Where are all the surgeons in clinical academia?

Abstract

Introduction

High-quality research into surgical disease will benefit surgical patients. Whereas nearly one-fifth of National Health Service (NHS) England consultants are surgeons, less than 5% of the government’s health research funding supports surgical research.

Methods

Using an observational study, we identified surgeons in active research fellowships and on selection panels for the three largest pan-specialty medical funding bodies in the UK. We quantified the proportion of editorial board members that are surgeons, and the proportion of surgical research published over a 1-year period in the New England Journal of Medicine, The Lancet and the British Medical Journal.

Results

Some 185/1,579 (12%) of research fellowships held by clinicians were awarded to surgeons, with relatively fewer surgeons holding senior fellowships compared with predoctoral fellowships. Across the three research funding bodies, 9/165 (5%) of the clinical panel members were surgeons, whereas for the three pan-specialty journals, 5/84 (6%) of the clinical editorial board members were surgeons. Of the 541 original articles published by the same three journals, only 45 (8%) were classified as surgical.

Conclusions

We show that surgeons were underrepresented across differing domains of clinical academia. The causes of this are likely multifactorial; there are fewer senior surgeons occupying decision-making positions, fewer role models in senior fellowship positions and surgical training may leave less time to engage in research. We propose further qualitative research within the surgical community, funding bodies and journals to understand the origins of the problem and begin to form evidence-based solutions.

Introduction

Clinical academics are medical doctors who are active in medical research and have undergone formal research training; the majority have undertaken a higher degree such as a PhD. They are fundamental in enabling the progression of scientific research into clinical practice, as highlighted in the UK Government’s Industrial Strategy for the future.¹ The Integrated Academic Training (IAT) Programme is a training pathway in the UK that incorporates protected research time into clinical training. The National Institute of Health Research (NIHR) funds 250 academic clinical fellowships (ACFs) and 100 academic clinical lectureships (ACLs) each year in England to support academic clinical trainees at pre- and postdoctoral levels, respectively.² Several UK funding bodies have research fellowship schemes to support clinicians at various career stages, such as the NIHR Clinical Doctoral Research Fellowship (doctoral studentship), the Wellcome Trust Clinical Research Career Development Fellowships (early independent researchers), and the Medical Research Council (MRC) Senior Clinical Fellowship (established researchers).

In 2014–2015, 34% of hospital admissions in England required surgical care.^3,4 In February 2020, 18% of the 49,581 consultants (most senior hospital doctors) working for the National Health Service (NHS) in England worked in a surgical specialty as defined by the Royal College of Surgeons of England (hereafter the RCS; Supplementary Table 1).^5,6 However, surgical research projects have been estimated to receive less than 5% of the government’s health research funding.⁷ The RCS states that undertaking research and understanding evidence-based research methodologies are beneficial to all surgeons,⁸ and those surgeons with a particular interest in academia are encouraged to pursue a clinician–scientist career. However, given the nature of surgery as a craft specialty, finding the time to engage in research while maintaining technical skills is particularly challenging.^9,10 Over the past two decades, not only has funding for surgical research diminished, but surgeons’ interest in research may also have fallen.^11–13 Although there has been an increase in applications from physicians for research grants over recent years, the application volume for surgeons has remained static.¹⁴ We therefore hypothesised that there is an underrepresentation of surgeons and surgical research within clinical academia in the UK.

First, we sought to quantify the representation of surgeons in different spheres of academic medicine by calculating: (i) the proportion of clinicians awarded research fellowships at the largest pan-specialty medical funding bodies in the UK who are surgeons; (ii) the proportion of clinicians who sit on the selection panels for those fellowships who are surgeons; and (iii) the proportion of surgeons on the editorial boards of the highest impact pan-specialty clinical journals. Subsequently, we quantified the proportion of surgical research published in those same journals.

Methods

Definitions

‘Non-clinicians’ were defined as any professional without a medical degree. ‘Non-surgical clinicians’ were defined as medical doctors who worked in any specialty other than surgery. Surgeons were defined as doctors who work in one of the ten surgical specialties as defined by the RCS,⁵ who were a fellow or member of one of the four Royal Colleges of Surgeons in the UK, or their overseas equivalents.

Data sources

Information was collected from research funding bodies and medical journals through direct correspondence using Freedom of Information requests. The data were provided in full, or collected from appropriate alternative sources suggested by the individual organisations.

