Epidemiology research.

Abstract

Since research in medicine, especially in such specialties as epidemiology, is done more and more by interdisciplinary teams, of whom at least some have no medical background, David Grimes and Kenneth Schulz's report (Jan 5, p 57)1Grimes DA Schulz KF An overview of clinical research: the lay of the land.Lancet. 2002; 359: 57-61Summary Full Text Full Text PDF PubMed Scopus (405) Google Scholar on different research designs could benefit from a synopsis of terms typically used in social sciences2Campbell DT Stanley JC Experimental and quasi-experimental designs for research. Rand McNally, Chicago1966Google Scholar to describe different research designs. This glossary (panel) might help more precise communication in such groups, and perhaps avoid some confusion.PanelTabled 1Terms used by Grimes and SchulzTerms typically used in social sciencesRandomised controlled trial(True)experimental designNon-randomised controlled trialPseudorandomised experimental designAnalytical study (with comparison group)Quasiexperimental designDescriptive study (without comparison group)Pre-experimental design Open table in a new tab Not all terms are readily understood by our colleagues in other sciences. Quasi-experimental design, for example, sounds at least strange, if not wrong, to people with a medical-based background. On the other hand, most social scientists do not know what precisely is meant by analytical study.Another possible source of confusion is Grimes and Schulz's definition of internal validity. Social scientists generally strictly differentiate between the use of the term validity in the context of tests or questionnaires, and in the context of research designs. Although Grimes and Schulz seem to mean the same thing, they state that internal validity means that a trial should measure what it sets out to measure. That is the definition that psychologists3Nunnally JC Bernstein IH Psychometric theory. 3rd edn. McGraw-Hill, New York1994Google Scholar or quality-of-life researchers4Streiner DL Norman GR Health measurement scales: a practical guide to their development and use. 2nd edn. Oxford University Press, New York1995Google Scholar use only in connection with the validity of tests or questionnaires.The term internal validity, in the context of research designs, means for social scientists how good possible outcome differences between an index and a control (or comparison) group can be causally traced back to the intervention (treatment, exposure) examined, or whether the results could, for example, be equally well explained by known or unknown other confounding variables. To show treatment efficacy and effectiveness, the randomised controlled trial (or true experimental design) is indeed the best choice to this end. Since research in medicine, especially in such specialties as epidemiology, is done more and more by interdisciplinary teams, of whom at least some have no medical background, David Grimes and Kenneth Schulz's report (Jan 5, p 57)1Grimes DA Schulz KF An overview of clinical research: the lay of the land.Lancet. 2002; 359: 57-61Summary Full Text Full Text PDF PubMed Scopus (405) Google Scholar on different research designs could benefit from a synopsis of terms typically used in social sciences2Campbell DT Stanley JC Experimental and quasi-experimental designs for research. Rand McNally, Chicago1966Google Scholar to describe different research designs. This glossary (panel) might help more precise communication in such groups, and perhaps avoid some confusion. Tabled 1Terms used by Grimes and SchulzTerms typically used in social sciencesRandomised controlled trial(True)experimental designNon-randomised controlled trialPseudorandomised experimental designAnalytical study (with comparison group)Quasiexperimental designDescriptive study (without comparison group)Pre-experimental design Open table in a new tab Not all terms are readily understood by our colleagues in other sciences. Quasi-experimental design, for example, sounds at least strange, if not wrong, to people with a medical-based background. On the other hand, most social scientists do not know what precisely is meant by analytical study. Another possible source of confusion is Grimes and Schulz's definition of internal validity. Social scientists generally strictly differentiate between the use of the term validity in the context of tests or questionnaires, and in the context of research designs. Although Grimes and Schulz seem to mean the same thing, they state that internal validity means that a trial should measure what it sets out to measure. That is the definition that psychologists3Nunnally JC Bernstein IH Psychometric theory. 3rd edn. McGraw-Hill, New York1994Google Scholar or quality-of-life researchers4Streiner DL Norman GR Health measurement scales: a practical guide to their development and use. 2nd edn. Oxford University Press, New York1995Google Scholar use only in connection with the validity of tests or questionnaires. The term internal validity, in the context of research designs, means for social scientists how good possible outcome differences between an index and a control (or comparison) group can be causally traced back to the intervention (treatment, exposure) examined, or whether the results could, for example, be equally well explained by known or unknown other confounding variables. To show treatment efficacy and effectiveness, the randomised controlled trial (or true experimental design) is indeed the best choice to this end. Epidemiology researchAuthors' reply Full-Text PDF