1
|
Olstad DL, Beall R, Spackman E, Dunn S, Lipscombe LL, Williams K, Oster R, Scott S, Zimmermann GL, McBrien KA, Steer KJD, Chan CB, Tyminski S, Berkowitz S, Edwards AL, Saunders-Smith T, Tariq S, Popeski N, White L, Williamson T, L'Abbé M, Raine KD, Nejatinamini S, Naser A, Basualdo-Hammond C, Norris C, O'Connell P, Seidel J, Lewanczuk R, Cabaj J, Campbell DJT. Healthy food prescription incentive programme for adults with type 2 diabetes who are experiencing food insecurity: protocol for a randomised controlled trial, modelling and implementation studies. BMJ Open 2022; 12:e050006. [PMID: 35168964 PMCID: PMC8852661 DOI: 10.1136/bmjopen-2021-050006] [Citation(s) in RCA: 5] [Impact Index Per Article: 2.5] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Indexed: 12/12/2022] Open
Abstract
INTRODUCTION The high cost of many healthy foods poses a challenge to maintaining optimal blood glucose levels for adults with type 2 diabetes mellitus who are experiencing food insecurity, leading to diabetes complications and excess acute care usage and costs. Healthy food prescription programmes may reduce food insecurity and support patients to improve their diet quality, prevent diabetes complications and avoid acute care use. We will use a type 2 hybrid-effectiveness design to examine the reach, effectiveness, adoption, implementation and maintenance (RE-AIM) of a healthy food prescription incentive programme for adults experiencing food insecurity and persistent hyperglycaemia. A randomised controlled trial (RCT) will investigate programme effectiveness via impact on glycosylated haemoglobin (primary outcome), food insecurity, diet quality and other clinical and patient-reported outcomes. A modelling study will estimate longer-term programme effectiveness in reducing diabetes-related complications, resource use and costs. An implementation study will examine all RE-AIM domains to understand determinants of effective implementation and reasons behind programme successes and failures. METHODS AND ANALYSIS 594 adults who are experiencing food insecurity and persistent hyperglycaemia will be randomised to a healthy food prescription incentive (n=297) or a healthy food prescription comparison group (n=297). Both groups will receive a healthy food prescription. The incentive group will additionally receive a weekly incentive (CDN$10.50/household member) to purchase healthy foods in supermarkets for 6 months. Outcomes will be assessed at baseline and follow-up (6 months) in the RCT and analysed using mixed-effects regression. Longer-term outcomes will be modelled using the UK Prospective Diabetes Study outcomes simulation model-2. Implementation processes and outcomes will be continuously measured via quantitative and qualitative data. ETHICS AND DISSEMINATION Ethical approval was obtained from the University of Calgary and the University of Alberta. Findings will be disseminated through reports, lay summaries, policy briefs, academic publications and conference presentations. TRIAL REGISTRATION NUMBER NCT04725630. PROTOCOL VERSION Version 1.1; February 2022.
