Career paths in medial research 887

of their male counterparts. Although women who apply have equal success rates for gaining NHMRC and other external research grants, a study of academics at the University of Melbourne in 1994 found that 78% of male academics applied for external grants compared with only 40% of female academics.’

Solutions

Every woman faces different crises in her quest for a career in medical research. Societal and family respon- sibilites may prevent one from accepting the rigours of a life in medical research. However, the ‘working mother’ is now the norm and child care facilities attached to, or close to, the hospitals, universities and research institu- tions, are more commonly available. Partners must also be encouraged to share parenting responsibilities. In male-dominated universities, procedural changes are still required such that women are represented on all important committees, in particular, those responsible for distributing university research funds, and that women of Level C appointments (Senior Lecturer) are considered senior enough for those committees. On NHMRC committees, for example, regional grants interviewing committees, women should be represented. Criteria for appointment and promotion (universities and Fellowship schemes) must be sensitive to periods of down-time due to relocation or family responsibilities. In 1995, collaborations are a key component of success in medical research. These could be set during a PhD candidature and for women could ensure productivity during, for example, maternity leave.

Many women in medical research find travelling and conference attendance ‘too difficult’ because of family commitments. Women should use information technol- ogy to their benefit with increased use of the Internet, e-mail and facilities such as videoconferencing. This same technology should be used to allow women in

research to form networks and establish channels of communication to keep each other informed of advances in their area of research or, more generally, of research politics or the identity of new sources of research money. The establishment of Women’s Groups can help more junior researchers identify with more senior ones and in many cases receive informal and constructive advice on applications for funding or promotion. The knowledge that many female researchers are having similar difficulties with career advancement can boost confidence levels. Manuscript rejection and grant failure has happened to us all, but it is very distressing when it first happens. Hopefully, the identification of and accessibility to role models and mentors will help to expand the ‘horizons’ of female medical researchers. Successful women in research should help the younger ones. Bonds can be very loose but the introduction at a conference to a leading researcher in the field or to a journal editor may reap unexpected returns. Women in medical research do not wish to achieve success through affirmative action but with increased confidence and extended horizons, women should soon be taking more positions in the higher levels of academia.

A study by the Australian Society for Medical Research showed that more than 70% of medical researchers were satisfied with their career. The breakdown of answers by men and women is not known. Research is rewarding but it is very demanding and fraught with insecurity. My hope is that the barriers are not too great for an increased number of women, with or without families, to stay in this career and to reach the highest levels if that is to what they aspire.

REFERENCES

1 Waller P, Grieve N. Women and Research. University of Melbourne, 1994.

Journal of Gastroenterology and Hepatology (1996) 11, 887-891

Myths and realities in clinical research

M WARD

Department of Gastroenterology, Royal Brisbane Hospital, Herston, Queensland, Australia

Abstract It is often argued that clinicians should be more actively involved in research and that this inactivity is due to shortage of time and funds. This may be so, but more important and less obvious constraints are the differences in the philosophical foundations of clincial practice and research. Researchers must strive to abolish uncertainty, to be unwaveringly committed to truth and to rid

Correspondence: Dr M Ward, Department of Gastroenterology, Royal Brisbane Hospital, Herston, QLD 4209, Australia. Accepted for publication 16 May 1996.

888 M Ward

themselves of all bias. Clincians, in contrast, must frequently manage patients in the absence of certainty, handle truth creatively and be willing to be biased on their patients’ behalf as part of a professional obligation. These differences constitute powerful selection pressures for individuals with particular skills, personalities and sources of job satisfaction. Clinical practice and research are instrinsically different activities and it is probably unreasonable to expect an individual talented in one to be similarly expert in the other.

Key words: clinical research, myths.

One of June Halliday’s particular talents, which I have greatly appreciated, is her ability to separate the wheat from the chaff; to take woolly ideas and expose them to the cold light of scientific scrutiny; to devise an experiment which will settle the matter. The Franciscan monk and archetypal scientist, Roger Bacon, must have had her in mind when he wrote, in the 13th century:

Neither the voice of authority nor the weight of argument are as significant as experiment, for thence comes a quiet mind

Jdne has been known to challenge the voice of authority from time to time, and rarely seems to be in a state of quiet mind, so presumably she is always between experiments. I thought therefore, that it may be appropriate to make a few obsgrvations about just what does propel clinicians towards, or away from, the quest for the quiet mind. To try to sort the wheat from the chaff, to separate the myths from the realities and end in true Baconian spirit by reviewing the results of a recent experiment.

