M, Robins JM, Richardson TS. Partial Identification of the Average Treatment Effect Using Instrumental Variables: Review of Methods for Binary Instruments, Treatments, and Outcomes. J Am Stat Assoc 2018; 113:933-947. [PMID: 31537952 PMCID: PMC6752717 DOI: 10.1080/01621459.2018.1434530] [Citation(s) in RCA: 42] [Impact Index Per Article: 7.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022]
Abstract
Several methods have been proposed for partially or point identifying the average treatment effect (ATE) using instrumental variable (IV) type assumptions. The descriptions of these methods are widespread across the statistical, economic, epidemiologic, and computer science literature, and the connections between the methods have not been readily apparent. In the setting of a binary instrument, treatment, and outcome, we review proposed methods for partial and point identification of the ATE under IV assumptions, express the identification results in a common notation and terminology, and propose a taxonomy that is based on sets of identifying assumptions. We further demonstrate and provide software for the application of these methods to estimate bounds. Supplementary materials for this article are available online.
Collapse
Affiliation(s)
- Sonja A Swanson
- Department of Epidemiology, Erasmus Medical Center, Rotterdam, The Netherlands.,Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA
| | - Miguel A Hernán
- Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA.,Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA.,Harvard-MIT Division of Health Sciences and Technology, Boston, MA
| | - Matthew Miller
- Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA.,Department of Health Sciences, Northeastern University, Boston, MA
| | - James M Robins
- Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA.,Department of Biostatistics, Harvard T. H. Chan School of Public Health, Boston, MA
| | | |
Collapse
|
11
|
Labrecque J, Swanson SA. Understanding the Assumptions Underlying Instrumental Variable Analyses: a Brief Review of Falsification Strategies and Related Tools. CURR EPIDEMIOL REP 2018; 5:214-220. [PMID: 30148040 PMCID: PMC6096851 DOI: 10.1007/s40471-018-0152-1] [Citation(s) in RCA: 80] [Impact Index Per Article: 13.3] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/21/2022]
Abstract
PURPOSE OF REVIEW Instrumental variable (IV) methods continue to be applied to questions ranging from genetic to social epidemiology. In the epidemiologic literature, discussion of whether the assumptions underlying IV analyses hold is often limited to only certain assumptions and even then, arguments are mostly made using subject matter knowledge. To complement subject matter knowledge, there exist a variety of falsification strategies and other tools for weighing the plausibility of the assumptions underlying IV analyses. RECENT FINDINGS There are many tools that can refute the IV assumptions or help estimate the magnitude or direction of possible bias if the conditions do not hold perfectly. Many of these tools, including both recently developed strategies and strategies described decades ago, are underused or only used in specific applications of IV methods in epidemiology. SUMMARY Although estimating causal effects with IV analyses relies on unverifiable assumptions, the assumptions can sometimes be refuted. We suggest that the epidemiologists using IV analyses employ all the falsification strategies that apply to their research question in order to avoid settings that demonstrably violate a core condition for valid inference.
Collapse
Affiliation(s)
- Jeremy Labrecque
- Department of Epidemiology, Erasmus MC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands
| | - Sonja A. Swanson
- Department of Epidemiology, Erasmus MC, P.O. Box 2040, 3000 CA Rotterdam, The Netherlands
- Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, MA USA
| |
Collapse
|
12
|
Nature as a Trialist?: Deconstructing the Analogy Between Mendelian Randomization and Randomized Trials. Epidemiology 2018; 28:653-659. [PMID: 28590373 DOI: 10.1097/ede.0000000000000699] [Citation(s) in RCA: 107] [Impact Index Per Article: 17.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/26/2022]
Abstract
Mendelian randomization (MR) studies are often described as naturally occurring randomized trials in which genetic factors are randomly assigned by nature. Conceptualizing MR studies as randomized trials has profound implications for their design, conduct, reporting, and interpretation. For example, analytic practices that are discouraged in randomized trials should also be discouraged in MR studies. Here, we deconstruct the oft-made analogy between MR and randomized trials. We describe four key threats to the analogy between MR studies and randomized trials: (1) exchangeability is not guaranteed; (2) time zero (and therefore the time for setting eligibility criteria) is unclear; (3) the treatment assignment is often measured with error; and (4) adherence is poorly defined. By precisely defining the causal effects being estimated, we underscore that MR estimates are often vaguely analogous to per-protocol effects in randomized trials, and that current MR methods for estimating analogues of per-protocol effects could be biased in practice. We conclude that the analogy between randomized trials and MR studies provides further perspective on both the strengths and the limitations of MR studies as currently implemented, as well as future directions for MR methodology development and application. In particular, the analogy highlights potential future directions for some MR studies to produce more interpretable and informative numerical estimates.