Collapse
Affiliation(s)
- Dana Lee Olstad
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Reed Beall
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Eldon Spackman
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Sharlette Dunn
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Lorraine L Lipscombe
- 2Department of Medicine, Women's College Hospital, University of Toronto, Toronto, Ontario, Canada
| | - Kienan Williams
- Indigenous Wellness Core, Alberta Health Services, Calgary, Alberta, Canada
| | - Richard Oster
- Department of Agricultural, Food & Nutritional Sciences, University of Alberta, Edmonton, Alberta, Canada
| | - Sara Scott
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Gabrielle L Zimmermann
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
- Knowledge Translation Platform, Alberta SPOR SUPPORT Unit, Department of Pediatrics, Faculty of Medicine & Dentistry, University of Alberta, Edmonton, Alberta, Canada
| | - Kerry A McBrien
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
- Department of Family Medicine, G012 Health Sciences Centre, 3330 Hospital Drive NW, Calgary, Alberta, Canada
| | - Kieran J D Steer
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Catherine B Chan
- Department of Agricultural, Food & Nutritional Sciences, University of Alberta, Edmonton, Alberta, Canada
- Department of Physiology, University of Alberta, Edmonton, Alberta, Canada
- Diabetes, Obesity and Nutrition Strategic Clinical Network, Alberta Health Services, Calgary, Alberta, Canada
| | - Sheila Tyminski
- Nutrition Services, Alberta Health Services, Edmonton, Alberta, Canada
| | - Seth Berkowitz
- Division of General Medicine and Clinical Epidemiology, Department of Medicine, University of North Carolina at Chapel Hill School of Medicine, Gatineau, Quebec, Canada
| | - Alun L Edwards
- Department of Medicine, Cumming School of Medicine, University of Calgary Foothills Medical Centre, Calgary, Alberta, Canada
| | - Terry Saunders-Smith
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Saania Tariq
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Naomi Popeski
- Diabetes, Obesity and Nutrition Strategic Clinical Network, Alberta Health Services, Calgary, Alberta, Canada
| | - Laura White
- Alberta Region, First Nations and Inuit Health Branch, Indigenous Services Canada, Edmonton, Alberta, Canada
| | - Tyler Williamson
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Mary L'Abbé
- Department of Nutritional Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario, Canada
| | - Kim D Raine
- School of Public Health, University of Alberta, 3-300 Edmonton Clinic Health Academy, Edmonton, Alberta, Canada
| | - Sara Nejatinamini
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - Aruba Naser
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | | | - Colleen Norris
- Faculty of Nursing, University of Alberta, Edmonton, Alberta, Canada
- Cardiovascular Health and Stroke Strategic Clinic Network, Alberta Health Services, Calgary, Alberta, Canada
| | - Petra O'Connell
- Diabetes, Obesity and Nutrition Strategic Clinical Network, Alberta Health Services, Calgary, Alberta, Canada
| | - Judy Seidel
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
- Primary Health Care Integration Network, Primary Health Care, Alberta Health Services, Calgary, Alberta, Canada
| | - Richard Lewanczuk
- Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Calgary, Alberta, Canada
| | - Jason Cabaj
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
| | - David J T Campbell
- Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, Calgary, Alberta, Canada
- Department of Medicine, Cumming School of Medicine, University of Calgary Foothills Medical Centre, Calgary, Alberta, Canada
- Department of Cardiac Sciences, Cumming School of Medicine, University Drive NW, Calgary, Alberta, Canada
| |
Collapse
|
2
|
Kaciroti NA, Little RJA. Bayesian sensitivity analyses for longitudinal data with dropouts that are potentially missing not at random: A high dimensional pattern-mixture model. Stat Med 2021; 40:4609-4628. [PMID: 34405912 DOI: 10.1002/sim.9083] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/18/2020] [Revised: 04/05/2021] [Accepted: 05/10/2021] [Indexed: 11/05/2022]
Abstract
Randomized clinical trials with outcome measured longitudinally are frequently analyzed using either random effect models or generalized estimating equations. Both approaches assume that the dropout mechanism is missing at random (MAR) or missing completely at random (MCAR). We propose a Bayesian pattern-mixture model to incorporate missingness mechanisms that might be missing not at random (MNAR), where the distribution of the outcome measure at the follow-up time t k , conditional on the prior history, differs across the patterns of missing data. We then perform sensitivity analysis on estimates of the parameters of interest. The sensitivity parameters relate the distribution of the outcome of interest between subjects from a missing-data pattern at time t k with that of the observed subjects at time t k . The large number of the sensitivity parameters is reduced by treating them as random with a prior distribution having some pre-specified mean and variance, which are varied to explore the sensitivity of inferences. The missing at random (MAR) mechanism is a special case of the proposed model, allowing a sensitivity analysis of deviations from MAR. The proposed approach is applied to data from the Trial of Preventing Hypertension.