Too busy The first of the myths is that we are too busy, so overburdened by the demands of our patients, that there is no time left for research. This is not convincing. Most doctors have a fair degree of freedom to organize their working day as they wish and most who spend all their time in clinical toil do so from preference. A preference born of the common tendency among professionals to do mostly what they mostly enjoy doing and to enjoy doing those things which talent or training, has made largely effortless and thus accomplished with relative ease, competence, and personal satisfaction. It must be conceded that there is often a shortage of protected time. The placidity needed for research is easily shattered by the irksome warbling of a pager, but even this should only be a problem where only a single clinician carries the full burden of a clinical service. Where two or more are gathered, lack of at least some protected time for research generally reflects lack of desire or lack of organisational ability.

Not enough money The second myth is that clinicians would do lots more research if only they had more money for the purpose, if only the National Health and Medical Research Council (NHMRC) weren’t so mean. Here the grain of truth

may be a little larger. There is no doubt that some very worthy projects are rejected, but it is probable that the proportion is not as large as is sometimes claimed. Last year, with some reluctance and a moderately heavy heart, I resigned as an NHMRC grant reviewer after some 15 years, mainly because, even in my own area of interest and knowledge, I was finding an increasing proportion of the applications almost completely incomprehensible. I accept that this is in part, perhaps a large part, a reflection of the struggles of an ageing clinical brain as it tries to catch each wave of molecular advance, paddling furiously to keep up with the younger riders as they surf effortlessly towards the shore. How- ever, I do not think this is the whole problem. Frequently, as I struggled to grasp just what profound questions lay behind the dense, turgid, technical prose, I came to the sad conclusion that all that was hidden, was a dense, turgid, technical idea. I was left with the same slightly bemused feeling as I experience when watching synchronized swimming, in which elegant young women put their heads under water and wave their feet in the air with exquisite precision; a feeling of: yes it’s all very clever, but is it really worth the effort? But I had no way of knowing, so it seemed time to ride off into the sunset.

It is also instructive to recall that John Snow, equipped with only paper and pencil, mapped the distribution of cases of cholera in the epidemic in London in the 184Os, deduced that the common factor was contaminated water from the Broad Street pump and aborted the epidemic by the simple expedient of removing the pump handle. Lest we think that all the cheap and easy questions have been answered, we should also note that Barry Marshall and his colleagues, equipped with only humble endoscopes and culture plates, unleashed Helicobacter pylori on an unsuspecting and sceptical world and re-wrote our understanding of the pathogenesis of peptic ulcer.’ What Snow and Marshall had in common was, not a large grant, but perception. The ability to realize the significance of evidence that was there for all to see.

Realities: Human frailties

The problems of the ‘curiosity challenged’ An under-recognized, or perhaps under-acknowledged, inhibitor of research is the simple fact that some people are blissfully unaware of the questions that need to be answered. That is not to say they are dim, they just resemble Asquith, the British Prime Minister who was said to be ‘... equipped with an excellent mind, undistracted by too much curiosity.’ Too much curiosity is not invariably beneficial, however, and others are

Career paths in medial research 889

hampered as much by lack of discipline, as by lack of ideas as they flit like butterflies from one intriguing notion to the next, never settling down to the tedious but crucial tasks of organizing the logistics of answering the answerable.

Burnet’s dark secret Any examination of human frailty in this context must also address a mystery: just why it is that a career in science is so appealing to so many bright young people? The pay is appalling, the job security non-existent and their vital contributions are often met with incompre- hension and seeming indifference. A clue to their persistence is perhaps to be found in Macfarlane Burnet’s retirement speech, when he obviously felt safe enough to disclose the guilty secret:

We can think of research essentially as intellectual play and equate it not with ethics or industry but with organized sport. Anyone who has ever made a significant discovery knows that experimental research is fun ... the pleasure of skilful manipulation of material things ... the pleasure of discussing strategy ... the elation of success.z

Fun, pleasure, elation? These words led me to wonder whether there may be an even darker secret, whether, as in other illicit pleasures, there is an element of compulsion, research as a disease process, an addiction. The evidence is there: scientists start as innocent youngsters, by asking a simple and apparently harmless question, but find the answer strangely unsatisfying. They then have to go on to bigger and harder questions, but still find no peace. The familiar downward spiral continues. They are soon to be seen in ragged groups hanging around the metaphorical park benches of QIMR, WEHI and other dens of iniquity, waiting for their dealers to arrive: elderly professors back from the latest conference, notebooks bulging with new ideas with which to feed the hopeless dependence of gullible young minds.

Those who find this fanciful should consider the questions we teach undergraduates to detect alco- holism. These are shown in Table 1; all I have done is change the word drinking to science. Bear in mind that positive answers to only two of these questions are considered indicative of a drinking problem where alcohol is concerned. I leave you to search your souls.

