Degenerative cervical myelopathy education in UK medical schools: a national cross-sectional survey of medical students

Abstract

Introduction

Degenerative cervical myelopathy (DCM) is a common and progressive neurological condition caused by injury of the cervical spinal cord by degenerative spinal pathology. Delayed diagnosis leading to avoidable and irreversible disability is a major current problem limiting patient outcomes. Lack of sufficient representation of DCM in undergraduate and postgraduate medical curricula may contribute to poor recognition of DCM by non-specialist doctors. The objective of this study was to assess the DCM teaching provision in UK medical schools and the DCM knowledge of UK medical students.

Methods

UK medical students completed a web-based survey distributed nationally through university social media pages, university email bulletins and the national student network of Myelopathy.org. The survey comprised a 19-item questionnaire capturing data on student demographics, myelopathy teaching and myelopathy knowledge. Advertisements were repeated monthly over a 12-month recruitment period and participation was incentivised by entry into an Amazon voucher prize draw. Ethical approval for the study was granted by the Psychology Research Ethics Committee, University of Cambridge (PRE.2018.099).

Results

A total of 751 medical students from 32 British medical schools completed the survey. Medical students from all year groups participated. Most students (520; 72%) had not received any medical school teaching about DCM. When students had received DCM teaching, the duration of teaching was minimal (75% < 1 h). A total of 350 students (47%) reported conducting private study on DCM. Modal student self-rating of their own knowledge of DCM was ‘terrible’ (356; 47%). There was no correlation between a student’s subjective rating of their knowledge and their answers to objective questions. A total of 723 (96%) of students expressed interest in learning more about DCM, with lectures the preferred format.

Conclusions

DCM appears to be a neglected condition in medical education which has implications for clinical practice. However, student enthusiasm to undertake private study suggests future teaching interventions will be well-received. Future work is necessary to characterise the format of DCM teaching that is most effective and to subsequently measure how educational interventions translate into clinical benefits.

Keywords:

• Cervical spine
• degeneration
• myelopathy
• neurology
• neurosurgery
• medical education

Introduction

Degenerative cervical myelopathy (DCM) is a common neurological condition of symptomatic cervical spinal cord compression secondary to degenerative changes of the spine.^1–4 DCM is the most common cause of spinal cord impairment in adults worldwide, with an estimated prevalence of up to 5% in the over 40s.^1,2,5,6 It is a progressive condition, with a very poor quality of life.^7–11 Surgery can halt progression but not reverse existing damage.^12,13

Substantial improvements in the care of patients with DCM could be achieved today without additional scientific understanding of the condition.^1,14,15 Patients with DCM almost universally experience diagnostic delays, with 2 years and 5 clinic appointments being typical.^16,17 Improvement in speed of diagnosis is therefore a key target that could provide immediate reductions in disability.¹⁸

Initially, patients with DCM are often seen by primary care, emergency and medical physicians.^17,19,20 These non-specialist triage points appear key to earlier diagnosis, yet DCM features poorly in medical curricula.²¹ This risks non-specialist doctors being poorly trained to detect DCM. The current provision and composition of DCM medical education are uncharacterised. Our recent evaluation of undergraduate and postgraduate medical curricula identified deficiencies in DCM representation; however, this did not correlate with student performance in multiple-choice question banks.²¹

The objective of this study was therefore to assess the teaching of DCM in UK medical schools and medical student knowledge of DCM. The aim is to identify both training and knowledge gaps to inform future initiatives to improve DCM medical education, ultimately to improve efficiency of diagnosis and patient outcomes. Improving DCM education is also an important component of the number one research priority of the AO Spine RECODE-DCM initiative: improving awareness.¹⁴ We hypothesise that DCM knowledge and awareness among medical professionals are related to their specific educational experiences.

