1135

Braz J Med Biol Res 36(9) 2003

Science funding and scientific productivity in Brazil

The growing competition in Brazilianscience: rites of passage, stress andburnout

1Departamento de Bioquímica Médica, Instituto de Ciências Biomédicas,Centro de Ciências da Saúde and 2Instituto de Psiquiatria,Universidade Federal do Rio de Janeiro, Rio de Janeiro, RJ, Brasil

L. de Meis1, A. Velloso1,D. Lannes1, M.S. Carmo1

and C. de Meis2

Abstract

Brazil’s scientific community is under pressure. Each year there is anincrease in its contribution to international science and in the numberof students who are trained to do research and teach at an advancedlevel. Most of these activities are carried out in state and federaluniversities, but with government funding that has decreased by morethan 70% since 1996. Interviews with graduate students, post-doctoralfellows and professors in one university department with a strongresearch tradition illustrate the level of stress engendered by theconflict between increasing competition and diminishing resources,and serve to underscore the negative effects on creativity and on thetendency to choose science as a career.

CorrespondenceL. de Meis

Departamento de Bioquímica Médica

Instituto de Ciências Biomédicas

CCS, UFRJ

Rio de Janeiro, RJ

Brasil

Fax: +55-21-2270-8647

E-mail: demeis@biqmed.ufrj.br

Research supported by CNPq

and FAPERJ.

Received August 4, 2003

Accepted August 5, 2003

Key words• Science funding• Public and private

universities• Mental suffering• Publish or perish• Academic pressure• MSc and PhD students

Introduction

The number of Brazilian authors and thenumber of Brazilian publications in the in-ternational scientific literature have grownsubstantially during the last decade. Con-trasting with this growth, the resources allo-cated by the Brazilian government to scienceand technology in the 1996-2002 period de-creased steadily (Figure 1). This remarkablediscrepancy led us to study the effect ofshrinking funds on scientists’ everyday lives.A brief account of our findings has recentlybeen published (1). The conclusion was thatthe shortage of funds promotes competition,job stress, burnout and mental suffering. Inorder to survive professionally, scientists needto publish at increasing rates and in journalswith ever greater impact, generating a fret sowell characterized recently by Lawrence (2).

Here we describe briefly the distribution ofscience in Brazil, and analyze 25 open inter-views and 124 semi-structured interviewscarried out among graduate students and pro-fessors of a biochemistry department of theFederal University of Rio de Janeiro that isconsidered highly qualified by Brazilian fund-ing agencies.

Brazilian science

Unlike most of the developed countries,Brazil’s population of young people (18 to30 years old) has increased during the pasttwo decades (Figure 2). Student enrollmentin both private and government (state andfederal) universities has also grown, as shownfor the years 1995-2000 in Figure 3C. Educa-tion at government universities is free. Thenumber of professors employed by private

Brazilian Journal of Medical and Biological Research (2003) 36: 1135-1141ISSN 0100-879X Concepts and Comments

1136

Braz J Med Biol Res 36(9) 2003

L. de Meis et al.

students are found in government universi-ties (Figure 4 and Table 2). The graduatecourses are evaluated regularly and rankedfrom 1 (lowest) to 7 (highest) by CAPES(Coordenação e Aperfeiçoamento de Pessoalde Nível Superior), a Brazilian agency of theMinistry of Education (Figure 4). Evaluationincludes analysis of the curriculum vitae ofthe teaching staff, the number of years stu-dents take to defend their theses, and whetheror not the thesis work leads to a scientificpublication. Fellowships are granted to thecourses rated 3 and above, to be distributedamong their students.

From 1997 to 2000, Brazilian scientistspublished 34,274 papers in journals indexedby the Institute for Scientific Information(ISI; Figure 1A) and of these, 91.3% werepublished from government institutions thathave graduate courses. Science and trainingof new scientists are concentrated in a fewgovernment universities: 78.5% of the gradu-ate courses rated 6 and 7 are located in 12government universities, and the investiga-tors in these institutions published 71.2% ofthe articles from Brazil that appeared in jour-nals indexed by ISI between 1997 and 2000(Table 3). Among these 12 universities, threeof them, the University of São Paulo (USP),the Federal University of Rio de Janeiro(UFRJ) and the University of Campinas(UNICAMP), house 51.1% of the level 6-7graduate courses and published 42.9% of theBrazilian articles indexed by ISI between1997 and 2000 (Table 3). From this informa-tion, it can be gathered that Brazilian scienceis concentrated in a very few institutionsand, since these centers are mainly govern-mental, science is basically supported bygovernment funds.