Collapse
|
13
|
Causal null hypotheses of sustained treatment strategies: What can be tested with an instrumental variable? Eur J Epidemiol 2018; 33:723-728. [PMID: 29721747 PMCID: PMC6061140 DOI: 10.1007/s10654-018-0396-6] [Citation(s) in RCA: 29] [Impact Index Per Article: 4.8] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/26/2017] [Accepted: 04/03/2018] [Indexed: 01/12/2023]
Abstract
Sometimes instrumental variable methods are used to test whether a causal effect is null rather than to estimate the magnitude of a causal effect. However, when instrumental variable methods are applied to time-varying exposures, as in many Mendelian randomization studies, it is unclear what causal null hypothesis is tested. Here, we consider different versions of causal null hypotheses for time-varying exposures, show that the instrumental variable conditions alone are insufficient to test some of them, and describe additional assumptions that can be made to test a wider range of causal null hypotheses, including both sharp and average causal null hypotheses. Implications for interpretation and reporting of instrumental variable results are discussed.
Collapse
|
14
|
Commentary: Can We See the Forest for the IVs?: Mendelian Randomization Studies with Multiple Genetic Variants. Epidemiology 2018; 28:43-46. [PMID: 27662595 DOI: 10.1097/ede.0000000000000558] [Citation(s) in RCA: 11] [Impact Index Per Article: 1.8] [Reference Citation Analysis] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 12/16/2022]
|
15
|
An introduction to instrumental variable assumptions, validation and estimation. Emerg Themes Epidemiol 2018; 15:1. [PMID: 29387137 PMCID: PMC5776781 DOI: 10.1186/s12982-018-0069-7] [Citation(s) in RCA: 75] [Impact Index Per Article: 12.5] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/08/2017] [Accepted: 01/07/2018] [Indexed: 12/03/2022] Open
Abstract
The instrumental variable method has been employed within economics to infer causality in the presence of unmeasured confounding. Emphasising the parallels to randomisation may increase understanding of the underlying assumptions within epidemiology. An instrument is a variable that predicts exposure, but conditional on exposure shows no independent association with the outcome. The random assignment in trials is an example of what would be expected to be an ideal instrument, but instruments can also be found in observational settings with a naturally varying phenomenon e.g. geographical variation, physical distance to facility or physician’s preference. The fourth identifying assumption has received less attention, but is essential for the generalisability of estimated effects. The instrument identifies the group of compliers in which exposure is pseudo-randomly assigned leading to exchangeability with regard to unmeasured confounders. Underlying assumptions can only partially be tested empirically and require subject-matter knowledge. Future studies employing instruments should carefully seek to validate all four assumptions, possibly drawing on parallels to randomisation.
Collapse
|
16
|
Cook JA, MacLennan GS, Palmer T, Lois N, Emsley R. Instrumental variable methods for a binary outcome were used to informatively address noncompliance in a randomized trial in surgery. J Clin Epidemiol 2017; 96:126-132. [PMID: 29157924 PMCID: PMC5862096 DOI: 10.1016/j.jclinepi.2017.11.011] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.7] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/15/2017] [Revised: 10/24/2017] [Accepted: 11/13/2017] [Indexed: 12/01/2022]
Abstract
Objectives Randomization can be used as an instrumental variable (IV) to account for unmeasured confounding when seeking to assess the impact of noncompliance with treatment allocation in a randomized trial. We present and compare different methods to calculate the treatment effect on a binary outcome as a rate ratio in a randomized surgical trial. Study Design and Setting The effectiveness of peeling versus not peeling the internal limiting membrane of the retina as part of the surgery for a full thickness macular hole. We compared the IV-based estimates (nonparametric causal bound and two-stage residual inclusion approach [2SRI]) with standard treatment effect measures (intention to treat, per protocol and treatment received [TR]). Compliance was defined in two ways (initial and up to the time point of interest). Poisson regression was used for the model-based approaches with robust standard errors to calculate the risk ratio (RR) with 95% confidence intervals. Results Results were similar for 1-month macular hole status across methods. For 3- and 6-month macular hole status, nonparametric causal bounds provided a narrower range of uncertainty than other methods, though still had substantial imprecision. For 3-month macular hole status, the TR estimate was substantially different from the other point estimates. Conclusion Nonparametric causal bound approaches are a useful addition to an IV estimation approach, which tend to have large levels of uncertainty. Methods which allow RRs to be calculated when addressing noncompliance in randomized trials exist and may be superior to standard estimates. Further research is needed to explore the properties of different IV methods in a broad range of randomized controlled trial scenarios.
