LAUSR

Countless textbooks in epidemiology and medical statistics describe characteristics of descriptive and experimental clinical studies. Early postwar textbooks tended to focus on core elements and principles on how to plan and execute clinical research, and provided examples of how data from weak study methodologies have the potential to cause a range of fallacies and harm (Bradford Hill, 1971). The naming of the different strategies adopted to assemble study participants and accumulate data combined with the approaches to draw statistical inferences has come later. Unfortunately, varieties of different terms used to describe clinical research have emerged, which creates confusing terminology recognized two decades ago (Gordis, 1996) that is persistent even today (Celentano & Szklo, 2018). Moreover, multiple examples exist where the text does not reflect the alleged study design implemented in the title in the materials and methods section of the paper. One major impetus for the need to characterize different clinical study designs was in the 60s to facilitate searching in emerging digital bibliographic databases, of which the predecessor of MEDLINE and PubMed, ie, MEDLARS, was the most important. Librarians could efficiently retrieve titles by combining so‐called Medical Subject Headings (MeSH) in Boolean expressions. Many authors today include in the title of their papers the MESH terms introduced in 1965 to describe persons, and not necessarily patients, to report outcomes or characteristics of patients that have received therapy. It is difficult and time‐consuming to assess the proportion of titles that reflects a study population consisting of patients that have received therapy amongst the papers where the title contains the term “prospective study” (n = 32,499 papers in PubMed), “retrospective study” (n = 22,249), and “follow‐up study”(n = 17,953). These terms became subheadings of “cohort studies” when this term was introduced as a MESH‐term in 1989. The term “case reports” (n = 15,778) was introduced as a type of publication in 1966 and does not require any description of outcomes of therapy but is only described as: “clinical presentations that may be followed by evaluative studies that eventually lead to a diagnosis.” (URL: https://www.ncbi.nlm.nih.gov/mesh/68002363). “Case series” (n = 16,423) is an intriguing term for several reasons, although it has never been a MEDLINE MESH‐term or publication