Intended for healthcare professionals

Feature Christmas 2009—Professional Matters

H-index pathology: implications for medical researchers and practitioners

BMJ 2009; 339 doi: https://doi.org/10.1136/bmj.b5356 (Published 15 December 2009) Cite this as: BMJ 2009;339:b5356
  1. Rob Horne, professor of behavioural medicine1,
  2. Keith J Petrie, professor of health psychology2,
  3. Simon Wessely, vice dean, institute of psychiatry2
  1. 1Centre for Behavioural Medicine, The School of Pharmacy, London WC1H 9JP
  2. 2Department of Psychological Medicine, University of Auckland, New Zealand
  1. rob.horne{at}pharmacy.ac.uk

    The h-index has quickly become the standard method by which medical schools judge the impact of medical researchers. Rob Horne, Keith Petrie, and Simon Wessely describe a cluster of potentially pathological behaviours associated with the index.

    In 2005 Jorge Hirsch proposed the h-index as a means of measuring the productivity and impact of a researcher1. A researcher’s h-index is determined by the highest number of papers they have published to receive at least that many citations (figure). So a scientist with an h-index of 40 has written 40 papers that have received at least 40 citations. The h-index can be obtained through the subscription databases of Web of Science and Scopus, or through using Publish or Perish software, which is based on the Google Scholar database, enabling brave (or reckless) authors to check their own h-index.

    Figure1

    Illustration of h-index calculation

    Although the h-index is not without its drawbacks, it has quickly become the standard measure by which medical schools judge the value of academic staff.2 The process of observing or assessing performance can influence behaviour and the h-index is no exception. The increasing importance of citation rate as an index of success has led to an increase in self citation (where the author’s earlier work is cited in their new publication).3 We have also noted the emergence of a range of socially undesirable behaviours associated with the h-index. We outline the behaviours and discuss their implications for medical researchers and practitioners.

    H-index behaviours

    Home-ophobia—Irrational hatred of people with similar names who may dilute or diminish your h-index. The name of this syndrome derives from the fact that having a less h-endowed namesake also reduces the likelihood of a Google search revealing the subject’s personal home page at the top of the search result. The fear of h-index dilution, coupled with home page obscurity within the Google search, creates the state of home-ophobia. Particularly prevalent among academics named Smith, Jones, Cohen, and Patel.

    H-bomb—Where disclosure or discovery of an individual’s h-index has an immediate, catastrophic effect on career aspirations, professional standing, and sense of self. Often manifested in the short term by an explosive venting of emotion, sometimes accompanied by fainting (H-ysteria) followed by chronic psychosomatic illness (post traumHatic stress disorder), and occasionally by psycHosis (see below).

    PsycHosis—A delusional state in which the sufferer perceives their h-index to be much higher than it really is and behaves accordingly (for example, with understated academic swagger). Sometimes linked to a failure to appreciate the influence of having a common surname, it is, in this respect, the reverse of home-ophobia. Unlike home-ophobia, psycHosis can also affect people with uncommon surnames, who can succumb to the delusional belief that they have authored a Nature paper.4

    One h-manship—Surrounding oneself with individuals with a lower h-index in order to boost self esteem. This may involve attending meetings which would ordinarily be avoided, such as seminars in cultural studies and general practice conferences.

    h-indexism—Appointing people to academic positions based on their h-index rather than the traditional factors of appearance, high school attended, or whether they are Chelsea Football Club season ticket holders.

    h-Cite—Self citation of a paper based solely on the fact that more citations of this particular paper will raise the author’s h-index.5 6 7 This should not be confused with general self citation3 where any one of an author’s papers is shamelessly referred to in the author’s own article.8 9

    HAART (highly articulate angry response to teaching)—Reaction exhibited when, on the basis of a low h-index, the academic is “invited” by the head of department to make the provision of undergraduate teaching his or her “core mission.” Not to be confused with highly active antiretoviral therapy.10

    Dropping your h’s—Letting one’s h-index slip in social company in order to boost social standing. Sometimes causing arguments with h-eretics who question the validity of the index. For those with a low h-index, this can take the form of a h-istory (story fabricated to explain a low h-score).

    Comment

    We believe that the cluster of behaviours described here has implications for medical researchers and practitioners. An awareness of the existence of these behaviours in others may help medical researchers to avoid any h-index linked professional embarrassment. However, retaining a dignified aloofness to the h-index is difficult for those with scores of less than 30. For this reason, researchers may also wish to increase their h-index as quickly as possible by publishing innovative work or through cunning self citation11.

    There are also implications for medical practitioners. Beliefs12 and behaviour13 can influence health and we anticipate that general and specialist physicians will see a large increase in the incidence of h-index related presentations, perhaps taking on new psychosomatic forms like those discussed above.14 Some cases may be managed by education15 and cognitive behaviour therapy16 but many will require medication.17 Prescribing creates a further challenge because many h-index patients will not perceive themselves to be ill, and consequently doubt that they need medication, leading to nonadherence.18 Moreover, being academics, they are likely to be dissatisfied with standard information19 demanding more detailed explanations and discussion about the condition and treatment.

    There are also important considerations for medicine as an academic discipline. The h-index is typically calculated for an individual. However, it can also be applied to groups of researchers.20 21 It could, in future, be applied to compare the research contribution of medical specialties in a medical school or of medical schools in universities with implications for the wellbeing of deans and editors of medical journals.

    Notes

    Cite this as: BMJ 2009;339:b5356

    Footnotes

    • We thank Amy Whitehead (research administrator, Centre for Behavioural Medicine, The School of Pharmacy, University of London) for help in preparing this manuscript and all the authors who have cited their papers. We also thank the reviewer for helpful comments and the suggestion that we should cite one of his papers in the article.

    • Competing interests: None declared. The h-indexes of the authors are 26 (KJP), 52 (SW), and 24 (RH).

    References