gms | German Medical Science

GMS German Medical Science — an Interdisciplinary Journal

Association of the Scientific Medical Societies in Germany (AWMF)

ISSN 1612-3174

Evaluation of medical research performance – position paper of the Association of the Scientific Medical Societies in Germany (AWMF)

Position Paper

  • corresponding author Christoph Herrmann-Lingen - Department of Psychosomatic Medicine and Psychotherapy, University of Göttingen Medical Center, Göttingen, Germany
  • Edgar Brunner - Institute for Medical Statistics, University of Göttingen Medical Center, Göttingen, Germany
  • Sibylle Hildenbrand - Institute of Occupational and Social Medicine and Health Services Research, University Hospital of Tübingen, Tübingen, Germany
  • Thomas H. Loew - Department of Psychosomatics, University Hospital of Regensburg, Regensburg, Germany
  • Tobias Raupach - Department of Cardiology and Pneumology, University of Göttingen Medical Center, Göttingen, Germany
  • Claudia Spies - Department for Anesthesiology and Surgical Intensive Care, Charité Berlin, Berlin, Germany
  • Rolf-Detlef Treede - CBTM Neurophysiology, Medical Faculty Mannheim of Heidelberg University, Mannheim, Germany
  • Christian-Friedrich Vahl - Department of Cardiothoracic and Vascular Surgery, University of Mainz Medical Center, Mainz, Germany
  • Hans-Jürgen Wenz - Clinic of Prosthodontics, Propaedeutics and Dental Materials, Christian-Albrechts-University Kiel, Kiel, Germany

GMS Ger Med Sci 2014;12:Doc11

doi: 10.3205/000196, urn:nbn:de:0183-0001969

This is the English version of the article.
The German version can be found at: http://www.egms.de/de/journals/gms/2014-12/000196.shtml

Received: June 23, 2014
Published: June 26, 2014

© 2014 Herrmann-Lingen et al.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by-nc-nd/3.0/deed.en). You are free: to Share – to copy, distribute and transmit the work, provided the original author and source are credited.


Abstract

Objective: The evaluation of medical research performance is a key prerequisite for the systematic advancement of medical faculties, research foci, academic departments, and individual scientists careers. However, it is often based on vaguely defined aims and questionable methods and can thereby lead to unwanted regulatory effects. The current paper aims at defining the position of German academic medicine toward the aims, methods, and consequences of its evaluation.

Methods: During the Berlin Forum of the Association of the Scientific Medical Societies in Germany (AWMF) held on 18 October 2013, international experts presented data on methods for evaluating medical research performance. Subsequent discussions among representatives of relevant scientific organizations and within three ad-hoc writing groups led to a first draft of this article. Further discussions within the AWMF Committee for Evaluation of Performance in Research and Teaching and the AWMF Executive Board resulted in the final consented version presented here.

Results: The AWMF recommends modifications to the current system of evaluating medical research performance. Evaluations should follow clearly defined and communicated aims and consist of both summative and formative components. Informed peer reviews are valuable but feasible in longer time intervals only. They can be complemented by objective indicators. However, the Journal Impact Factor is not an appropriate measure for evaluating individual publications or their authors. The scientific “impact” rather requires multidimensional evaluation. Indicators of potential relevance in this context may include, e.g., normalized citation rates of scientific publications, other forms of reception by the scientific community and the public, and activities in scientific organizations, research synthesis and science communication. In addition, differentiated recommendations are made for evaluating the acquisition of third-party funds and the promotion of junior scientists.

Conclusions: With the explicit recommendations presented in the current position paper, the AWMF suggests enhancements to the practice of evaluating medical research performance by faculties, ministries and research funding organizations.