The competition

The number of degrees granted everyyear has increased steadily during the pastdecade. The number of MSc and PhD stu-dents graduated in 1990 was 5,579 and 1,410,

Figure 2. Young population.Number of 18- to 30-year-oldsin Brazil, USA, UK, France andGermany. Source: United Na-tions Population Division,World population prospect,Population database.

Figure 1. Scientific publications, authors, and expenditures in science and technology. A,Brazilian authors appearing in publications indexed by the Institute for Scientific Informa-tion (ISI; open circles); number of scientific publications (articles, reviews and letters)originating in Brazil and published in journals indexed by ISI (filled circles). B, Total federalexpenditures on science and technology. Values in Brazilian currency (reais) were con-verted to US dollars using the average annual exchange rate. Source: “SIAFI/STM -Elaboração: Consultoria de orçamento/CD e PRODASEN”.

universities has increased, but this is not thecase in government universities (Figure 3B).In 2001 there were 1,208 private and 183government institutions (Figure 3A). Theenrollment of students in the various fieldsof knowledge in the two types of universitiesis different; examples are shown in Table 1.While a majority of undergraduate studentsare enrolled in private universities (Table 1),most of the courses that train MSc and PhD

1980 1990 2000Year

40

20

0

60

You

ng p

opul

atio

n (m

illio

ns)

Brazil USA XXXXX UK France Germany

Pub

licat

ions

and

a

utho

rs(h

undr

eds)

1980 1990

Year

2002

Year

A

2000 1996 1998

1500

1000

500

0

2000

Exp

endi

ture

s (m

illio

n U

S$)

B

2000

250

200

150

100

50

0

1137

Braz J Med Biol Res 36(9) 2003

Science funding and scientific productivity in Brazil

respectively, and this increased to 18,374and 5,344 in 2000. Most of the new profes-sionals entering the work market have beenabsorbed by private universities, where re-search is scant. Only a few research posi-tions are available in industry and in govern-ment universities each year. In 2001 theConselho Nacional de DesenvolvimentoCientífico e Tecnológico (CNPq), a Brazil-ian agency of the Ministry of Science andTechnology, created a new fellowship in-tended to allow 10 young investigators witha PhD degree and a good scientific back-ground to become affiliated with active re-search centers. The program was called“PROFIX”; 1,154 candidates competed forthe 10 positions available. In the same year,CNPq offered to the scientific community asa whole research grants varying fromUS$2,000 up to US$43,600 for a two-yearperiod. For the section that included bio-chemistry, biophysics, physiology, pharma-cology and neurosciences there were 437applications. Of this number, 267 were ap-proved on scientific merit but the resourcesavailable allowed the funding of only 20applications. Both for the PROFIX programand for the research grants the main selec-

Inst

itutio

ns o

ffer

ing

unde

rgra

duat

e co

urse

s 1000

800

600

400

200

0

Aca

dem

ic s

taff

(th

ousa

nds)

100

80

60

40

20

0

Stu

dent

s (x

105

)

20

15

10

5

01995 1998 2001

Year1995 1998 2001 1995 1998 2001

A B C

Figure 3. Number of institutionsoffering undergraduate courses(A), Academic staff (B) and Stu-dent enrollment (C) in govern-mental (open circles) and pri-vate Brazilian institutions (filledcircles). In 2001 there were939,225 students enrolled infederal and state universitiesand 2,091,521 students in pri-vate universities. Source: Insti-tuto Nacional de Estudos ePesquisas Educacionais AnísioTeixeira (INEP) - 2001 Census.

Table 1. Undergraduate enrollment and fields of knowledge taught in governmentand private universities.

Fields of knowledge Number of students Fraction of Fraction ofenrolled in both students enrolled students enrolled

governmental and in governmental in privateprivate universities universities (%) universities (%)

Law 414,519 13.9 86.9Business & Administration 356,156 12.8 87.2Medicine 57,930 53.4 46.6Mathematics 19,649 55.8 44.2Chemistry 18,224 61.6 38.4Earth sciences 22,302 81.8 18.2Physics 8,639 95.3 4.7All fields 3,030,754 31.0 69.0

Source: Instituto Nacional de Estudos e Pesquisas Educacionais Anísio Teixeira(INEP) - Census, 2001.

