Graduate education at the Faculty of Medicine of the University of São Paulo: Quo vadis?

EDITORIAL

Graduate education at the Faculty of Medicine of the University of São Paulo: Quo vadis?

Riad N. Younes^{I, II}; Daniel Deheinzelin^II; Dario Birolini^I

^IHospital das Clínicas, Faculty of Medicine, University of São Paulo

^IIHospital do Câncer AC Camargo - São Paulo/SP, Brazil. E-mail: [e-mail]

The aims of sensu strictu graduate education, as stated in the prospectus of the Faculty of Medicine of the University of São Paulo (FMUSP), are to form researchers with a broad view of their field of knowledge, and, at the same time, to be a system that generates knowledge in those fields. Undoubtedly lofty objectives, coherent with the general mission of universities, University of São Paulo being no exception: the continuous search for scientific knowledge, and the task of spreading the results of research to the public at large. Communication of knowledge and science is an essential part of the process. In fact, the number of Brazilian authors and their publications in international scientific literature has increased substantially¹. Brazil now stands as the 25^th science producer in the world (ISI Web of Science).

In the above scenario, it becomes clear that the graduate education process does not end at thesis presentation. It should be followed by the publication of the results reached during that particular study. Moreover, in any particular case, if the objectives of the program have been achieved, knowledge has been generated and a system has been established in the particular field, from which a series of other publications from the same author(s) should follow.

Is this the case at the FMUSP? In order to determine this, we have reviewed the fate of 1181 doctoral thesis presented between 1990 and 2000 at that Institution. In order to assess whether publications resulted from the respective authors in the years following thesis presentation, national and international publications databases (e.g., Medline, LILACS) were searched using the 'author' field in each database. The overall number of examined theses increased from 61 in 1990, to 158 in 2000, with a peak of 198 in 1999. Figure 1 depicts the number of doctorate holders remaining without any publication, either national or international, 1 -10 years after the thesis completion.

One may clearly see that publication only starts two years after completion of the theses. A similar study conducted recently in France, regarding an equivalent situation, disclosed rather different results. Twenty-seven percent of the theses appeared in print in the first year, nearly 50% after two years². This suggests a failure in our system or, at least, a missed window of opportunity. Our study shows that, after five years, more than 50% of the post graduates have not published a single study, not even a review article.

Such a low publication rate could be explained by the fact that national journals are underrepresented in the Journal of Citation Reports and that most of the successfully graduated PhDs are publishing in local journals. However, it has recently been shown that the number of Brazilian research published in internationally indexed journals has increased significantly³. Moreover, by adding the articles published in SciELO indexed journals, a Brazilian initiative that allows for the on line retrieval and consultation of many more articles, publications increase by 33%. In spite of this, bibliographic production recorded by the research directory of the Brazilian National Research Council (CNPq) is three times higher³. In other words what we found at FMUSP seems to happen all over the country, irrespective of the type of journals searched. It must be pointed out that the intention of publication in the medical field is to spread knowledge. As such, the value of publishing in a journal that cannot be retrieved is questionable.

One could speculate on likely explanations. One possibility is that, since theses are parts of research lines, authors, particularly the senior researcher, may be waiting for more data to convey a more convincing message. However, what probably happens is the exact contrary. Since the evaluation of researchers relies in the number of published articles, the tendency is to submit articles as soon as possible, to avoid loss of "power" and impact⁴. Such rush to print certainly endangers quality; however it does not appear to be a factor in FMUSP, as far as our discussion takes us.

Lack of institutional endorsement and highlighting may be factors that explain the observed results. This possibility was studied in Croatia, where 34% of Ph.D. theses over ten years resulted in MEDLINE articles. But a significant difference was found between two different Croatian Universities, suggesting different institutional stress directed towards publication⁵.

One might argue that the results were in fact presented at international symposia or congresses, but subsequently rejected for publication. Probability of publication of articles derived from abstracts has been shown to be related to presentation in small meetings, to selection for oral presentation and to presentation in meetings in the United States. Moreover, abstracts of basic science and abstracts describing positive results have a higher probability of full publication⁶. In the case of FMUSP, even if the results were presented at meetings, this did not result in more publications. This conclusion is based on the observed plateau in publication rates four years after thesis presentation.