Collapse
Affiliation(s)
- Niko A Kaciroti
- Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, Michigan, USA.,Department of Pediatrics, Medical School, University of Michigan, Ann Arbor, Michigan, USA
| | - Roderick J A Little
- Department of Biostatistics, School of Public Health, University of Michigan, Ann Arbor, Michigan, USA
| |
Collapse
|
3
|
Griswold ME, Talluri R, Zhu X, Su D, Tingle J, Gottesman RF, Deal J, Rawlings AM, Mosley TH, Windham BG, Bandeen-Roche K. Reflection on modern methods: shared-parameter models for longitudinal studies with missing data. Int J Epidemiol 2021; 50:1384-1393. [PMID: 34113988 PMCID: PMC8407871 DOI: 10.1093/ije/dyab086] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Download PDF] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Accepted: 04/01/2021] [Indexed: 11/12/2022] Open
Abstract
A primary goal of longitudinal studies is to examine trends over time. Reported results from these studies often depend on strong, unverifiable assumptions about the missing data. Whereas the risk of substantial bias from missing data is widely known, analyses exploring missing-data influences are commonly done either ad hoc or not at all. This article outlines one of the three primary recognized approaches for examining missing-data effects that could be more widely used, i.e. the shared-parameter model (SPM), and explains its purpose, use, limitations and extensions. We additionally provide synthetic data and reproducible research code for running SPMs in SAS, Stata and R programming languages to facilitate their use in practice and for teaching purposes in epidemiology, biostatistics, data science and related fields. Our goals are to increase understanding and use of these methods by providing introductions to the concepts and access to helpful tools.
Collapse
Affiliation(s)
- Michael E Griswold
- The MIND Center, University of Mississippi Medical Center, 2500 N State Street, Jackson, MS, 39216, USA
| | - Rajesh Talluri
- Department of Data Science, University of Mississippi Medical Center, Jackson, MS, USA
| | - Xiaoqian Zhu
- Department of Data Science, University of Mississippi Medical Center, Jackson, MS, USA
| | - Dan Su
- The MIND Center, University of Mississippi Medical Center, 2500 N State Street, Jackson, MS, 39216, USA
| | - Jonathan Tingle
- The MIND Center, University of Mississippi Medical Center, 2500 N State Street, Jackson, MS, 39216, USA
| | - Rebecca F Gottesman
- Department of Neurology, School of Medicine, Johns Hopkins University, Baltimore, MD, USA
| | - Jennifer Deal
- Department of Epidemiology, Johns Hopkins University, Baltimore, MD, USA
| | - Andreea M Rawlings
- Department of Epidemiology, Johns Hopkins University, Baltimore, MD, USA
| | - Thomas H Mosley
- The MIND Center, University of Mississippi Medical Center, 2500 N State Street, Jackson, MS, 39216, USA
| | - B Gwen Windham
- The MIND Center, University of Mississippi Medical Center, 2500 N State Street, Jackson, MS, 39216, USA
| | - Karen Bandeen-Roche
- Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA
| |
Collapse
|
4
|
Scharfstein DO, Steingrimsson J, McDermott A, Wang C, Ray S, Campbell A, Nunes E, Matthews A. Global sensitivity analysis of randomized trials with nonmonotone missing binary outcomes: Application to studies of substance use disorders. Biometrics 2021; 78:649-659. [PMID: 33728637 PMCID: PMC10392106 DOI: 10.1111/biom.13455] [Citation(s) in RCA: 3] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/30/2019] [Revised: 02/03/2020] [Accepted: 02/24/2021] [Indexed: 11/30/2022]
Abstract
In this paper, we present a method for conducting global sensitivity analysis of randomized trials in which binary outcomes are scheduled to be collected on participants at prespecified points in time after randomization and these outcomes may be missing in a nonmonotone fashion. We introduce a class of missing data assumptions, indexed by sensitivity parameters, which are anchored around the missing not at random assumption introduced by Robins (Statistics in Medicine, 1997). For each assumption in the class, we establish that the joint distribution of the outcomes is identifiable from the distribution of the observed data. Our estimation procedure uses the plug-in principle, where the distribution of the observed data is estimated using random forests. We establish n asymptotic properties for our estimation procedure. We illustrate our methodology in the context of a randomized trial designed to evaluate a new approach to reducing substance use, assessed by testing urine samples twice weekly, among patients entering outpatient addiction treatment. We evaluate the finite sample properties of our method in a realistic simulation study. Our methods have been implemented in an R package entitled slabm.