Research funding

First, we quantified the number and proportion of surgeons who are awarded active research fellowships from the largest pan-specialty medical funding bodies in the UK: the NIHR; UK Research and Innovation (UKRI), which includes the Medical Research Council (MRC); and the Wellcome Trust.¹⁵ NIHR fellowships are available in England rather than the UK. These data were categorised by seniority of fellowship with senior research fellowships defined as those at postdoctoral level. Next, we calculated the number and proportion of surgeons on the selection panels for these funding bodies.

Research publication

We quantified the number and proportion of surgeons on the editorial boards, defined as active editors, associate editors or members of the wider editorial board, for three of the highest impact pan-specialty clinical journals: New England Journal of Medicine (NEJM), The Lancet and the British Medical Journal (BMJ). These journals were selected because they publish high-quality research from the full spectrum of medical specialities.

Next, we analysed all full-length research articles published by these three journals over a 1-year period from 1 February 2019 to 1 February 2020, prior to the Covid-19 global pandemic. Articles were classified according to the medical specialty that most closely aligned with the research content. They were further classified as either surgical articles or non-surgical articles. Surgical articles were defined as those that assessed the efficacy of a surgical technique, or those that pertained to disease states treated by surgeons. For example, any study that uses a surgical technique (eg heart valve replacement) was included as ‘surgical’, whereas studies that use drugs to treat ‘surgical disease’ (eg chemotherapy for breast cancer) were not included as ‘surgical’. Titles and abstracts were screened and classifications were independently assigned by three authors. Any disagreements were resolved by mutual agreement after review of the full text.

Comparison of research fellowships with clinical workforce

Using public data from the 2017 MRC fellowship survey¹⁶ and the NHS England Workforce Monthly Statistics for February 2020,⁶ we calculated ratios for each specialty comparing the number of fellowships held by clinicians with the number of consultants working in these specialties. We used the number of consultants as a proportion of the total workforce as a surrogate measure to quantify the relative size of the specialty. Consultant data were used because consultants are senior, hospital-based clinicians on the specialist register for their specialty who have completed their entire specialist training.

Results

Research fellowships

Across the NIHR, UKRI and Wellcome Trust, surgeons held 12% of the 1,579 active research fellowships awarded to clinicians. Surgeons held 15% of fellowships awarded to clinicians at the NIHR, compared with 9% at UKRI and 3% at the Wellcome Trust. Consistently across the three research bodies, surgeons held a progressively smaller proportion of fellowships as seniority increased (Table 1).

Table 1 .

Breakdown of active fellowships held by clinicians awarded by the National Institute of Health Research (2019), UK Research and Innovation (2019) and Wellcome Trust (2020)

	Fellowships held by clinicians	Non-surgical clinicians (%)	Surgeons (%)
National Institute of Health Research	1,097	937 (85)	160 (15)
Postdoctoral	352	301 (86)	51 (14)
Professorships (chair/pre-chair)	17	16 (94)	1 (6)
Other post-doctoral (excluding professorships)	335	285 (85)	50 (15)
Predoctoral	743	634 (85)	109 (15)
Doctoral*	51	46 (90)	5 (10)
Predoctoral	692	588 (85)	104 (15)
Miscellaneous	2	2 (100)	0 (0)
UK Research and Innovation	177	161 (91)	16 (9)
Postdoctoral	57	55 (96)	2 (4)
Senior clinical fellowships	6	5 (83)	1 (17)
Clinician scientist fellowships	46	45 (98)	1 (2)
Clinical research training fellowships (postdoctoral)†	5	5 (100)	0 (0)
Predoctoral – Clinical research training fellowships	120	106 (88)	14 (12)
Wellcome Trust	305	296 (97)	9 (3)
Postdoctoral	172	169 (98)	3 (2)
Principal fellowships	14	14 (100)	0 (0)
Senior fellowships	50	50 (100)	0 (0)
Intermediate fellowships	33	33 (100)	0 (0)
Early career fellowships	75	72 (96)	3 (4)
Predoctoral	133	127 (95)	6 (5)
PhD training fellowship for clinicians	106	101 (95)	5 (5)
Research training fellowships	14	13 (93)	1 (7)
International training fellowships	9	9 (100)	0 (0)
International Masters fellowships	4	4 (100)	0 (0)
Total	1,579	1,394 (88)	185 (12)

NIHR = National Institute of Health Research; UKIR = UK Research and Innovation

*NIHR doctoral fellowships support individuals undertaking a PhD, NIHR predoctoral fellowships support individuals undertaking a Master’s degree.