1 Status quo

As early as in 1999 the AWMF (Association of the Scientific Medical Societies in Germany) has critically commented on the inappropriate use of methods such as the unadjusted journal impact factor for evaluating medical research performance [1]. These comments have been widely received in German academic medicine and partly been followed by many faculties of medicine [2]. Also from the perspective of individual medical (e.g., [3]) and non-medical disciplines (e.g., [4]), the dominance of journal impact factors in evaluating the performance of individual researchers has been criticized repeatedly. Nevertheless, methodologically questionable quality indicators still play an important role when it comes to evaluating research performance of individuals and institutions, as recently stated in an editorial in Science [5]. Based on the San Francisco Declaration on Research Assessment (DORA) there is a growing opposition against the allocation of public resources and the daily practice to decide about scientific careers mainly on the basis of cumulated journal impact factors. Peter Higgs, the current Nobel laureate in physics, recently criticized in an interview with the Guardian: “Today I wouldn’t get an academic job… I don’t think I would be regarded as productive enough” (http://www.theguardian.com/science/2013/dec/06/peter-higgs-interview-underlying-incompetence).


2 Recommendations

The AWMF makes the following recommendations concerning the evaluation of medical research performance by faculties, ministries of research and research funding organizations:

  • The evaluation of medical research performance should be based on a priori and explicitly phrased and communicated aims.
  • Informed peer review procedures are particularly useful for evaluating medical research performance. However, because of the high effort involved in such peer review, it appears feasible only at longer time intervals.
  • The most important parameter of evaluation is the importance of research for the advancement of scientific medicine or a particular medical discipline.
  • For this purpose, the journal impact factor is not an appropriate measure. Therefore it shall not be used for evaluating the research performance of individuals or institutions. It should rather be replaced, as soon as possible, with more appropriate indicators, such as adequately normalized citations rates.
  • Besides the reception by the scientific community, also the usefulness for the practice of medicine (e.g., guideline relevance, transfer into practice) or the society as a whole (e.g., disease prevention, economic impact) are considered appropriate indicators of scientific impact in medicine.
  • In view of increasing problems with attracting junior scientists and physicians, adequate measures to attract and support young academics make up a second highly important parameter of evaluation.
  • The structure and processes of undergraduate academic teaching, measures to support postgraduate junior scientists, and the respective results of these measures are considered appropriate indicators for successful promotion of young scientists. [A separate position paper of the AWMF and the MFT (German Association of Medical Faculties) will address issues of evaluation in curricular medical teaching and will be issued in the near future.]
  • Depending on the aims of the evaluation, attracted or disbursed third party funds can also be used as parameters for evaluation.
  • When evaluating attracted third party funds, public grants or comparable funds based on independently peer-reviewed grant proposals shall receive a higher score than funds from other sources, especially those without a competitive review process.
  • Besides simply adding the total amounts of third party funds, the scientific “yield” per sum of money spent should be considered. Suitable algorithms should be developed for defining this ratio.
  • Suitable indicators should also be developed for the evaluation of research performed in larger, typically interdisciplinary groups, such as research consortia and multi-author publications. These indicators should take both the individual contribution and the achievement of the group as a whole (added value by networking, coordination etc.) into account. This refers to the scientific impact as well as to jointly attracted third party funds.

3 Rationale for the recommendations

3.1 Overarching aspects of evaluation

Three overarching aspects of evaluation can be identified with regards to medical research performance:

  • Aims of the evaluation: optimization of research performance by means of the regulating effects of summative and formative evaluation on different levels (evaluation of individual researchers versus evaluation of institutes, clinical departments, centers or entire faculties)
  • Methods of evaluation: dimensions (input/output) [The term “output” as used here covers both the “impact” of research and promotion for junior scientists.] and instruments (e.g., informed peer review, metrics)
  • Consequences of the evaluation: material and immaterial appraisal and reward for good performance, adaptation of general research environments etc.

From the perspective of the AWMF, evidence base, transparence and acceptance of evaluations are pivotal prerequisites for their success. Evaluations should not be considered acts of top-down control but rather as interactive processes for quality assurance and development of science, and fair allocation of limited research resources and career opportunities. The methods of evaluation follow its defined aims. Conversely, the methods define, what consequences can plausibly be based on the evaluation. Aims and consequences therefore create a necessary framework for methods of evaluation. The main focus of the current paper will be placed on these methods.