Collapse
Affiliation(s)
- Daniel O Scharfstein
- Department of Population Health Sciences, University of Utah School of Medicine, Salt Lake City, Utah, USA
| | - Jon Steingrimsson
- Department of Biostatistics, Brown University, Providence, Rhode Island, USA
| | - Aidan McDermott
- Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, USA
| | - Chenguang Wang
- Division of Biostatistics and Bioinformatics, The Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University School of Medicine, Baltimore, Maryland, USA
| | - Souvik Ray
- Department of Statistics, Stanford University, Stanford, California, USA
| | - Aimee Campbell
- Department of Psychiatry, Columbia University Irving Medical Center and New York State Psychiatric Institute, New York, USA
| | - Edward Nunes
- Department of Psychiatry, Columbia University Irving Medical Center and New York State Psychiatric Institute, New York, USA
| | | |
Collapse
|
5
|
Bica I, Alaa AM, Lambert C, van der Schaar M. From Real-World Patient Data to Individualized Treatment Effects Using Machine Learning: Current and Future Methods to Address Underlying Challenges. Clin Pharmacol Ther 2020; 109:87-100. [PMID: 32449163 DOI: 10.1002/cpt.1907] [Citation(s) in RCA: 64] [Impact Index Per Article: 16.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 11/05/2019] [Accepted: 05/14/2020] [Indexed: 12/21/2022]
Abstract
Clinical decision making needs to be supported by evidence that treatments are beneficial to individual patients. Although randomized control trials (RCTs) are the gold standard for testing and introducing new drugs, due to the focus on specific questions with respect to establishing efficacy and safety vs. standard treatment, they do not provide a full characterization of the heterogeneity in the final intended treatment population. Conversely, real-world observational data, such as electronic health records (EHRs), contain large amounts of clinical information about heterogeneous patients and their response to treatments. In this paper, we introduce the main opportunities and challenges in using observational data for training machine learning methods to estimate individualized treatment effects and make treatment recommendations. We describe the modeling choices of the state-of-the-art machine learning methods for causal inference, developed for estimating treatment effects both in the cross-section and longitudinal settings. Additionally, we highlight future research directions that could lead to achieving the full potential of leveraging EHRs and machine learning for making individualized treatment recommendations. We also discuss how experimental data from RCTs and Pharmacometric and Quantitative Systems Pharmacology approaches can be used to not only improve machine learning methods, but also provide ways for validating them. These future research directions will require us to collaborate across the scientific disciplines to incorporate models based on RCTs and known disease processes, physiology, and pharmacology into these machine learning models based on EHRs to fully optimize the opportunity these data present.