†Five of the UKRI clinical research training fellowships were held by postdoctoral scientists who had been out of research for an extended period.

Selection panel membership

Fellowship selection panel data were analysed for the NIHR, UKRI and Wellcome Trust (Table 2). Of the 59 clinical panel members across the NIHR boards, 5 (8%) were surgeons. Across the 11 UKRI research boards and panels for which data are published, 3 of the 85 clinical panel members were surgeons (4%).¹⁷ The Science Interview Panel and Clinical Interview Committee at the Wellcome Trust had 1 surgeon among the 21 clinical panel members.^18,19

Table 2 .

Breakdown of clinical members of fellowship selection panels for the National Institute of Health Research (2018), UK Research and Innovation (2020) and the Wellcome Trust (2019)

	Clinical panel members	Non-surgical members (%)	Surgical members (%)
National Institute of Health Research	59	54 (92)	5 (8)
UK Research and Innovation	85	82 (96)	3 (4)
Wellcome Trust	21	20 (95)	1 (5)
Total	165	156 (95)	9 (5)

Membership of editorial boards

For the NEJM, The Lancet and the BMJ, 6% of the total clinicians who were editors, associate editors, and members of the editorial board were surgeons (Table 3). This represents two surgeons at the NEJM and The Lancet, and one at the BMJ.

Table 3 .

Breakdown of editors, associate editors and the wider editorial board for the New England Journal of Medicine (2019), The Lancet (2019) and the British Medical Journal (2020)

	Clinical members of the editorial board	Non-surgical clinicians (%)	Surgeons (%)
New England Journal of Medicine	32	30 (94)	2 (6)
The Lancet	31	29 (94)	2 (6)
British Medical Journal	21	20 (95)	1 (5)
Total	84	79 (94)	5 (6)

Representation of surgical research in high-impact clinical journals

We identified 541 original articles in total published between 1 February 2019 and 1 February 2020: 201 articles from the NEJM (Supplementary Table 2), 184 articles from The Lancet (Supplementary Table 3) and 156 articles from the BMJ (Supplementary Table 4). There were 45 (8%) surgical articles in total: 12 (6%) were published by the NEJM, 23 (13%) by The Lancet and 10 (6%) by the BMJ (Table 4 and Figure 1). Subsequent breakdown of the surgical articles into subspecialty is shown in Figure 2.

Table 4 .

Classification of articles published by the New England Journal of Medicine, The Lancet and the British Medical Journal (1 February 2019 to 1 February 2020)

Article classification	NEJM	The Lancet	BMJ	Total (%)
Public Health & Epidemiology	2	12	52	66 (12)
Infectious Diseases & Immunology	33	21	5	59 (11)
Cardiology	27	18	5	50 (9)
Oncology	24	20	6	50 (9)
Surgery	12	23	10	45 (8)
Neurology	16	14	3	33 (6)
Obstetrics & Gynaecology	7	18	7	32 (6)
Endocrinology & Metabolic Medicine	11	6	13	30 (6)
Haematology	20	4	1	25 (5)
Respiratory	12	6	2	20 (4)
Psychiatry	2	7	8	17 (3)
Anaesthetics, Pain & Intensive Care Medicine	8	2	6	16 (3)
Paediatrics & Neonatology	7	5	4	16 (3)
Gastroenterology	4	4	7	15 (3)
Renal	6	4	2	12 (2)
Primary Care	3	1	4	8 (1)
Rheumatology	2	5	0	7 (1)
Dermatology	3	2	0	5 (1)
Ophthalmology	0	5	0	5 (1)
Emergency Medicine	0	1	2	3 (1)
Geriatric Medicine	0	2	0	2 (0)
Other	2	4	19	25 (5)
Total	201	184	156	541

BMJ = British Medical Journal; NEJM = New England Journal of Medicine

Figure 1 .

Breakdown of articles published by the New England Journal of Medicine, The Lancet and the British Medical Journal (1 February 2019 to 1 February 2020). All 541 original articles in the three journals were classified according to the medical specialty most closely aligned with the content of article.

Figure 2 .