In Germany, almost 90% of all practicing physicians present themselves for doctorates⁷. Three main reasons are appointed to explain this: (i) there is a general belief that a doctorate will help in successful job hunting; (ii) most physicians believe that the doctorate will help them in their continuing medical education and (iii) the opportunity to become familiar with new experimental methods or even specialization in particular medical subjects attracts students. However, not all medical theses result in publication. One study conducted in Heidelberg⁸ showed that publication was influenced by three factors: a positive result, the publishing activity of the dissertation tutor, and the quality of the statistical analysis.

It may concluded that the very low publishing rates found at FMUSP may benefit from such a comparative analysis. While the quality of statistical analysis could be easily improved, we believe that the publishing activity of the dissertation tutor may be the point that deserves most attention. The evaluation of graduate education programs, conducted by the Education Ministry through its Coordenadoria de Aprimoramento de Pessoal de Ensino Superior (CAPES) is based, in part, on the number of published articles of the tutors and, more importantly, on the number of articles published in conjunction with the graduate students. Although there is still heterogeneity in the figures related to a recently published evaluation⁹, there seems to be no doubt about the need of such criteria. The numbers disclosed by this study may explain why most of the programs of FMUSP did not attain the highest grades in such evaluation. Clearly here is another window of opportunity to be explored by the FMUSP.

As a bottom line, the more likely explanation may lie in the graduate education system itself. The whole system continues to work as an extension of medical education. The concept of graduate education as an initiation for researchers, although theoretically accepted, is not what is being practiced. And worse, this is common knowledge in the system¹⁰.

Although desirable, no one expects that all doctoral theses produce truly innovative research. The need of investigators capable of understanding and tuned up to the newest developments is essential to any country and should be one of the aims of any graduate education system. Unfortunately, if these "middle category" researchers do not participate in the publishing system, we cannot ensure that such a goal will ever be achieved. Certainly, the lack of publication as evidenced here compromises the quality of our medical education as a whole¹¹.

REFERENCES

1. de Meis L, Velloso A, D Lannes, Carmo de Meis C. The growing competition in Brazilian science: rites of passage, stress and burnout. Braz J Med Biol Res 2003;36(9):1135-1141.

2. Salmi LR, Gana S, Mouillet E. Publication pattern of medical thesis, France, 1993-98. Med Educ. 2001;35(1):18-21.

3. Coura JR, Willcox Lde C. Impact factor, scientific production and quality of Brazilian medical journals. Mem Inst Oswaldo Cruz. 2003;98(3):293-7.

4. Lawrence PA. The politics of publication. Nature. 2003; 422(6929):259-61.

5. Frkovic V, Skender T, Dojcinovic B, Bilic-Zulle L. Publishing scientific papers based on Master's and Ph.D. theses from a small scientific community: case study of Croatian medical schools. Croat Med J. 2003;44(1):107-11.

6. von Elm E, Costanza MC, Walder B, Tramer MR. More insight into the fate of biomedical meeting abstracts: a systematic review. BMC Med Res Methodol. 2003;3(1):12.

7. Diez C, Arkenau C, Meyer-Wentrup F. The German medical dissertation—time to change? Acad Med. 2000;75(8):861-3.

8. Vogel U, Windeler J. Factors modifying frequency of publications of clinical research results exemplified by medical dissertations. Dtsch Med Wochenschr. 2000;125(5):110-3.

9. Marchini JS, Leite JP, Velasco IT. Avaliação da Pós Graduação da Capes: homogenia ou heterogenia. Infocapes 2001; 9:7-16

10. Borges DR. Postgraduate studies in the medical field. Rev Assoc Med Bras. 1994;40(4):271-5.

11. Does research make for better doctors? Lancet.1993;342:1063-64.

Publication Dates

• Publication in this collection
  01 Mar 2005
• Date of issue
  Feb 2005