3.2 Aims of evaluation

The evaluation of medical research performance should always occur under an a priori clearly defined aim. The aim of the evaluation determines the type, intensity and frequency of evaluations. Evaluation of the quality of research performance is the most critical component in this context. Depending on the precise aims, quality can be operationalized as progress in scientific knowledge or the benefit of research for patient care, undergraduate, postgraduate and continued medical education (including attraction and promotion of junior scientists) or for other societal aims (e.g., prevention, ethical issues, economic relevance). In contrast to the summative evaluations predominating so far (i.e. evaluations of research results) the relative weight of formative evaluations (i.e. evaluation used for optimizing scientific processes) should be increased. Such formative evaluations can serve for giving constructive feedback with the aim of advancing individual careers, scientific programs and institutions and supporting the implementation of good scientific practice.

3.3 Methods of evaluation

Informed peer review procedures such as those that have been endorsed and performed by the German Council of Science and Humanities (“Wissenschaftsrat”) are particularly suitable for evaluating medical research performance. Because of the high effort required for this method (including, e.g., burden on reviewers) informed peer reviews appear feasible solutions only for select purposes (e.g., evaluation of whole faculties, appointment procedures) and at larger time intervals.

Therefore, less costly evaluation methods must also be available. In this context, quantitative parameters can be applied under the precautions mentioned below. However, using an automatic link between certain numeric scores and subsequent (e.g. financial or career-related) consequences is strongly discouraged. Metric indicators should rather serve for informed discussions between the evaluating bodies and the evaluated researchers or institutions. Metric indicators should be interpreted in the context of the specific background of the particular medical discipline evaluated, its research culture, local conditions etc.

In addition to formative evaluations, summative evaluations can be applied for specific regulatory processes in research funding or in preparation for decisions on career development and promotion. These evaluations shall, however, be used with a good sense of proportion and never without a critical appraisal of their regulatory effects, in order to counteract unwanted regulatory effects (e.g., effects on quantity instead of quality [6]; inappropriate overestimation of mainstream research, which tends to produce higher summative scores and may therefore be favored over innovative research). Any unwanted effects should lead to immediate modification of the summative parameters used. Any excessive “evaluatis” is disapproved by the AWMF. This includes, e.g., evaluations that are (in total or in parts) performed without a clear aim or clear consequences, an excessively high frequency of evaluations without appropriate increase in information, or evaluations at time intervals that are too short for initiating meaningful regulatory processes in research planning.

The evaluation of medical research performance shall mainly focuse on three core areas:

  • The “impact” of research activities in a broader sense, i.e. their contribution to scientific, medical/clinical and other societal progress
  • The “input”, i.e. especially the performance in generating competitive third party funds
  • The “attraction and promotion of young scientists” as a crucial factor of sustainability

These areas conform with recent recommendations on future research rankings issued by the German Council of Science and Humanities [7]. From the perspective of the AWMF also the criteria additionally proposed by the Council of Science and Humanities, i.e. science transfer, knowledge transfer and reputation can be subsumed under these three dimensions.

3.3.1 Evaluation of “impact”

As a signer of the San Francisco Declaration on Research Assessment (DORA: http://am.ascb.org/dora/) the AWMF commits itself to the requirements of evaluation of publications as stated there. The following aspects are of particular relevance in the context of evaluation of research performance:

  • “Do not use journal-based metrics, such as Journal Impact Factors, as a surrogate measure of the quality of individual research articles, to assess an individual scientist’s contributions, or in hiring, promotion, or funding decisions.”
  • The scientific content of a publication is much more important than publication metrics or names of journals. This is of particular importance when evaluating junior scientists.
  • Funding agencies and institutions should not only consider research publications but also the value of other research outcomes (data sets, software, patents etc.) and keep in mind a broader, also qualitative spectrum of impact measures including influence on politics and medical practice.

When evaluating the impact of an individual’s research performance, it is the position of the AWMF that the core question is whether this individual has contributed to progress in his or her discipline.

This can be assessed or measured on different levels:

  • 1st Level: Evaluation of publications
    a) In recognized scientific journals with peer review
    b) In other media (books, guidelines etc.)
    c) Citation by guideline recommendations
  • 2nd Level: Active contributions to scientific organizations or boards and editorships
  • 3rd Level: Leadership in organizing scientific conferences

It is considered difficult to combine these three levels into a single scale, since no useful conversion factors are available. The levels 1b to 3 should rather be considered relevant indicators in their own right and should be used to supplement the level 1a indicators predominating so far, leading to a multidimensional appraisal with separate subdimensions.