Realities: The peculiarities of the clinical practice

Uncertainty management If research is so much fun, another mystery immediately presents itself: why don’t more clinicians indulge themselves? This may well be due to the peculiarities of the clinical mind-set and environment. The first of these is the way clinicians is deal with uncertainty:

The scientist must strive to reduce uncertainty, the clinician must accept and manage the intrinsic uncertainty of the human condition in all its splendid variety. The scientist must be reductionist, the clinician, globaL3

This was illustrated by Bolinger who measured the reduction in uncertainty accompanying clinical manage- ment and showed that the major drop was between first contact and problem form~lation.~ It is in this zone of maximum uncertainty that clinicians are most skilled and effective as pattern re cognizer^,^ whereas researchers are at their best when all the variables are defined and controlled. Clinicians obtain job satisfaction from managing uncertainty, scientists by abolishing it.

Mechanisms and outcomes Clinicians and scientists also differ in their approaches to mechanisms and outcomes: the scientist must have a primary interest in mechanisms, the clinician in outcomes. The patient is not terribly enthralled by an explanation of their pain at a molecular level, they just want it to go away. Humans vary in their interest in mechanism and outcomes and tend to occupy different zones of a matrix (Fig. 1) In the bottom left corner are the ‘total apathetics’ interested in nothing, to the right the ‘clinical myopics’ who don’t care how something works as long as it does. Their counterparts are the ‘academic indulgents’ who are only concerned with how things( work and not whether they do and in the top right corner the ‘agitated paralytics,’ so interested in everything that they do nothing. The message is clear and well known to Buddhists: we should seek ‘the middle way’.

Tri’ing with the truth Clinicians and scientists approach the truth in rather different ways:

The scientist must be honest, often to the point of discomfort, the clinician must sometimes withhold the unvarnished truth if the hopeless are to keep hope, the incompetent are to be kept afloat, and the anxious are to be ~ a l r n e d . ~

My conscience is clear when I say to a patient with inoperable oesophageal cancer not ‘The average survival of patients in your condition is 3 months,’ which is true, but rather ‘I have known patients in your condition who have lived several years without major problems’, which is also true. In clinical practice, the bell curve must be handled creatively.

Giving bias a good name Bias is not an axlathema to clinicians :

It is also sometimes overlooked that the clinician has a professional obligation to be biased. To be biased that is,

Table 1 Detecting problem researchers

Have you ever felt the need to cut down on your Science ? Have you ever been annoyed by other people criticising your Science? Do you ever feel bad or guilty about your Science? Do you ever have to do Science fmt thing in the morning to steady your nerves?

M Ward 890

on the patient’s behalf, to be an advocate: a role which is at odds with the need to be an impartial judge in matters scientific and a source of conflict of interest in clinical research; does the greater scientific good justify the potential hazard to the individual ~at ient .~

I felt rather better about admitting this bias when I came across Sonnenberg’s recent spectrum of culpability of the lies that researchers tell (Table 2).6 It was with smug satisfaction that I noted bias was considered the mildest of sins, although clinicians do sometimes flit into the red and orange zone when they overlook or ignore awkward facts that don’t fit their pet clinical hypothesis. We rarely completely fabricate patients however. I suppose we steal them occasionally.

The answerable versus the interesting

Clinicians also differ in the type of questions they would like to research, but often conclude that the interesting questions are often unanswerable and the answerable questions often uninteresting. In our journal club, the seasons are marked not by the return of migrant birds but by the return of migrant questions. We thus have a fondness for debating at least once a year whether steroids are beneficial in the treatment of alcoholic hepatitis, as we carefully pore over the latest meta- analysis. My mind, I must confess, tends to wander on such occasions. I am reminded of the apocryphal economics professor who was chided by his colleagues for setting the same examination questions each year. He replied that the questions may be the same, but he changed the answers each year.

Perhaps one day we will hear tell of how to stop people poisoning themselves with alcohol in the first place.

The peculiarities of the clinical environment When clinicians do get involved in research they tend to work in their comfort zone with patients or patient populations, an area of study that has its own particular

v)

E ? C m .c V m E E

Academic Agitated self-indulgence paralysis

Total Clinical apathy myopia

Interest in outcomes

Figure 1 The outcome-mechanism matrix.

Table 2 The Sonnenberg spectrum of scientific lies

Infrared Red Orange Yellow Green Blue Violet Ultraviolet

Bias Overlooking or dropping data Ignoring data that do not fit the hypothesis Beautifying data Altering data Inventing data Inventing experiments Stealing data

problems, as noted by the clinical epidemiologist Alvan Feinstein:

This type of research is both difficult and unfashionable, it is difficult because the study of sick people requires much more effort and intellectual versatility than the study of phenomena observable in the laboratory. A rat or test tube is always available and replaceable, keeps its appointments promptly, does not move out of town, makes no demands upon the investigator, requires no informed consent and can readily be destroyed to find out what happened.’

