Abstract
The ‘citation classic’ notion was coined by Eugene Garfield at the Institute for Scientific Information (ISI) as a means of understanding citation behaviour. Later on, he termed the 15 000 most-cited papers of all-time ‘superstar papers’ [1].
We used the ISI database [2] to locate the 20 most cited articles (1945–2006) in the 10 psychiatric journals of highest impact [3]. From the full reports, we extracted details of studies. By identifying characteristics of such extreme outliers we expected to gain insight into the status of the field in the last four decades.
Results are shown in Table 1. Apart from a few important theoretical contributions (on the catecholamine hypothesis of depression, stress and social support, the pathogenesis of schizophrenia), population-based studies on the prevalence of mental disorders and empirical studies (on dementia, the laboratory diagnosis of melancholia and the treatment of resistant schizophrenia), the majority are validation studies of either rating scales or diagnostic interviews.
Characteristics of the 20 most cited articles in the top 10 psychiatric journals
These superstars appear to conform well to Mayes and Horwitz's recent historical essay [4]. Thus, the development of rating scales in the 1960s was an important step towards a change to a paradigm that equated measurable symptoms with the presence of disease. Reliable interviews, developed in the 1970s, allowed for the introduction of DSM-III, which had the explicit goal of increasing the reliability of diagnoses; in 1980, nationwide prevalence surveys followed. This citation behaviour could be interpreted as an indication of success in the attempt of research-oriented psychiatrists – mainly based at Washington and Columbia University – to bring the profession into the mainstream of scientific inquiry by redesigning views of mental illnesses as symptom-based entities.