Methods

The protocol for this study has previously been published.²²

Study partners

Myelopathy.org

Myelopathy.org is an international DCM charity, which aims to improve health and wellbeing in DCM.²³ This study forms part of a wider project to evaluate DCM education and awareness among medical professionals and develop interventions to improve this.^1,21 The charity is focused on improving medical education to address the current problems of delayed, missed and underdiagnosis.¹

Student society of myelopathy.org

The Student Society of Myelopathy.org was established in 2018 with the aim of improving DCM awareness among medical students. The society has hosted a series of educational events; established national medical student essay and research prizes; designed and initiated this study and secured sponsorship from corporate partners to support these projects.

Study design

An electronic survey was agreed to be the most efficient way to simultaneously assess teaching, perceptions of teaching and current knowledge and to efficiently reach students from all UK medical schools. Part 1 of the survey was designed to capture basic demographic characteristics and teaching on DCM. Part 2 was designed to capture knowledge and awareness of fundamental DCM facts including prevalence, symptoms, diagnostic imaging modalities, surgery and appreciation of contemporaneous issues, such as prolonged time to diagnosis and poor patient quality of life. Therefore, a breadth of DCM knowledge was assessed, encompassing multiple points in the disease pathway. Part 2 was also designed to capture participants’ perceptions of their DCM knowledge. These questions were felt to capture sufficient data while maintaining an acceptably short survey completion time of approximately 5 min.

The study management team comprised representation from second-year to final-year students from multiple UK medical schools. The steering committee comprised an academic neurosurgical registrar (BD) and an academic neurosurgical consultant (MK), both with clinical and academic interests in DCM, and an academic consultant neurologist (SRLS) who is a neuroscience teaching fellow and neuroscience course lead at the University of Cambridge. Importantly, this was a student-motivated, student-led project, conducted under the supervision and mentorship of a team of senior researchers.

The survey was iteratively refined by the study management team until Part 1 was felt to capture sufficient information on current DCM teaching provision in UK medical schools. Similarly, Part 2 was refined until it was felt to evaluate fundamental DCM facts at a level sufficient for a non-specialist doctor to appropriately recognise DCM. In total, a 19-item questionnaire was developed (Figure 1).

Figure 1. Final survey design. Reproduced with permission from Myelopathy.org.

Survey development was closely aligned with the Association for Medical Education in Europe (AMEE) seven-step guide to medical education survey development:¹ literature review,² focus group establishment,³ synthesis of literature review/focus group,⁴ item development,⁵ expert validation,⁶ feedback and expert validation and⁷ piloting,²⁴ with the exception that a quantitative gap analysis was performed in the place of a literature review due to the lack of research on DCM medical education at the time of study design.²¹

Survey piloting

The survey was piloted by a group of 20 medical students who were not involved in study design. No inconsistencies, ambiguities, nor problematic survey items were identified when data were inspected. Students reported the survey to be acceptable on both desktop and mobile devices. Informal cognitive interviews of five students revealed that survey items were interpreted as intended, without ambiguity. No changes were therefore made.

Survey administration

To maximise reach across the UK, the survey was hosted on Qualtrics survey software (Provo, UT), a commercial online survey platform. Qualtrics survey software includes both desktop and mobile-compatible versions, maximising convenience for participants.

The survey comprised two pages: a first page with a participant information sheet (PIS) and a question capturing informed consent and a second page containing all 19 questions to minimise the number of clicks required for participants to complete the survey.

Survey dissemination

A network of Myelopathy.org student representatives was recruited, with representation of students from 25 UK medical schools. Some were also representatives for the Society of British Neurological Surgeons and Neurology and Neurosurgery Student Interest Group (NANSIG). A standard advertisement approved by the research ethics committee was used for all advertising. Representatives from all medical schools were sent monthly reminder emails to disseminate adverts at their institution over a 12-month period from February 2019 to January 2020. There was no difference in the timing of reminder emails between different medical schools. Although prompted to advertise monthly, the final decision on frequency of advertising at each institution was left to the discretion of representatives, guided by local policies.

On completion of the survey participants were invited to leave an email address in a separate linked survey to be entered into a prize draw to win one of five £10 Amazon vouchers. The prize draw was agreed by the study management group and approved by the research ethics committee. The rationale was that prizes of this value may have encouraged participation without providing an inappropriate incentive to participate.