Level 1a

In addition to the requirements stated in the DORA, the following points are suggested:

  • As far as bibliometric indicators shall be used for summative evaluations, it must previously be made sure that these indicators are evidence-based, transparent and feasible. For testing the evidence base of indicators, the aim of the evaluation must be kept in mind (e.g. the desired regulatory effects of the evaluation). The simplicity of performance cannot be the main criterion and must not lead to the application of inappropriate instruments. However, also an evaluation that appears useful in terms of contents must remain feasible under the given conditions. For methodologically adequate evaluation, bibliometric expertise must be “purchased”, if necessary. Such services are commercially available.
  • Neither the journal impact factor nor the H-index are suitable measures for evaluation of individuals’ research performance. The journal impact factor is a measure for citations to a journal over a relatively short time frame. It is not sufficiently correlated with citation rates of individuals articles, does not take into account the variance in publication cultures among disciplines and must therefore not be used for evaluating individuals and institutions. The H-index has been seen skeptically due to its numerical instability and dependency on age.
  • The use of more differentiated bibliometric analyses is therefore preferred [5], especially the use of field and article type-normalized citation rates. [I.e. the standardization of citation rates of individual articles is based on their respective disciplines (as a reflected in the subject area of the journal in which an article is published) and on the article type: original papers, review articles and letters to the editor are weighted separately due to very different patterns of citation. It still needs to be determined how field normalization can be performed for interdisciplinary publications or publications from cross-sectional research areas as well as from those journals that are listed in the web of science in a category that differs from the German specification of medical disciplines.]
  • Field normalization reflects and adjusts for the differences in publication and citation cultures across disciplines [8], [9], [10]. Article type normalization takes into account the different average citation rates of e.g., original publications and reviews [11]. [Of course, also systematic reviews, meta-analyses and guidelines must be considered as genuine scientific publications. See also AWMF Statement dated 9 November 2013. http://www.awmf.org/fileadmin/user_upload/Stellungnahmen/Forschung_und_Lehre/AWMF-Resolution_Wiss-Anerkennung-LL-Arbeit.pdf]. The evaluation window for citations should cover several (e.g. five) years, as suggested by bibliometric research findings [11].
  • When evaluating individual research performance, the “dos and don’ts” of individual-related bibliometrics according to Glänzel & Wouters [12] should be followed.
  • The individual contributions of each author of a publication should be named in a standardized manner in all journals. Courtesy or honorary authorships are unacceptable.
Level 1b

The following contributions should be counted as discrete publication forms with individually determined weights:

  • Monographs and book chapters
  • Guidelines and health technology assessment (HTA) reports, even when they have neither been published in a book nor in a scientific journal
  • Publications of original data, software developments, patents etc. (see DORA)
  • Provision of scientific findings to lay persons for non-scientific practice (e.g., patient guidebooks, press releases etc.)
Level 1c

If a publication is cited by guidelines when immediately justifying specific recommendations (“guideline relevance”) this should be considered an appropriate measure of clinical impact of this publication and should be considered separately.

Level 2

Under this heading, activities for intra- and interdisciplinary networking and quality assurance of research should be evaluated. Relevant aspects of particular importance are

  • Editorship in scientific journals as a core instrument for disseminating research results
  • Active positions on boards, sections and working groups of scientific societies and organizations
  • Active involvement in scientific councils of recognized national or international research-funding and science organizations.
  • Outstanding scientific reviewer positions (e.g., collegiates of the German Research Fund [DFG])
Level 3

Organization and leadership in scientific meetings is an important medium of research communication and shall be considered as discrete achievement.

3.3.2 Evaluation of “input”

Input-related parameters of research performance can be defined on various levels. Those factors should be preferred that can be directly influenced by the evaluated person or institution:

  • General research framework (basic funding, expertise, strategic concepts, proportion of protected time for research, quality of promotion for junior scientists); this can be influenced on the level of faculties or centers and will not be in the main focus of this statement.
  • Attraction and effective use of third party funds; this can be influenced by individual researchers.