Collapse
Affiliation(s)
- Ioana Bica
- University of Oxford, Oxford, UK.,The Alan Turing Institute, London, UK
| | - Ahmed M Alaa
- University of California - Los Angeles, Los Angeles, California, USA
| | - Craig Lambert
- Clinical Pharmacology and Safety Sciences, Research and Development, AstraZeneca, Cambridge, UK
| | - Mihaela van der Schaar
- The Alan Turing Institute, London, UK.,University of California - Los Angeles, Los Angeles, California, USA.,University of Cambridge, Cambridge, UK
| |
Collapse
|
6
|
Aktary ML, Caron-Roy S, Sajobi T, O'Hara H, Leblanc P, Dunn S, McCormack GR, Timmins D, Ball K, Downs S, Minaker LM, Nykiforuk CI, Godley J, Milaney K, Lashewicz B, Fournier B, Elliott C, Raine KD, Prowse RJ, Olstad DL. Impact of a farmers' market nutrition coupon programme on diet quality and psychosocial well-being among low-income adults: protocol for a randomised controlled trial and a longitudinal qualitative investigation. BMJ Open 2020; 10:e035143. [PMID: 32371514 PMCID: PMC7228519 DOI: 10.1136/bmjopen-2019-035143] [Citation(s) in RCA: 3] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Download PDF] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 10/20/2019] [Revised: 02/19/2020] [Accepted: 04/02/2020] [Indexed: 01/04/2023] Open
Abstract
INTRODUCTION Low-income populations have poorer diet quality and lower psychosocial well-being than their higher-income counterparts. These inequities increase the burden of chronic disease in low-income populations. Farmers' market subsidies may improve diet quality and psychosocial well-being among low-income populations. In Canada, the British Columbia (BC) Farmers' Market Nutrition Coupon Programme (FMNCP) aims to improve dietary patterns and health among low-income participants by providing coupons to purchase healthy foods from farmers' markets. This study will assess the impact of the BC FMNCP on the diet quality and psychosocial well-being of low-income adults and explore mechanisms of programme impacts. METHODS AND ANALYSIS In a parallel group randomised controlled trial, low-income adults will be randomised to an FMNCP intervention (n=132) or a no-intervention control group (n=132). The FMNCP group will receive 16 coupon sheets valued at CAD$21/sheet over 10-15 weeks to purchase fruits, vegetables, dairy, meat/poultry/fish, eggs, nuts and herbs at farmers' markets and will be invited to participate in nutrition skill-building activities. Overall diet quality (primary outcome), diet quality subscores, mental well-being, sense of community, food insecurity and malnutrition risk (secondary outcomes) will be assessed at baseline, immediately post-intervention and 16 weeks post-intervention. Dietary intake will be assessed using the Automated Self-Administered 24-hour Dietary Recall. Diet quality will be calculated using the Healthy Eating Index-2015. Repeated measures mixed-effect regression will assess differences in outcomes between groups from baseline to 16 weeks post-intervention. Furthermore, 25-30 participants will partake in semi-structured interviews during and 5 weeks after programme completion to explore participants' experiences with and perceived outcomes from the programme. ETHICS AND DISSEMINATION Ethical approval was obtained from the University of Calgary Conjoint Health Research Ethics Board, Rutgers University Ethics and Compliance, and University of Waterloo Office of Research Ethics. Findings will be disseminated through policy briefs, conference presentations and peer-reviewed publications. TRIAL REGISTRATION NUMBER NCT03952338.