The survey was advertised in social media groups and on social media pages for relevant student groups and societies at universities across the UK. In addition, university mailing lists and email bulletins were used, where appropriate, in accordance with local rules. Adverts were also placed in the monthly NANSIG newsletter, emailed out to medical student subscribers across the UK. The survey link was also hosted on the Student Society of Myelopathy.org webpage and Facebook page.

Eligibility and representation

All UK medical students were eligible to complete the survey. We aimed to secure representation of all UK medical schools.

Consent and confidentiality

The PIS emphasised the rationale, purpose, study aims and the voluntary nature of participation. By necessity, the background information on DCM in the PIS was carefully selected to avoid biasing responses to the DCM knowledge-based survey questions.

Participants assigned themselves a unique identifier based on a defined combination of characters from their mother’s maiden name, the street they grew up on and their mobile phone number; this ensured anonymity and confidentiality while allowing the possibility of linking together future surveys from the same participant.

Data security

All data were stored exclusively on the secure, online Qualtrics platform until closure of the survey. Thereafter, survey data was extracted directly from the Qualtrics platform into a password-encrypted Excel spreadsheet (Microsoft, San Francisco, CA) on a password-protected computer. Only the immediate research team was granted access to this on a need-to-access basis. After completion of data analysis, all data that are no longer required will be deleted. No participant-identifiable data was collected or stored.

Statistical analysis

Data analysis and visualisation were performed using R version 4.0.5 (R Core Team, Vienna, Austria) and RStudio version 1.4.1106 (RStudio Team, Vienna, Austria)).

Due to all items in the survey being mandatory, completed responses had no missing data. Analysis was performed on all collected values. Statistical analysis was mainly descriptive; however, inferential statistics in the form of Kruskall − Wallis tests were used, as the data were non-parametric.

Patient and public involvement

Patients were involved in the conception, design, development and conduct of this study. A DCM Patient and Public Engagement Day hosted at the University of Cambridge was captured by Cambridge TV in a documentary.²⁵ During this day, a focus group of patients with DCM emphasised their experiences of diagnostic delays. This has been frequently echoed by patients in Myelopathy.org social media groups and in a DCM word-cloud initiative to understand patient’s perceptions of the condition for a related project.²⁶ Patients with DCM therefore inspired the conception of this study and approved the final survey design. Patients also assisted in the essential administrative tasks of the project. The online survey for the study was hosted on Myelopathy.org, which is a patient-maintained website for people with DCM. Patients were, therefore, active in aiding the dissemination of the survey to UK medical students. Patients who were involved in preparation of the manuscript are among the authors. Patients with DCM are involved in plans to disseminate this research to the patient community, including blog articles on Myelopathy.org, posts in online patient support groups and presence at spinal conferences in the UK.

Results

Summary

A total of 751 medical students from 32 (86%) UK medical schools completed the survey. Medical students from all year groups participated, with 4th-year students the modal year group (26%). A specialist interest in neurology or neurosurgery was declared by 120 students (16%). Dundee (91, 12%), Manchester (77, 10%) and Cambridge (52, 7%) were the best represented medical schools, comprising 29% of the responses. The full dataset can be found in Supplementary Material 1.

Current standard of education

The majority of students (520; 72%) had not received any medical school teaching on DCM (Figure 2). This still holds true when early years/preclinical students (1^st- and 2nd-year students) are excluded (333/543; 61%).

Figure 2. Student responses to the question ‘Have you received any cervical myelopathy teaching from your medical school?’ for each medical school.

A total of 269 students (36%) reported that they had never heard of a condition called either DCM or cervical spondylotic myelopathy. A few medical schools, namely Plymouth and Norwich UEA, were exceptional cases in which most respondents had received DCM teaching.

The reported duration of teaching was minimal for most students (Figure 3). Amongst the minority of students who had received teaching, only 25% reported that they had received at least 1 hour of DCM teaching.

Figure 3. Duration of DCM teaching received by students.

A minority of students (350, 47%) reported conducting private study on DCM. The modal student self-rating of their DCM knowledge was ‘terrible’ (Figure 4). Only 156 students (21%) rated their knowledge as average or better.