Parameters for the evaluation of research performance within faculties:

  • Explicit and transparent rules should be established what third party funds will be accepted for evaluation and how they will be weighted.
  • When weighting third party funds for evaluation, funds granted after an independent review process must be given higher weights than those provided without independent review.
  • Also the source of funding should be weighted: public funding and neutral foundations should be given a higher weight than funds from special interest groups or industrial sponsors.
  • Contract research is sufficiently honoured by the funds provided and is appropriate for financing preliminary research for preparing competitive grand applications. It does not justify an additional bonus from public sources. However, in the area of applied science, it can be utilized for evaluating individual researches.
  • A fair and transparent evaluation of individual grant money from collaborative research projects and industry-independent multicenter studies should adequately reflect both the achievement of the principal investigator as well as those of the collaboration partners. This is a necessary prerequisite for a culture of scientific collaboration. Eventually general weighting algorithms must be defined for the participants in various types of research consortia and studies (e.g., one third of evaluation weights for the principal applicant, distribution of the remaining two thirds among all collaboration partners or study centers).
  • Funds should be weighted according to the number of positions for scientists. The acquisition of expensive equipment should not be counted as scientific quality criterion. In clinical trials, weights based on the amount of case payments can be considered. Accordingly, flat rates can be used for evaluating third party-funded scientific services.
  • The cost effectiveness in terms of scientific output (as described under the chapters “impact” and “attraction and promotion of junior scientists”) per funds granted should be considered as a measure for appropriate spending of resources when evaluating performance of researchers and institutions. For this purpose, suitable algorithms have to be developed.

When evaluating medical research, a bonus for systemic, translational and human subject or patient-related research with concrete reference to practical medicine should be introduced.

3.3.3 Evaluation of attraction and promotion of junior scientists

The attraction and promotion of junior scientists from the beginners’ stage up to the independent researcher is a core issue of evaluation.

The guiding principle should be to get students into science at an early stage and to support them in a sustainable way until they acquire sufficient competences for promotion to full professor. Target groups of promotions are undergraduate students, doctoral students and post-doctoral fellows as well as associate professors of medicine and dentistry, physicians, dentists and scientists from related areas. They should be taught a multidimensional model of research that includes scientific work itself, its application in practical medicine, lifelong learning, and teaching. This is in accordance with the role of a “scholar” as defined in the CanMEDs model of teaching [13], in which the role of a scholar is defined as a core role of graduates from medical school curricula in accordance with the international outcome framework. In so far, also the support for scientific competence for students going through the basic medical curriculum should be considered an instrument of promotion of junior scientists. Its evaluation, however, will not be addressed here in detail and only under the aspect of relevance for research, since the evaluation of achievements in teaching will be the subject of a separate position paper which is currently being developed between the AWMF and the German Association of Medical Faculties (“Medizinischer Fakultätentag”).

3.3.3.1 Quantitative indicators

Aspects of junior scientist promotion measurable at the level of institutions (faculty, institute, clinical department) include:

  • Broadness and depth of measures for junior scientist promotion e.g.,
    • Structured programs for acquainting students with research: exchange programs, curricular (teaching of scholar competencies, compulsory and elective courses) and hypothesis-based offerings (e.g., journal clubs, term or master theses, “how-to” courses), graduate schools/MD/PhD programs [14], clinical/physician scientist programs [15], mentoring programs in interdisciplinary networks
    • Number of students mentored during medical school (as a modular bridge before the start of scientific specialization [16])
    • Amount of protected time guaranteed [17], i.e. dedicated time for research without competing obligations in teaching and patient care for junior scientists in all steps of their careers until promotion to full professorship. Indicators: Time of exemption from clinical duties in percentages of a full employment (per duration of employment), amount of institution-wide conference times for research (per week or month) with percent participation rates of junior scientists.
  • Results of measures for junior scientist promotion e.g.
    • Number of graduates from the programs named above
    • Sustainability of programs; indicators may be e.g., career development, publication, attraction of grant money by junior scientists
    • Number of tenure track professorships in research and teaching [18]
    • Number of appointments of junior researchers to scientific leadership positions or to clinical leadership positions with a minimum of requirement in the three dimensions of scholar competences (application of scientific results, lifelong learning, teaching)

Individually measurable quantitative criteria of junior scientist promotion include