Num

ber

of g

radu

ate

cour

ses

600

500

400

300

200

100

01 2 3 4 5 6 70

CAPES evaluation score

Figure 4. Evaluation of graduatecourses in Brazil in 2001 in gov-ernmental (open circles) and pri-vate universities (filled circles).A total of 1,544 graduatecourses were rated from level 1to 7 by CAPES (Coordenação eAperfeiçoamento de Pessoal deNível Superior), including 1,343from governmental and 201from private universities.Source: CAPES.

1138

Braz J Med Biol Res 36(9) 2003

L. de Meis et al.

tion criterion was the scientific backgroundof the candidates, as evaluated by the num-ber of scientific publications and the impactof the journals where the papers were pub-lished.

Perception of the problem by theacademic staff

It was not surprising therefore that thenarratives of graduate students, postdoctoralfellows and teaching staff when interviewedrevolved around publications and journalimpact, increasing competition, disappoint-ment and uncertainty about their careers. Acommon feature of the respondents in theinterviews was a high degree of involvementwith their work. Typical statements of gradu-ate students were:

1. “You can fly, you can imagine some-thing out of the blue, something that Natureitself thought about... and you can discoverit!”;

2. “A fellow that earns a meager salarylike that one, well, he must really like whathe does... otherwise, he doesn’t live”;

3. “Why do you think there are so manycouples in science... they literally live lockedin the laboratory. They forget that there is awhole world outside, and the only partnerthey can meet is in the laboratory”.

Concerning this last statement we foundout that 26 of the 39 established investiga-tors of the department were married amongthemselves (13 married couples - 66.7%), 2were currently unmarried but had been mar-ried to fellow scientists (5.1%), 3 were single(7.7%), and only 8 were married to partnerswho had jobs unrelated to academia (20.5%).

The publication of articles and the im-pact of the journals where papers are pub-lished were the main concern of both gradu-ate students and investigators. Typical narra-tives were:

4. “He is a very helpful fellow... he hasgreat ideas and I always learn a lot fromhim, but this is not enough... In this place

Table 2. Graduate courses.

Fields of knowledge Government Privateuniversities universities

Life sciences 569 53Exact sciences 285 23Earth sciences 168 5Environment 15 0Social sciences 306 120Total 1,343 201

Source: Instituto Nacional de Estudos e PesquisasEducacionais Anísio Teixeira (INEP) - Census,2001.

Table 3. . . . . Publications (1997 to 2000) and outstand-ing graduate courses.

Universities* Publications** Courses level6 and 7***

USP 7,999 36

UFRJ 3,301 19

UNICAMP 3,413 14

UFMG 1,915 9

UFRGS 1,693 7

UNB 723 6

UFSC 901 4

UFPE 734 3

UNIFESP 1,150 3

UFSCAR 1,021 3

UFPR 784 1

UFF 755 1

Subtotal 24,389 (71.2%) 106 (78.5%)

Total for Brazil 34,274 (100%) 135 (100%)

*USP = University of São Paulo, UFRJ = FederalUniversity of Rio de Janeiro, UNICAMP = Univer-sity of Campinas, UFMG = Federal University ofMinas Gerais, UFRGS = Federal University of RioGrande do Sul, UNB = University of Brasília, UFSC= Federal University of Santa Catarina, UFPE =Federal University of Pernambuco, UNIFESP =Federal University of São Paulo, UFSCAR = Fed-eral University of São Carlos, UFPR = FederalUniversity of Paraná, and UFF = Federal Flumi-nense University. **Publications: includes ar-ticles, letters and reviews originating in Brazil andpublished in journals indexed by the Institute forScientific Information (ISI). ***Classified byCAPES (Coordenação e Aperfeiçoamento dePessoal de Nível Superior) on a scale of 1 (low) to7 (high).

1139

Braz J Med Biol Res 36(9) 2003

Science funding and scientific productivity in Brazil

you are only acknowledged when you havepublications... the rest doesn’t matter”;

5. “He [the thesis advisor] doesn’t careabout my thesis. He believes that a thesis isthe consequence of good work and goodwork means papers published in good jour-nals”;

6. “You feel a constant pressure to pub-lish...”;

7. “What we hear is that you are worthwhat you publish... the currency in this arenais publications”;

8. “My boss has a lot of publications...but when you go and look for what he dis-covered... you can define it in three sen-tences: he discovered this, this and that. Therest was to train students and get grants”.