Figure 4. Student rating of their own knowledge about DCM.

Objective student knowledge

When asked 6 objective questions to test knowledge of DCM, 56 ± 29% of student responses were correct.

There appeared to be no correlation between a student’s subjective rating of their own knowledge and their answers to objective questions (Figure 5; Supplementary Material 2). Furthermore, no correlation was observed when these data were divided by question topic. The number of correct answers was also not significantly different when considering students with a declared specialist interest in neurology/neurosurgery (Figure 6).

Figure 5. The proportion of correct answers as a function of student subjective rating of DCM knowledge.

Figure 6. The proportion of correct answers as a function of student specialist interest.

Quality of life perception

When students were asked to rank five conditions in terms of predicted quality of life from best¹ to worst,⁵ DCM was ranked as the condition with the second-best quality of life (Figure 7). The mean ranking of DCM was worse than diabetes, better than cancer and similar to depression and heart failure.

Figure 7. Student estimates of patient quality of life from best¹ to worst⁵ in five conditions.

Future learning

Most students (723, 96%) were interested learning more about DCM, with lectures the modal preferred format for future learning (Figure 8).

Figure 8. Preferred format for future learning about DCM.

Discussion

Our results suggest that most students at UK medical schools have never received DCM teaching. Of the few students who have received DCM teaching, the median duration of teaching was 15 min. Although students rated their mean subjective knowledge of DCM as ‘terrible’, performance in objective questions exceeded this expectation. There was a desire amongst most students to learn more about DCM, with a preferred format of lectures.

DCM is a neglected condition in medical education

With an estimated prevalence of 2.7%,⁶ the reported absence of DCM teaching in UK medical schools is at odds with the General Medical Council’s ‘Outcomes for Graduates’, Outcome 2 – Professional Skills.²⁷ Of note, graduates are expected to ‘synthesise findings from the history…examinations and investigations…and make proposals about underlying causes or pathology’. They are also expected to ‘assess and determine the severity of a clinical presentation’. A knowledge of prevalent conditions, such as DCM, is necessary to fulfil these criteria.

The responses to our survey align with a prior study investigating references to DCM in student curricula and commonly used medical texts.²¹ DCM was shown to have the lowest modal rank of references relative to conditions of similar prevalence.

Together, these findings have a growing relevance to clinical practice. There are currently significant delays to DCM diagnosis,¹ with the largest contributor being misdiagnosis in primary care.¹⁶ Process mapping of the diagnostic pathway in the UK has provided a detailed assessment of this journey to diagnosis.²⁰ Knowledge of DCM is therefore essential for the generalist, and this necessitates introduction at the undergraduate level.

Students underestimate the effect of DCM on quality of life

Students ranked DCM as the second-best condition out of five for its impact on quality of life. Whilst the ranking relative to heart failure and depression may be equivocal, the impact of cancer was ranked considerably greater than DCM. This contrasts with patient quality of life data, demonstrating that DCM causes the worst cumulative mental and physical disability.²⁸ It is reasonable to expect that such assumptions may persist into clinical practice. Such a mismatch between patient experience and clinician expectation can hinder effective patient care, with underestimation risking translation into under prescribing and referral. Examples of this can be seen in the pain literature, concerning biases regarding race and gender.^29,30

DCM Findings may reflect broader issues in neurological education

Diagnostic delays exist for other neurological diseases, including dementia,³¹ amyotrophic lateral sclerosis (ALS)³² and epilepsy.³³ Whilst this can be partially attributed to the insidious nature of many neurological conditions, educational shortcomings are frequently cited.^34,35 Such shortcomings are thought to result from the phenomenon of neurophobia. First formally described in 1994,³⁶ neurophobia describes an aversion amongst medical students and young doctors to understanding neurological disease. This is to some extent self-fulfilling, with students believing that neurology is the most difficult medical specialty.³⁷ In our study, neurophobia may well have influenced the subjective perceptions of student DCM knowledge.