Collapse
Affiliation(s)
- Michelle L Aktary
- Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
| | | | - Tolulope Sajobi
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| | - Heather O'Hara
- British Columbia Association of Farmers' Markets, Vancouver, British Columbia, Canada
| | - Peter Leblanc
- British Columbia Association of Farmers' Markets, Vancouver, British Columbia, Canada
| | - Sharlette Dunn
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| | - Gavin R McCormack
- Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
- School of Architecture, Planning and Landscape, University of Calgary, Calgary, Alberta, Canada
| | - Dianne Timmins
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| | - Kylie Ball
- Institute for Physical Activity and Nutrition (IPAN), Deakin University, Burwood, Victoria, Australia
| | - Shauna Downs
- School of Public Health, Rutgers University, Newark, New Jersey, USA
| | - Leia M Minaker
- School of Planning, University of Waterloo, Waterloo, Ontario, Canada
| | | | - Jenny Godley
- Department of Sociology, University of Calgary, Calgary, Alberta, Canada
| | - Katrina Milaney
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| | - Bonnie Lashewicz
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| | - Bonnie Fournier
- School of Nursing, Thompson Rivers University, Kamloops, British Columbia, Canada
| | - Charlene Elliott
- Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
- Department of Communication Media and Film, University of Calgary, Calgary, Alberta, Canada
| | - Kim D Raine
- School of Public Health, University of Alberta, Edmonton, Alberta, Canada
| | - Rachel Jl Prowse
- School of Public Health, University of Alberta, Edmonton, Alberta, Canada
| | - Dana Lee Olstad
- Faculty of Kinesiology, University of Calgary, Calgary, Alberta, Canada
- Department of Community Health Sciences, University of Calgary, Cumming School of Medicine, Calgary, Alberta, Canada
| |
Collapse
|
7
|
Sadinle M, Reiter JP. Sequentially additive nonignorable missing data modelling using auxiliary marginal information. Biometrika 2019. [DOI: 10.1093/biomet/asz054] [Citation(s) in RCA: 5] [Impact Index Per Article: 1.0] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/12/2022] Open
Abstract
Summary
We study a class of missingness mechanisms, referred to as sequentially additive nonignorable, for modelling multivariate data with item nonresponse. These mechanisms explicitly allow the probability of nonresponse for each variable to depend on the value of that variable, thereby representing nonignorable missingness mechanisms. These missing data models are identified by making use of auxiliary information on marginal distributions, such as marginal probabilities for multivariate categorical variables or moments for numeric variables. We prove identification results and illustrate the use of these mechanisms in an application.
Collapse
Affiliation(s)
- Mauricio Sadinle
- Department of Biostatistics, University of Washington, Seattle, Washington, U.S.A
| | - Jerome P Reiter
- Department of Statistical Science, Duke University, 214 Old Chemistry Building, Durham, North Carolina, U.S.A
| |
Collapse
|
8
|
Fitzmaurice GM, Lipsitz SR, Weiss RD. Sensitivity analysis for non-monotone missing binary data in longitudinal studies: Application to the NIDA collaborative cocaine treatment study. Stat Methods Med Res 2019; 28:3057-3073. [PMID: 30146938 PMCID: PMC6393220 DOI: 10.1177/0962280218794725] [Citation(s) in RCA: 2] [Impact Index Per Article: 0.4] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/16/2022]
Abstract
Conventional approaches for handling missingness in substance use disorder trials commonly rely upon a single deterministic "worst value" imputation that posits a perfect relationship between missingness and drug use ("missing value = presumed drug use"); this yields biased estimates of treatment effects and their standard errors. Instead, deterministic imputations should be replaced by probabilistic versions that encode researchers prior beliefs that those with missing data are more likely to be using drugs at those occasions. Motivated by this problem, we present a method for handling non-monotone missing binary data in longitudinal studies. Specifically, we consider a joint model that combines a not missing at random (NMAR) selection model with a generalized linear mixed model for longitudinal binary data. The selection model links the distribution of a missing outcome to the corresponding distribution of the outcome for those observed at that occasion via a fixed and known sensitivity parameter. The mixed model for longitudinal binary data assumes the random effects have bridge distributions; the latter yields regression parameters that have both subject-specific and marginal interpretations. This approach is completely transparent about what is being assumed about missing data and can be used as the basis for sensitivity analysis.
Collapse
Affiliation(s)
- Garrett M Fitzmaurice
- Division of Alcohol and Drug Abuse, McLean Hospital, Belmont, MA, USA
- Department of Psychiatry, Harvard Medical School, Boston, MA, USA
| | - Stuart R Lipsitz
- Division of General Internal Medicine, Brigham and Women's Hospital, Boston, MA, USA
- Department of Medicine, Harvard Medical School, Boston, MA, USA
| | - Roger D Weiss
- Division of Alcohol and Drug Abuse, McLean Hospital, Belmont, MA, USA
- Department of Psychiatry, Harvard Medical School, Boston, MA, USA
| |
Collapse
|
9
|
Abstract
In this article, I review the key elements of the proposed International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use E9 Addendum, present a constructive critique, and provide recommendations of how it can be improved. To highlight ideas, I present a case study involving a confirmatory trial for a chronic pain medication.