Collapse
Affiliation(s)
- Jonathan A Cook
- Centre for Statistics in Medicine, Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Botnar Research Centre, Nuffield Orthopaedic Centre, Windmill Road, Oxford, OX3 7LD, UK.
| | - Graeme S MacLennan
- The Centre for Healthcare Randomised Trials (CHaRT), Health Sciences Building, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, UK
| | - Tom Palmer
- Department of Mathematics and Statistics, Fylde College, Lancaster University, Lancaster, LA1 4YF, UK
| | - Noemi Lois
- Wellcome-Wolfson Institute of Experimental Medicine, School of Medicine, Dentistry and Biomedical Sciences, Queens University, 97 Lisburn Road, Belfast, BT9 7BL, UK
| | - Richard Emsley
- Centre for Biostatistics, School of Health Sciences, Manchester Academic Health Science Centre, The University of Manchester, Jean McFarlane Building, Oxford Road, Manchester, M139PL, UK
| |
Collapse
|
17
|
Affiliation(s)
- Miguel A Hernán
- From the Departments of Epidemiology and Biostatistics, Harvard T.H. Chan School of Public Health (M.A.H., J.M.R.), and the Harvard-MIT Division of Health Sciences and Technology (M.A.H.), Boston
| | - James M Robins
- From the Departments of Epidemiology and Biostatistics, Harvard T.H. Chan School of Public Health (M.A.H., J.M.R.), and the Harvard-MIT Division of Health Sciences and Technology (M.A.H.), Boston
| |
Collapse
|
18
|
Moura LMVR, Westover MB, Kwasnik D, Cole AJ, Hsu J. Causal inference as an emerging statistical approach in neurology: an example for epilepsy in the elderly. Clin Epidemiol 2016; 9:9-18. [PMID: 28115873 PMCID: PMC5221551 DOI: 10.2147/clep.s121023] [Citation(s) in RCA: 4] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/23/2022] Open
Abstract
The elderly population faces an increasing number of cases of chronic neurological conditions, such as epilepsy and Alzheimer's disease. Because the elderly with epilepsy are commonly excluded from randomized controlled clinical trials, there are few rigorous studies to guide clinical practice. When the elderly are eligible for trials, they either rarely participate or frequently have poor adherence to therapy, thus limiting both generalizability and validity. In contrast, large observational data sets are increasingly available, but are susceptible to bias when using common analytic approaches. Recent developments in causal inference-analytic approaches also introduce the possibility of emulating randomized controlled trials to yield valid estimates. We provide a practical example of the application of the principles of causal inference to a large observational data set of patients with epilepsy. This review also provides a framework for comparative-effectiveness research in chronic neurological conditions.
Collapse
Affiliation(s)
- Lidia MVR Moura
- Massachusetts General Hospital, Department of Neurology, Epilepsy Service, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
| | - M Brandon Westover
- Massachusetts General Hospital, Department of Neurology, Epilepsy Service, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
| | - David Kwasnik
- Massachusetts General Hospital, Department of Neurology, Epilepsy Service, Boston, MA, USA
| | - Andrew J Cole
- Massachusetts General Hospital, Department of Neurology, Epilepsy Service, Boston, MA, USA
- Harvard Medical School, Boston, MA, USA
| | - John Hsu
- Massachusetts General Hospital, Mongan Institute, Boston, MA, USA
- Harvard Medical School, Department of Medicine, Boston, MA, USA
- Harvard Medical School, Department of Health Care Policy, Boston, MA, USA
| |
Collapse
|
19
|
Chaibub Neto E. Using instrumental variables to disentangle treatment and placebo effects in blinded and unblinded randomized clinical trials influenced by unmeasured confounders. Sci Rep 2016; 6:37154. [PMID: 27869205 PMCID: PMC5116680 DOI: 10.1038/srep37154] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.1] [Reference Citation Analysis] [Abstract] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/07/2016] [Accepted: 10/25/2016] [Indexed: 11/09/2022] Open
Abstract
Clinical trials traditionally employ blinding as a design mechanism to reduce the influence of placebo effects. In practice, however, it can be difficult or impossible to blind study participants and unblinded trials are common in medical research. Here we show how instrumental variables can be used to quantify and disentangle treatment and placebo effects in randomized clinical trials comparing control and active treatments in the presence of confounders. The key idea is to use randomization to separately manipulate treatment assignment and psychological encouragement conversations/interactions that increase the participants' desire for improved symptoms. The proposed approach is able to improve the estimation of treatment effects in blinded studies and, most importantly, opens the doors to account for placebo effects in unblinded trials.
Collapse
|