Design and statistical analysis of oral medicine studies: common pitfalls

Authors

  • L Baccaglini,

    1. Department of Community Dentistry & Behavioral Science, College of Dentistry, University of Florida, Gainesville, FL
    Search for more papers by this author
  • JJ Shuster,

    1. Division of Biostatistics, Department of Epidemiology and Health Policy Research, College of Medicine, University of Florida, Gainesville, FL
    2. Shands Clinical Research Unit, Clinical and Translational Science Institute, University of Florida, Gainesville, FL
    Search for more papers by this author
  • J Cheng,

    1. Division of Biostatistics, Department of Epidemiology and Health Policy Research, College of Medicine, University of Florida, Gainesville, FL
    Search for more papers by this author
  • DW Theriaque,

    1. Regulatory Knowledge and Research Support, Clinical and Translational Science Institute, University of Florida, Gainesville, FL
    Search for more papers by this author
  • VJ Schoenbach,

    1. Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC, USA
    Search for more papers by this author
  • SL Tomar,

    1. Department of Community Dentistry & Behavioral Science, College of Dentistry, University of Florida, Gainesville, FL
    Search for more papers by this author
  • C Poole

    1. Department of Epidemiology, Gillings School of Global Public Health, University of North Carolina, Chapel Hill, NC, USA
    Search for more papers by this author

Lorena Baccaglini, DDS, DDS, MS, PhD, Department Epidemiology and Biostatistics, College of Public Health and Health Professions, Department of Community Dentistry & Behavioral Science, University of Florida College of Dentistry, 1329 SW 16th Street, Suite 5182, PO Box 103628, Gainesville, FL 32610 3628, USA. Tel: (352) 273 5962, Fax: (352) 273 5985, E-mail: lbaccaglini@dental.ufl.edu

Abstract

Oral Diseases (2010) 16, 233–241

A growing number of articles are emerging in the medical and statistics literature that describe epidemiologic and statistical flaws of research studies. Many examples of these deficiencies are encountered in the oral, craniofacial, and dental literature. However, only a handful of methodologic articles have been published in the oral literature warning investigators of potential errors that may arise early in the study and that can irreparably bias the final results. In this study, we briefly review some of the most common pitfalls that our team of epidemiologists and statisticians has identified during the review of submitted or published manuscripts and research grant applications. We use practical examples from the oral medicine and dental literature to illustrate potential shortcomings in the design and analysis of research studies, and how these deficiencies may affect the results and their interpretation. A good study design is essential, because errors in the analysis can be corrected if the design was sound, but flaws in study design can lead to data that are not salvageable. We recommend consultation with an epidemiologist or a statistician during the planning phase of a research study to optimize study efficiency, minimize potential sources of bias, and document the analytic plan.

Ancillary