Breakdown of surgical articles (n=45) published by the New England Journal of Medicine, The Lancet and the British Medical Journal (1 February 2019 to 1 February 2020) by surgical subspecialty

Comparison of research fellowships to clinical workforce

In 2017, the MRC conducted a UK-wide survey of clinical and health research fellowships¹⁶ and published the data online.²⁰ The survey identified 2,145 research fellowships held by clinicians and these were categorised by clinical specialty (Figure 3). We used public data from the NHS England Workforce Statistics to identify the number of consultants working in surgery and each medical specialty.⁶ Using these data and the MRC 2017 fellowship data, for surgery and the medical specialties, we subsequently calculated ratios comparing the number of fellowships with the number of consultants (Table 5). Strikingly, the proportion for surgery was less than for any medical specialty, and in some medical specialties the ratio was five- to tenfold higher than the ratio for surgery, further suggesting an underrepresentation of surgeons in these fellowship positions.

Figure 3 .

Distribution of research fellowships held by clinicians across different medical specialties, using publicly available data²⁰ from the 2017 UK-Wide Survey for Clinical and Health Research Fellowships¹⁶

Table 5 .

Research fellowships awarded to clinicians from surgical and medical specialties, compared with the proportion of NHS England consultants working in these specialties

Clinical specialty	Number of fellowships (% of total fellowships held by clinicians)	Number of consultants (% of total consultant workforce)	Ratio of number of fellowships to number of consultants
Infection/microbiology	127 (5.9)	612 (1.2)*	0.208
Neurology	97 (4.5)	820 (1.7)	0.118
Oncology	143 (6.7)	1214 (2.4)	0.118
Renal	69 (3.2)	599 (1.2)	0.115
Cardiology	144 (6.7)	1422 (2.9)	0.101
Haematology	77 (3.6)	875 (1.8)	0.088
Rheumatology	47 (2.2)	629 (1.3)	0.075
Endocrinology	51 (2.4)	762 (1.5)	0.067
Respiratory medicine	76 (3.5)	1149 (2.3)	0.066
Gastroenterology	70 (3.3)	1350 (2.7)	0.052
Physicians/medicine	115 (5.4)	2709 (5.5)†	0.042
Surgery	221 (10)	8837 (18)	0.025

Percentages are calculated using the data for every clinical specialty (including specialties from outside medicine and surgery). Using public data from the 2017 UK-Wide Survey for Clinical and Health Research Fellowships²⁰ and the NHS Workforce Monthly Statistics for February 2020.⁶

*Includes consultants from infectious diseases, medical microbiology and virology

†Includes consultants from acute internal medicine, general internal medicine and geriatric medicine

Discussion

We investigated the presence of surgeons in research fellowship positions, selection panels and journal editorial boards to give a picture of the representation of surgeons in clinical academia in the UK.

For the three largest pan-specialty UK research-funding bodies, surgeons were overall awarded 12% of the 1,579 active research fellowships awarded to clinicians. There were large differences between funding bodies. The NIHR was well represented with surgeons awarded 15% of its fellowships held by clinicians, compared with 9% and 3% for the UKRI and Wellcome Trust, respectively. Across all funders, there was a marked decrease in surgeons holding more senior fellowships. For example, at the NIHR, 50 of the 335 postdoctoral research fellowships were held by surgeons, in contrast to 1 of the 17 professorial fellowships. Furthermore, we found evidence that surgeons are underrepresented on the selection panels for these fellowships; only 5% of clinical panel members for the NIHR, UKRI and Wellcome Trust were surgeons.

We also identified underrepresentation of surgeons in editorial positions and on editorial boards for three high-impact pan-specialty medical journals: the NEJM, The Lancet and the BMJ. Across these three journals, only 6% of the clinical editors, associate editors or members of the wider editorial boards were surgeons, representing just five individuals in total.

We quantified the proportion of articles published in the same journals over a 1-year period and found that surgical research articles comprised 8% of the total. By way of comparison, 9% of articles were classified as belonging to the medical subspecialty cardiology, and 9% to the subspecialty of oncology.

For the 2017 MRC survey on clinical fellowships, although it may at first appear that surgeons are well represented in these fellowship positions, this is an artefact resulting from general medical specialties (cardiology, oncology, infection/microbiology, physicians/medicine, neurology, haematology, respiratory, gastroenterology, renal, endocrinology, rheumatology) being presented by subspecialty rather than by the umbrella term of ‘general/internal medicine’. By summing the numbers of fellowships awarded to these medical specialties, they outnumber the surgical fellowships by a factor of 4.6 to 1. Similar results were also seen in the 2009 survey that the MRC conducted.²¹ Our calculated ratios for surgery and each medical specialty, comparing fellowship numbers with number of consultants, highlight that when compared with their medical counterparts, surgeons are underrepresented in research fellowship positions relative to the number of surgeons in the workforce.