CONCLUSION

A recent editorial* posed the question of ‘does research make for better doctors?’ and concluded that it did, but mostly on the same grounds as exercise is promoted: for probable but uncertain future benefit. Clinical practice and research are intrinsically different activities and individuals who function at the highest levels in both are rare. We should celebrate those who can and do occupy this difficult dual role, but they are the exceptions that prove the rule. In most cases persuading reluctant clinicians to undertake research is rather like persuading rugby league forwards to become ballet dancers: possible, but not, as a rule, a pretty sight.

Finally, if we do want to produce more hybrid clinician-researchers, how would we appease Roger Bacon? What experiment would we devise? I would suggest that all we need is the traditional technique that is always used to produce hybrids, a technique well known to another monk experimenter Gregor Mendel and to the royal families of Europe: selective breeding. The evidence is before us in the shape of the Halliday family tree in which two academic researchers, June and Bill, have produced three clinicians: a general prac- titioner, a gastroenterologist and an orthopaedic surgeon. This leads to the inevitable question of what happens if you cross two clinicians? We know. We only have to consider the family tree of the Wards to find the answer: two clinicians, a general practitioner and a gastroenterologist, have produced 1 academic researchers, a molecular microbiologist and a computer engineer with entrepreneurial tendencies. I’m not sure whether there is a Mendelian pattern here, or some peculiar recombination or whether a knockout mouse

Career paths in medical research 89 1

would help resolve the issue, but clearly this is a problem for June to ponder in her retirement. A suitable question for a delightfully unquiet mind.

REFERENCES

1 Marshall BJ, Warren JR. Unidentified curved bacillus in the stomach of patients with gastritis and peptic ulceration. Lancet 1984; 1: 1311-15.

2 Charlesworth M, Farral L, Stokes T, Turnbull D. Life among the scientists. Oxford: Oxford University Press, 1989.

3 Ward M. Clinicians in research: Why so few? Roc. R.

4 Bolinger RE, Ahlers PA. The science of ‘pattern recognition’. 3AMA 1975; 233: 1289-90.

5 Blois MS. Clinical judgement and computers. N. Engl. J. Med. 1980; 303: 192-7.

6 Sonnenberg A. How to tell the truth and avoid the urge to lie: Planning a career in medical research. HepatoZogy.

7 Feinstein AR. The quality of data in the medical record.

8 Editorial. Does research make for better doctors? Lancet

Coll. PhyS. 1993; 23: 90-2.

1995; 22: 974-81.

Comp. Biomed. Res. 1970; 3: 426-35.

1993; 342: 1063-4.

Journal of Gastroenterology and Hepatology (1996) 11, 89 1-894

Career paths for clinical scientists

GEOFFREY C FARRELL

Storr Liver Unit, Department of Medicine, University of Sydney at Westmead Hospital, Westmead, New South Wales, Australia

Abstract In the 16 years since James Wyngaarden proclaimed the clinical investigator to be an endangered species, attempts to revive this fragile beast have met with limited success in North America, the UK and Australia. The situation may be more healthy in some western European countries and Japan, but in many parts of Asia clinical investigators have vanished without trace. An analysis of the Australian context during the past 16 years suggests a gradual decline in absolute numbers of clinical scientists reaching maturation, coupled with an extraordinary diminution of their research fertility relative to that of basic scientists. In the present review, it is argued that clinical scientists have a vital role to play in medical research and, particularly, in clinical research. The reason why fewer medical graduates are entering and even fewer are being retained in medical research careers cannot be attributed to restrictions at entry, according to the availability of and competition for training scholarships. Other possible explanations include the late age of entry and the negative influence of role modelling. The latter operates directly by the attitudes and pathways of peers and indirectly through the biases of peer review. There is also a perception, possibly a realistic one, that insuperable barriers exist to obtaining a stable career position at the end of training. Finally, there is real concern about whether clinical investigators will be able to compete successfully with basic researchers for research grants. If this summation is correct, the solutions include active recruitment for research training at an earlier age, simultaneous research and medical training and truncation of postgraduate clinical training in the medical specialties with earlier introduction of medical research. In addition to these strategies, the proper training of clinical scientists must afford them generic research skills and social adaptation to a team approach with basic scientists.

Key words: careers, clinical scientist, funding, medical research, team approach, training.

INTRODUCTION medical graduates and young faculty.’ As evidence, he cited decreased participation of MD graduates in

In 1979, James Wyngaarden, then president of the postdoctoral traineeships, fellowships and career develop- Association of American Physicians and himself a ment awards, as well as a static number of MD-qualified distinguished investigator, decried the declining interest in Principal Investigators (PI) for National Institutes of biomedical research on the part of medical students, Health (NH) research grants compared with a bur-

Correspondence: Professor Geoffrey C Farrell, Storr Liver Unit, Department of Medicine, Westmead Hospital, Westmead,

Accepted for publication 16 May 1996. NSW 2145, Australia.