When the respondents referred to theircolleagues’ work, they usually referred tothe number of publications and the journals,not knowing clearly what it was that hadbeen discovered. Scientific reputation there-fore seems to be built on the basis ofscientometrics, assuming that the impact ofthe journals reflects the novelty and impor-tance of the discoveries:

9. “They evaluate people by the numberof publications... and then they classify themby the impact of the journals: high-impactand low-impact scientists...”.

The submission of a paper evokes strongemotions, and frequently the respondent isconcerned with what his peers will think ofhim when they learn about the acceptance orrejection:

10. “My major concern was to publish, tobe recognized... it was a kind of self-affirma-tion, so I could tell them, ‘Look, I am good!’”;

11. “When the journal does not accept,you feel as if it is not only your paper, butyou yourself that is rejected... They look atyou as if you do not deserve to be there... it isa very bad feeling!”;

12. “It is a great feeling to have a paperaccepted, mainly when you know that somany people have their papers rejected... ”.

The difficulty in obtaining research sup-

port, in a country where funding is mainlygovernmental, generates anxiety and a strongfeeling of insecurity both in individuals thathave recently been awarded a PhD and inestablished investigators:

13. “When I got my PhD degree I though:‘Now what am I going to do? It isn’t possiblethat I have worked so hard not to be ab-sorbed by the system’”;

14. “You fulfill all requirements, you workvery hard... but you are not sure if at the endyou will get a job...”;

15. “If you stop publishing, you lose yourgrant... You are ejected from the system, itdoesn’t matter what you did in the past - itonly matters what you have done in the last2 to 3 years”;

16. “At times I feel so anxious... you mustcomplete your thesis in a short period oftime, you have an advisor who guides youbut at the same time continuously demandsresults, because we live in a system thatconstantly demands more and more from theadvisor, and so it goes on down the line, in acascade...”;

17. “I began to feel paranoid. I did notwant to work any longer, I wanted to quit.Everything was dark... when I arrived at thelab I did not feel well, I felt sick...”.

The result of grant evaluation is alwaysassociated with strong feelings, regardlessof whether or not it is approved:

18. “I hate all of this, including the peoplewho work here... I want to retire, to goaway... I feel a knot in my stomach everytime the car gets near the university... I hatethose people that got the money... they areno better than I, nor do they publish morethan I do, and I have to use my salary to keepthe laboratory going”;

19. “I knew they would publish the resultof the evaluation that night, so I went homeand after dinner, I went to the computerquietly. ...my heart was pounding inside me,I could feel it thump, thump, thump while Ilooked up the list and then... My name wasthere, I was awarded! ...and then I started

1140

Braz J Med Biol Res 36(9) 2003

L. de Meis et al.

crying, and I cried and cried... I went andhugged my wife, crying... and all of that justfor a lousy grant of less than 8,000 dollarsper year”.

In a highly competitive atmosphere, Bra-zilian science might be improving, but whatis the impact of this situation on the indi-vidual, particularly on students who haveworked very hard to get their PhDs and needto find their place in the scientific commu-nity?

Rites of passage

There are moments, defined as rites ofpassage, which demarcate the transition inthe subject’s life from one clearly definedsocial position to another, still unfamiliar tohim. According to Van Gennep (3), the ritesare divided into three stages: the “death” ofthe old social role, the “margin” or “transi-tion”, and the “rebirth” into a new role insociety. The margin is a time when the sub-ject lives in a situation between death andrebirth. In this stage social rules are sus-pended because the subject no longer be-longs to the “Old World”, nor has the personentered the “New World” where he or shewill exist from now on. At this moment thesubject is in nowhereland. According toTurner (4), “the attributes of liminality orlaminar personae are necessarily ambigu-ous, since the condition and these personselude or slip through the network of classifi-cations that normally locate states and posi-tions in cultural space. Laminar entities areneither here nor there; they are between thepositions assigned and arrayed by law, cus-tom and conventions”.

The transition period in the ritual of pas-sage and the liminality are conditions whereanxiety, doubt about one’s capacity to copewith the situation, stress and mental suffer-ing are highlighted.