Although proposals, such as better integration of the preclinical sciences and more small group teaching, are often discussed,³⁸ a recent systematic review³⁹ has found that there is limited high-quality evidence for interventions in neurology education. Consequently, a cultural inertia may limit future improvements in DCM knowledge.

Student enthusiasm provides a cause for optimism

There was an explicit desire from virtually every student in our survey to have additional DCM teaching. This demand, alone, is promising.

Moreover, private study of DCM was conducted by 350 students (47%), indicating that at least 118 (16%) tried to compensate for what was not provided by their medical school. The average proportion of correct answers to our objective questions was 56%. This is perhaps also influenced by student encounters with DCM patients, which was not assessed in this study.

The willingness to undertake private study and a reasonable level of baseline knowledge suggest that only a modest addition of DCM teaching to medical curricula may be sufficient to substantially increase student competence and confidence.

Limitations and future work

Our survey depended primarily on student recollection of teaching, making it vulnerable to recall bias. This was minimised by having a large sample size.

Some medical schools may have been disproportionately represented, with Dundee, Cambridge and Manchester comprising 29% of our responses. However, high representation from these medical schools may reflect a strength in this study, as each school is situated in a different region of the UK. Furthermore, responses were similar from schools with smaller representation. In addition, included students from each medical school comprise only a small percentage of all students at the school, risking introducing a selection bias. Steps to mitigate against this were taken, including a wide and varied advertising strategy and sub-analysis comparing students with and without a specialist interest in neurology or neurosurgery.

The lack of correlation between student subjective and objective knowledge is interesting. Five questions may well have been too small a number to identify a significant association. Furthermore, it may be that some questions were simple for all students, while the more complex questions may have relied on guesswork. Alternatively, the Dunning − Kruger effect may provide a contribution to this finding.⁴⁰ It should also be noted that following the introduction of DCM educational materials, performance for DCM questions in student question banks is above the bank average,²¹ suggesting that the application of general neurological principles may counterbalance student perceptions of their own knowledge.

Similarly, we observed that the number of correct answers was not significantly different for students with a declared specialist interest in neurology/neurosurgery. Although the above factors may contribute, we speculate that this may also arise from student emphasis on better recognised neurological topics, such as stroke and epilepsy, while DCM remains neglected.

The most pressing future work is to determine what constitutes an effective educational intervention. This is aligned with the number one DCM research priority identified by AO Spine RECODE-DCM: raising awareness.¹⁴ To achieve this, it will first be necessary to understand why two medical schools were exceptional. Establishing a focus group with faculty from these schools will help to define how they are delivering more effective teaching. With this understanding, we plan to propose and deliver a pilot intervention at several medical schools and to evaluate outcomes using the survey from this study. It may also be useful to investigate the methods of promotion of previously neglected subjects to the mainstream of undergraduate education. Although there are limitations, migraine is one such example.^41,42 A more ambitious goal will be to engage a larger working group of educators and neurosurgeons to define a national undergraduate neurosurgery curriculum. This could be delivered by a national network of neurosurgeons at individual medical schools using centralised resources.

Conclusion

DCM education in UK medical schools is at best deficient, and at worst, absent. This impedes student confidence, alters perception of the condition, and has evident consequences in delayed DCM diagnosis. Inclusion of DCM on the undergraduate curriculum is more necessary than ever, and students have expressed a strong desire to receive more teaching.

Acknowledgements

This research aligns with the AO Spine RECODE DCM, James Lind Alliance number one research priority of raising awareness of DCM. This was selected by people living and working with DCM. For further information on how this process was conducted, why this question was prioritised and updates on currently aligned global research, please visit aospine.org/recode/raising-awareness.

Disclosure statement

This report is independent research arising from a Clinical Doctoral Research Fellowship supported by the National Institute for Health Research. The views expressed in this publication are those of the authors and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health and Social Care.

Additional information

Funding

MRK gratefully acknowledges support by the Cambridge NIHR Brain Injury MedTech Cooperative and is funded by a NIHR Clinician Scientist Award CS-2015-15-023. BMD is supported by an NIHR Clinical Doctoral Research Fellowship. ODM is supported by an Academic Clinical Fellowship. There was no additional external funding received for this study.