  • Number of adequately supported qualification theses (indicator e.g., number of doctoral students per completed doctorate)
  • Number of junior staff supported by structured research and funding programs or with leading role in development of evidence based guidelines
  • Career development and research success of junior staff from own research group (criteria see above)
3.3.3.2 Qualitative indicators

Indicators that can be measured on the level of the institution (faculty, institute, clinical department) include:

  • Availability and integration (both horizontal and vertical) of appropriate measures to promote junior scientists in the different stages of their careers, e.g. structured doctoral programs and sustainable support programs, internal and external peer review and coaching procedures (learning from the best, from common sense to scientific excellence, soft skills development). These programs shall foster interactions between junior scientists and established experts as suggested by the recommendations of the German Council of Science and Humanities [19] and the German Medical Association [20]. Further indicators may be the existence of research tracks and representatives of scientific societies within the institution.
  • Availability of quality standards (e.g., good scientific practice) and scientific infrastructure including, among others, programs for startup financing (for e.g., one year) in order to pay young scientists on the way to their first DFG grant, infrastructure directly related to grant applications (among others, courses about good clinical and scientific practice or good laboratory practice, grant counselling, support with related paperwork when dealing with e.g., animal protection authorities, availability of electronic laboratory diaries, use of core facilities or core research units) and clinical trials infrastructure (to support formal requirements from health authorities etc., among others: data protection, ethics committees, Federal Institute for Pharmaceuticals and Medical Products (BfArM), registration with state authorities and trial registries such as clinicaltrials.gov, writing of safety reports, procedures for prepublication of methods, evidence based monitoring of documents).
  • Availability of measures to increase transparency and equal opportunities. E.g., transparency of the scientific profile as an instrument for targeted choice of the university by junior scientists.
  • Flat hierarchies with e.g., trainee representation/speakers for assistant researchers, tandem professorships (e.g. following the Swiss example: assistant professorship with 50% time for research and teaching plus 50% for patient care, followed by a tenure track with 70% for research and teaching plus 30% for patient care or reverse); equal payment for scientific and clinical work over the whole course of the career; sustainability by better compatibility of career and family (childcare, eventually also at night, during weekends and with priority for scientists, childcare during meetings, seminars and conferences, emergency childcare).

Individually measurable qualitative criteria include:

  • Active participation in the promotion of young scientists by collaborating in e.g., DFG young scientist academies, summer schools for outstanding doctoral students, personal dedication as a model function, teaching of basic scientific competences to group members, e.g., teaching the difference between practice (everyday knowledge) profession (professional knowledge) and science (scientific knowledge) as well as bridging the gaps between analytics, transformation and theory; early integration of junior scientists in working groups with increasing individual responsibilities.
  • Quality of research performance of group members, e.g. reproducibility of results of members from the own group by other research groups, self-assessment: evaluation of both the most important results of own research and of the independence of own research by junior scientists as a measure of junior scientists’ promotion by their respective mentors, consequent tracking of junior scientists’ research, which should be reflected in a line of research and make clear the relevance of this research.

3.4 Consequences of evaluation

Possible consequences of the evaluation have to be clearly defined a priori. They should follow the aims of the evaluation and keep in mind its methodological limitations. Beside immediate feedback and joint discussion of evaluation results they can consist of a targeted application of instruments for organizational (e.g., creation of new research foci), project (e.g., funding decision) and career planning (e.g. appointment, tenure, mentoring), as long as they are based on a well-balanced and transparent procedure.

The performance-related allocation of funds is only one of numerous possible consequences of evaluation and its regulatory effects are seen controversially [21]. Appreciation of good work is of particular importance for the vast majority of scientists who are highly and intrinsically motivated. This should equally cover achievements in research, teaching and (in clinical medicine) also in patient care. The recognition can also be expressed by provision of time resources. In contrast, over the long term a predominance of financial incentives as extrinsic motivators runs the risk to undermine intrinsic motivation. This is particularly true, when the underlying evaluation processes are perceived as intransparent or unjust.

Therefore a sufficient basic funding of institutions is of high importance and should be subject to comprehensive evaluations only in longer time intervals in order to be adapted to new developments.