Collapse
Affiliation(s)
- Daniel O Scharfstein
- Department of Biostatistics, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
| |
Collapse
|
10
|
Li P, Stuart EA. Best (but oft-forgotten) practices: missing data methods in randomized controlled nutrition trials. Am J Clin Nutr 2019; 109:504-508. [PMID: 30793174 PMCID: PMC6408317 DOI: 10.1093/ajcn/nqy271] [Citation(s) in RCA: 34] [Impact Index Per Article: 6.8] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 05/11/2018] [Revised: 10/22/2017] [Accepted: 09/11/2018] [Indexed: 11/13/2022] Open
Abstract
Missing data ubiquitously occur in randomized controlled trials and may compromise the causal inference if inappropriately handled. Some problematic missing data methods such as complete case (CC) analysis and last-observation-carried-forward (LOCF) are unfortunately still common in nutrition trials. This situation is partially caused by investigator confusion on missing data assumptions for different methods. In this statistical guidance, we provide a brief introduction of missing data mechanisms and the unreasonable assumptions that underlie CC and LOCF and recommend 2 appropriate missing data methods: multiple imputation and full information maximum likelihood.
Collapse
Affiliation(s)
- Peng Li
- School of Nursing, University of Alabama at Birmingham, Birmingham, AL
| | - Elizabeth A Stuart
- Departments of Mental Health
- Biostatistics
- Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD
| |
Collapse
|
11
|
Linero AR, Daniels MJ. Bayesian Approaches for Missing Not at Random Outcome Data: The Role of Identifying Restrictions. Stat Sci 2018; 33:198-213. [PMID: 31889740 PMCID: PMC6936760 DOI: 10.1214/17-sts630] [Citation(s) in RCA: 22] [Impact Index Per Article: 3.7] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/19/2022]
Abstract
Missing data is almost always present in real datasets, and introduces several statistical issues. One fundamental issue is that, in the absence of strong uncheckable assumptions, effects of interest are typically not nonparametrically identified. In this article, we review the generic approach of the use of identifying restrictions from a likelihood-based perspective, and provide points of contact for several recently proposed methods. An emphasis of this review is on restrictions for nonmonotone missingness, a subject that has been treated sparingly in the literature. We also present a general, fully-Bayesian, approach which is widely applicable and capable of handling a variety of identifying restrictions in a uniform manner.
Collapse
|
12
|
Scharfstein DO, McDermott A. Global sensitivity analysis of clinical trials with missing patient-reported outcomes. Stat Methods Med Res 2018; 28:1439-1456. [DOI: 10.1177/0962280218759565] [Citation(s) in RCA: 5] [Impact Index Per Article: 0.8] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Indexed: 11/15/2022]
Abstract
Randomized trials with patient-reported outcomes are commonly plagued by missing data. The analysis of such trials relies on untestable assumptions about the missing data mechanism. To address this issue, it has been recommended that the sensitivity of the trial results to assumptions should be a mandatory reporting requirement. In this paper, we discuss a recently developed methodology (Scharfstein et al., Biometrics, 2018) for conducting sensitivity analysis of randomized trials in which outcomes are scheduled to be measured at fixed points in time after randomization and some subjects prematurely withdraw from study participation. The methodology is explicated in the context of a placebo-controlled randomized trial designed to evaluate a treatment for bipolar disorder. We present a comprehensive data analysis and a simulation study to evaluate the performance of the method. A software package entitled SAMON (R and SAS versions) that implements our methods is available at www.missingdatamatters.org .
Collapse
Affiliation(s)
| | - Aidan McDermott
- Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA
| |
Collapse
|