Scientific publications from departments of clinical physiology and nuclear medicine in Denmark. A bibliometric analysis of 'impact' in the years 1989-1994

This study reports a bibliometric analysis of scientific publications emanating from departments of clinical physiology and nuclear medicine, Denmark, during the years 1989-1994. The total number of publications during this period was 860 (763 scientific journal papers, 71 book/book chapters and 26 theses). Whereas the number of publications per year (188-113) decreased significantly with time (r = -0.94, P < 0.02), the number of authors (mean 4.1) was almost constant over time. University/university-related departments accounted for 96% of the papers. Only 8% of the papers resulted from a collaboration between two or more departments of clinical physiology and nuclear medicine, but the collaboration with other medical specialities and institutions was much greater (85%). The 763 papers were published in 239 different scientific journals, 80% in journals with an official 'impact factor', a bibliometric measure of quality (the average number of times a paper is cited in a journal in the publishing year and the subsequent year). Twenty per cent (20%) and 8.4% were printed in journals with an impact factor, respectively, of above 2.1 (the 500 journals most cited) and 3.7 (the 200 most cited), which is significantly above the national average (16.6% and 6.0%, P < 0.001). The 'cumulated impact' (i.e. the impact of all papers) showed a borderline significant decrease over time (r = -0.77, P = 0.1), whereas the average impact per paper (1.53) remained almost constant and was significantly above the national average (1.10, P < 0.001). A close relationship was found between the number of papers from a department and its cumulated impact (r = -0.97, P < 0.001). It is concluded that the total number of scientific papers from Danish departments of clinical physiology and nuclear medicine fell in the period, whereas the volume of quality, as assessed on the cumulated impact, only fell with borderline significance, and the impact per paper was almost constant from 1989 to 1994, and was above the national average.