In universities where there is no research,the thesis defense is a rite of passage, legiti-mization as part of the teaching staff coming

mainly from the academic title of PhD. Incontrast with this situation, in the biochem-istry department studied, the ritual of pas-sage is the publication of a paper in a journalindexed by ISI, and the thesis is simply “theconsequence of good work and good workmeans papers published in good journals”(see statements 5, 10). In the departmentstudied, where the graduate course is ranked7 by CAPES, legitimization never really ar-rives. The trajectory of the scientist becomesan increasingly difficult battle for grants,where the individual may lose support at anytime (see for instance statements 13 to 16).The idea of a continuous and stable career isblurred, and the individual must continuallyprove his or her capacity, being always in atransition state or liminality with the chancesof being eliminated from the system increas-ing every year (Figure 1B and statements 11,14 to 16).

Job burnout

The extended transition state may contri-bute to a condition of emotional and mentalexhaustion at work, defined as burnout, aphenomenon that has a significant impact onmodern culture (5,6). The interviews sug-gested that the situation characterized in Fig-ure 1B promotes this condition among thepopulation studied. Those most prone to de-velop burnout are the dedicated and commit-ted workers (see statements 1 to 3). Burnoutseems to be caused by frustrated hopes andexpectations, a feeling of inadequate controlover one’s work and a feeling of losing themeaning of life. These elements are presentin statements 11 and 13 to 19. The keysymptoms of burnout are emotional andmental exhaustion, feelings of cynicism withnegative and detached response to variousaspects of the job and a sense of ineffective-ness and lack of accomplishment. Thesesymptoms were described in statements 4,11, and 13 to 19. The likelihood of burnoutincreases when there is a significant and

1141

Braz J Med Biol Res 36(9) 2003

Science funding and scientific productivity in Brazil

prolonged occupational stress, includingwork overload and inadequate resources tomeet the stress. These situations are charac-terized by Figure 1 and statements 6, 15, 17and 18. The burnout syndrome leads to aprogressive loss of idealism and has beenassociated with impaired job performanceand poor health, including headaches, hy-pertension, anxiety and depression - see state-ments 11, 14 and 17 to 19. Different authorsreport that burnout may lead to alcohol anddrug abuse and deterioration of relationshipswith family and friends.

Conclusion

The number of Brazilians attending uni-versities is increasing steadily, thereby in-creasing the chance of identifying and nur-turing new scientific talents (Figures 2 and3). This reservoir of young talent is probablycontributing to the increase in the number ofBrazilian authors appearing in internationalscientific journals (Figure 1A). However,the discrepancy between the scarcity of fundsand the need for publications is leading tothe “publish or perish” condition experi-

enced in the USA during the seventies, withthe difference that in the USA this occurredat a time when science was well established,while in Brazil it is taking place at an earlystage of its scientific development. This pres-sure to publish is leading to an exaggerateddegree of competitiveness, propagating acultural distortion where scientometrics pre-vails over knowledge and where mental suf-fering is making up for the lack of funds, i.e.,is directly proportional to the lack of funds.One of the possible future scenarios is that asignificant proportion of Brazilian scientistswill leave academia and new talents will bediscouraged from choosing a scientific ca-reer due to its bleak perspective. The combi-nation of these two factors may lead to adecline of Brazilian science. Thus, if thispattern is not reversed in the near future,Brazilian science may go from growth todecay without ever realizing its full poten-tial.

Acknowledgments

We thank Dr. V. Rumjanek and Dr. M.Sorenson for useful discussions.

References

1. de Meis L, Carmo MS & de Meis C (2003). Impact factors: just partof a research treadmill. Nature, 424: 723.

2. Lawrence PA (2003). The politics of publication. Nature, 422: 259-261.

3. Van Gennep A (1960). The Rites of Passage. Routledge & KeganPaul, London, England.

4. Turner V (1969). The Ritual Process: Structure and Antistructure.

Aldine Publishing Company, Chicago, IL, USA.5. Maslach C, Schaufeli WB & Leiter MP (2001). Job burnout. Annual

Review of Psychology, 52: 397-442.6. Iacovides A, Fountoulakis KN, Kaprinis St & Kaprinis G (2002). The

relationship between job stress, burnout and clinical depression.Journal of Affective Disorders, 75: 209-221.