Study implications

From Joseph Murray’s first solid organ transplant to Shinya Yamanaka’s pioneering work on induced pluripotent stem cells, surgeons are well placed to make important contributions to medical science. It is therefore of some concern that we have shown surgeons to be underrepresented in clinical academia in the UK across every area investigated.

Although we are unable to ascertain causal relationships for this underrepresentation, we can suggest a number of factors that may be contributory. In the process of trying to develop a technical skill, surgeons may ascribe less importance to academia than their medical counterparts, and consequently develop less interest in research. In the same vein, less investment may be allocated to active research by surgeons for fear of losing these technical skills.

Also, it may be that the findings we see in this paper are self-perpetuating; there are fewer surgical role models within academia for young surgeons to follow, as highlighted by the especially low number of the highest-level research fellowships held by senior surgeons. Given the low proportion of surgeons on the selection panels for research fellowships, there may even be an element of unconscious bias against surgeons in the selection process for such posts.

Although these suggestions are speculative, it is important that further qualitative research is conducted within the surgical community and research funding bodies to understand the origins of surgical underrepresentation. As previously stated, our findings are likely to be multifactorial and therefore a number of stakeholders are required to contribute to ensure better surgical representation across all aspects of academia. One could draw careful parallels from the approach towards addressing inequality in other underrepresented groups, such as women in surgery. By comparing the underrepresentation of surgeons to the mission set by the RCS to encourage women to fulfil their surgical career ambitions, such goals should involve a change in attitude and greater support for the underrepresented group.²²

Pan-specialty funding bodies could ensure equitable surgical representation on panels, potentially by introducing a quota, and have designated surgical fellowship schemes, especially at more senior levels where the underrepresentation appears greatest. Surgical research is rarely prioritised by research bodies.²³ Including surgery in priority areas for research could help to improve this underrepresentation. The RCS has an established research fellowship scheme, but this only provides funding for 1 year and not the 3–4 years needed for a PhD.²⁴ Recently, the RCS has taken steps to increase surgical multicentre studies, through introducing surgical specialty leads across the UK.²⁵ This is to complement their clinical trial research networks that allow clinicians of all levels to engage in multicentre national surgical research.⁷ Such interventions should be encouraged to facilitate more involvement of surgeons in research.

There are certain limitations to our study. First, we did not have access to data regarding the ratio of successful applicants by specialty for research fellowship positions; we are therefore unable to draw any conclusions about whether surgeons are disproportionately unsuccessful at the application or interview stage.

With regard to the classification of research articles, we acknowledge that while a disproportionately low proportion were considered to be ‘surgical’ articles, we used a narrow definition for this category to make the classification consistent across the three journals. Many oncological research articles pertaining to the non-surgical treatment of diseases that are managed jointly by surgeons and oncologists would still be of interest to surgeons. Furthermore, it must be noted that many high-quality surgical articles will be published in high-impact pan-surgical journals, rather than the three pan-specialty medical journals that we investigated. These pan-specialty medical journals typically prioritise publication of randomised controlled trials, which are less likely to be conducted within surgery, although further studies are required to investigate this potential study design bias.

These limitations are mitigated to some degree by the principal strength of this study, which is that we approached the question of surgical underrepresentation in academia from a very broad perspective. We quantified the representation of surgeons in three distinct domains relevant to medical academia, and corroborated the findings by ascertaining the representation of surgical articles in the highest impact medical journals. For the latter, we scrutinised over 500 journal articles published over a year-long period, and ensured that the findings were not biased by the plethora of research articles pertaining to the Covid-19 pandemic that have emerged from February 2020 onwards.

Conclusion

We have shown that surgeons are underrepresented across all areas of clinical academia in the UK that we investigated. Fewer surgeons occupy fellowships awarded by research bodies in the UK, with surgeons well represented in the NIHR, compared with the UKRI and Wellcome Trust. Surgeons are significantly underrepresented on the selection panels of these research bodies. We have shown that less surgical research is published in leading journals, which correlates with poor surgical presence on journal editorial boards. Further qualitative research is required to identify reasons for our findings in order to form constructive solutions to encourage surgeons to engage more in clinical academia.

Conflicts of interest statement

AW is a former MRC Clinical Research Training Fellow and is a NIHR Clinical Lecturer in Plastic Surgery. DF is a former Wellcome Intermediate Fellow and is a Professor of Plastic and Reconstructive Surgery.