Notes

Acknowledgement

The present position paper is based on the results of the Berlin Forum of the AWMF held on 18 October 2013 under the headline “Methods for evaluation of medical research performance”. At this forum, invited experts, representatives from major science organizations (German Research Fund, Association of Medical Faculties, German Aerospace Center Project Management Agency, German Council of Science and Humanities) and participants of ad-hoc writing committees presented proposals which were subsequently subject to discussions within the AWMF Committee for Evaluation of Performance in Research and Teaching, condensed into a draft version and consented by the executive board of the AWMF. The authors thank all participants for their contributions to the present paper. In particular the following person should be mentioned here (in alphabetical order): Prof. K.-M. Debatin, Prof. R. Deinzer, Prof. W. Glänzel, Prof. C. Graf, Prof. H.-J. Heinze, Prof. S. Hornbostel, Dr. M. Kordel-Bödigheimer, Dr. T. Kostuj, Prof. H. K. Kroemer, Dr. A. Lücke, Prof. P. Meier-Abt, Dr. S. Moritz, Prof. A. F. J. van Raan, Prof. K. Rahn, Prof. G. Theilmeier, Dr. W. Warmuth.

Conflicts of interest

The authors declare the following conflicts of interest:

All authors are affiliated with German universities, which includes leadership duties and participation in academic self-government. All authors are also members of scientific societies, some (THL, RDT, CFV) of them in leading positions. CHL, RDT and CS are AWMF board members. All authors are authors of scientific publications. In preparing this manuscript, CHL, SH, RDT and HJW have received travel support from the AWMF and/or scientific societies for attending meetings. CHL, RDT and CS have received research grants from public sources. CHL, SH and RDT have received research grants from private entities. CHL and RDT have received honoraria from private entities. CS has been vice dean for teaching until May 2014, RDT is vice dean for research. CHL, EB, THL, RDT, CFV and CS are editors, associate editors or editorial board members for scientific journals. CHL and EB are book editors. CHL, EB, RDT, CFV and CS are reviewers for research funding organisations or endowments. THL is endowment council member.

Authorship

The co-authors are listed in alphabetical order.


