DÄ internationalArchive7/2009Statistical Methods in Medical Research
LNSLNS Medical, especially patient-centered research is held to high ethical, legal, methodological, and scientific standards. The purpose of patient-centered research is to improve patient care, and increasingly, evidence for the efficacy of diagnostic and therapeutic interventions is required. Evidence-based medicine means that treatment decisions should be based on data arising from convincing, methodologically sound studies, and no longer simply on individual clinicians' experience and preferences. Clinical guidelines are based on this type of evidence, and include the declaration of levels of evidence. The highest levels of evidence are attained by meta-analyses of published data from controlled trials, followed by published results from individual multicenter, randomized, controlled studies. In order to make a judgment about the relevance of research results for patients, clinicians need to understand the design and analysis of a study, as well as its strengths and limitations.

Understanding scientific studies
The aim of the following series of articles on statistical methods in medical research is to promote understanding of scientific articles. We will be presenting key methodological insights into study type and design, the selection of the correct type of study, and avoiding bias and errors. We will present descriptive data analysis methods and statistical tests, as well as basic statistical ideas such as p values, confidence intervals, power calculation, and ways of correcting for multiple testing. We will also deal with the interpretation, application, and special characteristics of trials, epidemiological studies, reviews, and meta-analyses.

A good grasp of methodology puts a critical reader in the position of being able to interpret the results of a study appropriately and to apply them to patient care. We chose the title "Statistical Methods in Medical Research" because statistical methodology underlies and unites a range of types of research, encompassing both clinical trials and epidemiological studies, for example.

The Institute of Medical Biostatistics, Epidemiology, and Informatics (IMBEI) has worked with the Interdisciplinary Center for Clinical Trials (Interdisziplinäres Zentrum Klinische Studien, IZKS), the Center for Pediatric Prevention (Zentrum für Präventive Pädiatrie) at the Department of Pediatrics (all of which are departments of the University Medical Center at the Johannes-Gutenberg University of Mainz), and the editorial team at Deutsches Ärzteblatt to craft a series of articles which we hope present the content, context, planning, implementation, and interpretation of trials and observational studies in an accessible manner.

The series is based loosely around a series of themes which touch on the critical appraisal of studies, the meaning and correct presentation of key statistical values, the correct choice and use of statistical tests, as well as possible pitfalls for studies.

February should see the publication of an article on "Critical appraisal of scientific articles," followed by "Study design in medical research" in March. The articles build on one another; and important points will be repeated, quite deliberately.

This series cannot hope to, and is not intended to, replace essential collaboration during the planning, execution, interpretation, and publication process with statisticians and other experts in research methodology. Instead it is intended as an accessible but well founded introduction to research methods.

Conflict of interest statement
The authors declare no conflict of interest in the terms of the guidelines of the International Committee of Medical Journal Editors.

Translated from the original German by Dr Sandra Goldbeck-Wood.

Dtsch Arztebl Int 2009; 106(7): 99
DOI: 10.3238/arztebl.2009.0099

Corresponding author
Dr. rer. nat. Bernd Röhrig
MDK Rheinland-Pfalz
Referat Rehabilitation/Biometrie
Albiger Str. 19 d, 55232 Alzey, Germany
Bernd.Roehrig@mdk-rlp.de