Missing Data in Longitudinal Clinical Trials Part A: Design and Conceptual Issues
Authors:
- Philip W. LavoriPLPhilip W. Lavori
- C. Hendricks BrownCBC. Hendricks Brown
- Naihua DuanNDNaihua Duan
- Robert D. GibbonsRGRobert D. Gibbons
- Joel GreenhouseJGJoel Greenhouse
- 1 December 2008
- journal article
- Published by SLACK, Inc. in Psychiatric Annals
- Vol. 38 (12) , 784-792
Abstract
There are numerous types of missing data that can occur in clinical trials. Some types of missing data cannot be prevented and are beyond the research team’s control. For example, a patient may relocate and be unavailable for an assessment. Other types occur because the research team actually designs the study to generate incomplete data. For example, it may not be cost-effective to obtain a full diagnostic assessment on each subject. Instead, an inexpensive screening tool is used to assess everyone. All screen positives, plus a random sample of screen negatives, are then given the full diagnostic. The incomplete diagnostic data on the majority of those screened negative are taken into account in assessing treatment effects.1-3 These types of data are missing by design, and there are statistical procedures to handle such planned missing data if designed appropriately. The third type occurs because of a faulty design plan by the research team. In this case, data are lost because a less than adequate protocol is followed. For example, if no screen negatives are given the full diagnostic, it will not be possible to ascertain the false negatives. ABOUT THE AUTHORS Philip W. Lavori, PhD, is Professor of Biostatistics, and Chair, Department of Health Research and Policy, Stanford University School of Medicine. C. Hendricks Brown, PhD, is with the Prevention Science and Methodology Group, Departments of Epidemiology and Biostatistics, College of Public Health, University of South Florida, Tampa. Naihua Duan, PhD, is Professor of Biostatistics in Psychiatry, Departments of Biostatistics and Psychiatry, Columbia University, New York; and Director, Division of Biostatistics, N.Y. Psychiatric Institute. Robert D. Gibbons, PhD, is Professor of Biostatistics and Psychiatry, and Director of the Center for Health Statistics, University of Illinois at Chicago. Joel Greenhouse, PhD, is with the Department of Statistics, Carnegie Mellon University, Pittsburgh. Address correspondence to: Philip Lavori, HRP/Redwood Building, 259 Campus Drive, Stanford, CA 94305-5405; fax 650-725-6951; or e-mail lavori@stanford.edu. Dr. Lavori, Dr. Brown, Dr. Duan, Dr. Gibbons, and Dr. Greenhouse have disclosed no relevant financial relationships. EDUCATIONAL OBJECTIVES Describe problems associated with missing data in psychiatric research. Define how missing data can be minimized by improved experimental designs. Identify methods that should be avoided when analyzing studies with missing data.Keywords
This publication has 16 references indexed in Scilit:
- Methods for testing theory and evaluating impact in randomized field trials: Intent-to-treat analyses for integrating the perspectives of person, place, and timeDrug and Alcohol Dependence, 2008
- The design versus the analysis of observational studies for causal effects: parallels with the design of randomized trialsStatistics in Medicine, 2006
- Analyzing incomplete longitudinal clinical trial dataBiostatistics, 2004
- A mixed‐effects multinomial logistic regression modelStatistics in Medicine, 2003
- Beyond Intention to TreatEpidemiologic Reviews, 2002
- A case study on the choice, interpretation and checking of multilevel models for longitudinal binary outcomesBiostatistics, 2001
- The efficiency of two-phase designs in prevalence surverys of mental disordersPsychological Medicine, 1990
- Statistics and Causal InferenceJournal of the American Statistical Association, 1986
- Bayesian Inference for Causal Effects: The Role of RandomizationThe Annals of Statistics, 1978
- Errors of the Third Kind in Statistical ConsultingJournal of the American Statistical Association, 1957