References

1.
Frömter E, Brähler E, Langenbeck U, Meenen NM, Usadel KH. Das AWMF-Modell zur Evaluierung publizierter Forschungsbeiträge in der Medizin [The AWMF model for the evaluation of published research papers in medicine. Arbeitsgemeinschaft der wissenschaftlichen medizinischen Fachgesellschaften (Working Group of the Scientific Medical Specialty Societies)]. Dtsch Med Wochenschr. 1999 Jul;124(30):910-5.
2.
Brähler E, Strauss B. Leistungsorientierte Mittelvergabe an Medizinischen Fakultäten : Eine aktuelle Ubersicht [Performance-oriented allocations of financial resources at medical schools: an overview]. Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2009 Sep;52(9):910-6. DOI: 10.1007/s00103-009-0918-1 External link
3.
Vahl CF. Forschungsbewertung: Fairness für forschende Chirurgen: ein Plädoyer. Dtsch Arztebl. 2008;105(12):A-625-8.
4.
Adler R, Ewing J, Taylor P. Citation Statistics – A Report from the International Mathematical Union (IMU) in Cooperation with the International Council of Industrial and Applied Mathematics (ICIAM) and the Institute of Mathematical Statistics (IMS). Stat Sci. 2009;24(1):1-14. DOI: 10.1214/09-STS285 External link
5.
Alberts B. Impact factor distortions. Science. 2013 May;340(6134):787. DOI: 10.1126/science.1240319 External link
6.
Glasziou P, Altman DG, Bossuyt P, Boutron I, Clarke M, Julious S, Michie S, Moher D, Wager E. Reducing waste from incomplete or unusable reports of biomedical research. Lancet. 2014 Jan;383(9913):267-76. DOI: 10.1016/S0140-6736(13)62228-X External link
7.
Wissenschaftsrat. Empfehlungen zur Zukunft des Forschungsratings. Drs. 3409-13, Mainz 25.10.2013. Available from: http://www.wissenschaftsrat.de/download/archiv/3409-13.pdf External link
8.
Van Raan AFJ. The use of bibliometric analysis in research performance assessment and monitoring of interdisciplinary scientific developments. Technikfolgenabschätzung – Theorie und Praxis. 2003;12(1):20-9.
9.
Van Raan AFJ. Statistical Properties of Bibliometric Indicators: Research Group Indicator Distributions and Correlations. J Am Soc Inf Sci Technol. 2006;57(3):408-430. DOI: 10.1002/asi.20284 External link
10.
Schulze N, Michels C, Frietsch R, Schmoch U, Conchi S. 2. Indikatorbericht Bibliometrische Indikatoren für den PFI Monitoring Bericht 2013. Hintergrundbericht für das Bundesministerium für Bildung und Forschung (BMBF). Berlin, Karlsruhe, Bielefeld: Institut für Forschungsinformation und Qualitätssicherung (iFQ), Fraunhofer-Institut für System- und Innovationsforschung ISI und Universität Bielefeld, Institut für Wissenschafts- und Technikforschung (IWT); 2012 [cited 23.6.2014]. Available from: http://www.bmbf.de/pubRD/Indikatorbericht_PFI_2013.pdf External link
11.
Van Raan AFJ. Comparison of the Hirsch-index with standard bibliometric indicators and with peer judgment for 147 chemistry research groups. Scientometrics. 2006;67(3):491-502. DOI: 10.1556/Scient.67.2006.3.10 External link
12.
Glänzel W, Wouters P. The dos and don’ts in individual level bibliometrics. Paper presented at the 14th ISSI Conference, Vienna, 15-18 July 2013. Available from: http://de.slideshare.net/paulwouters1/issi2013-wg-pw External link
13.
Frank JR, Danoff D. The CanMEDS initiative: implementing an outcomes-based framework of physician competencies. Med Teach. 2007 Sep;29(7):642-7. DOI: 10.1080/01421590701746983 External link
14.
Basu Ray I, Henry TL, Davis W, Alam J, Amedee RG, Pinsky WW. Consolidated academic and research exposition: a pilot study of an innovative education method to increase residents' research involvement. Ochsner J. 2012;12(4):367-72.
15.
Gaehtgens C. Clinical Scientist – Neue Karrierewege in der Hochschulmedizin. Werkstattgespräch am 27.-28.9.2013 in Schloss Herrenhausen, Hannover. Ergebnisse und Schlussfolgerungen. [cited 20.5.2014]. Available from: http://www.volkswagenstiftung.de/fileadmin/downloads/publikationen/veranstaltungsberichte/Veranstaltungsbericht_Clinical_Scientist.pdf External link
16.
Bierer SB, Chen HC. How to measure success: the impact of scholarly concentrations on students--a literature review. Acad Med. 2010 Mar;85(3):438-52. DOI: 10.1097/ACM.0b013e3181cccbd4 External link
17.
Sullivan R, Badwe RA, Rath GK, Pramesh CS, Shanta V, Digumarti R, D'Cruz A, Sharma SC, Viswanath L, Shet A, Vijayakumar M, Lewison G, Chandy M, Kulkarni P, Bardia MR, Kumar S, Sarin R, Sebastian P, Dhillon PK, Rajaraman P, Trimble EL, Aggarwal A, Vijaykumar DK, Purushotham AD. Cancer research in India: national priorities, global results. Lancet Oncol. 2014 May;15(6):e213-22. DOI: 10.1016/S1470-2045(14)70109-3 External link
18.
Kelley WN, Stross JK. Faculty tracks and academic success. Ann Intern Med. 1992 Apr;116(8):654-9. DOI: 10.7326/0003-4819-116-8-654 External link
19.
Wissenschaftsrat. Empfehlungen zur Bewertung und Steuerung von Forschungsleistung. Drs. 1656-11, Halle 11.11.2011. Available from: http://www.wissenschaftsrat.de/download/archiv/1656-11.pdf External link
20.
Bundesärztekammer. Curriculum Ärztliches Peer Review. 2. Auflage. Berlin: Bundesärztekammer; 2013. (Texte und Materialien der Bundesärztekammer zur Fortbildung und Weiterbildung; 30). Available from: http://www.bundesaerztekammer.de/downloads/CurrAerztlPeerReview2013.pdf External link
21.
Krempkow R, Schulz P. Welche Effekte hat die leistungsorientierte Mittelvergabe? Das Beispiel der medizinischen Fakultäten Deutschlands. Hochschule. 2012;(2):122-42.