An investigation into the contextual admissions information available at UK medical schools’ websites: what are the opportunities for enhancement?

ABSTRACT We audited the websites of all 57 undergraduate medicine course websites at UK universities for 2020 entry, looking in particular for the clarity of information regarding contextual admissions (CA). 49 programmes featured 47 distinct CA policies, using 26 different contextual markers, 8 programmes had no CA policy. Half (51%) of these courses with CA described clearly how they used CA, some allowed candidates to evaluate their own eligibility. However, the burden currently falls on individual applicants to understand the CA system at each medical school. Medical schools should present CA information clearly and accessibly so that applicants can make informed choices where to apply, having a one-stop for accessing all CA information across medical schools would also lessen the research burden on applicants. This could increase applications from a more diverse range of applicants. Other degree courses might use this study as a springboard to reflect on their CA policies.


Introduction
Studying medicine has long been competitive and thus selective across the globe, with applications to admission rates being 17/1 in the US, 12-20/1 in Australia/New Zealand, 10-20/1 in Canada and ∼11/1 in the UK (Association of American Medical Colleges 2020; Judy 2010; Shemmassian Academic Consulting 2019; Medical Schools Council Selection Alliance 2014).However, despite this competition, there are still under-represented groups in application and admission to medical schools.The profession has successfully embraced female applicants but has not been able to achieve similar success in terms of increasing or widening access to applicants from less affluent social backgrounds or those otherwise disadvantaged (Medical Schools Council 2018c;Kirch 2018).For example, Black or African Americans only make up 7.1% of US medical undergraduates compared with ∼13.4% of the population and the under-representation of students from rural backgrounds in Australian medical schools (Association of American Medical Colleges 2019; United States Census Bureau 2020; Medical Deans Australia and New Zealand 2017).In the UK, 80% of medicine applicants came from 20% of schools and colleges, and there is a diversity gap between medicine when compared to general entrants to UK higher education (Garrud 2014;Medical Schools Council 2018c).
There are thus concerns that medical school admission 'effectively maintains' social inequalities by being more easily accessible to already privileged strata of society (Lucas 2001).Drawing on a capability framework, this misses out those with potential and ability to become doctors who may have fewer credentials than their more advantaged peers (Nussbaum 2003).
Inclusiveness and representativeness then are key principles in thinking about medical schools and the future medical workforce (Young et al. 2012).Increasing diversity within medical schools and ultimately the health care workforce will ensure the workforce is representative of the society it serves, which can enhance empathy and understanding (BMA Equal Opportunities Committee 2009).It may also ensure geographic coverage with some indication that those originally from lower socio-economic and rural communities are more likely to return to serve these often harder-to-recruitfor communities (Puddey, Playford, and Mercer 2017;Steven et al. 2016;Laven and Wilkinson 2003).Minority patients tend both to seek care from minority physicians and to report greater satisfaction when they receive care from a minority physician (Young et al. 2012).Diversity also benefits the learning experience and enhances professionalism at medical schools (Young et al. 2012).
Over the years, some have voiced concerns in the UK and elsewhere that broadening the profile of the professions and changing admissions standards would lead to a lowering of academic standards and a conflict with meritocracy.Indeed a former president of Princeton University and colleagues are one of many who have tackled the issue head-on in their book 'Equity and Excellence' where they argue that contextualising qualification by opportunities makes sense for educators as well as the students.In the UK, the University of Edinburgh practices an evidence-based lowering of offers based on 'minimum preparedness': The idea here is that it would be unhelpful for both, students and the university, if students were admitted who were unable to succeed in higher education.However, setting exceptionally high bars of, e.g. three A*** might be a helpful way of 'deselcting' among many talented applicants but the bar is so high that many who would be able to succeeed are deselected.This is where contextualising prior opportunities allows universities to provide a place to those who have the potential to thrive in a course even though their attainment to date might not be as stellar as some of their peers.
Within the UK, outreach initiatives to 'widen access' tend to take the form of engaging school-aged children with the opportunities of higher education and supporting them through attainment rising, information giving, and activities such as university taster days to consider higher education and sometimes specifically medicine (Office for Students 2019a).Widening access medicine courses are specifically designed to increase diversity in the medical profession and are targeted at disadvantaged students who would have been unlikely to gain a place on a standard entry programme (Medical Schools Council 2018a).These might be run by consortia of universities like the Realising Opportunities Scheme or driven by a charity like the Sutton Trust, which runs a pathway to medicine programme.New policy initiatives like the OfS-commissioned body Transforming Access and Student Outcomes (TASO) are seeking to enhance the evidence-base of 'what works' in such initiatives.This is because, despite years and indeed decades of initiatives in the widening participation field, there remain gaps in prior attainment and opportunities to progress to higher education and an under-theorization of the field (Mountford-Zimdars and Harrison 2017).Making Theory of change approaches an integral part of research designs and evaluation is a promising shift in how initiatives are framed (Dent, Mountford-Zimdars, and Burke 2022).
Currently, there remain strong patterns whereby children from more advantaged and professional families, through their economic, social and cultural capital and parental support are most likely to progress into higher education and into medicine in particular.Because we do not have great measures of this individual advantage, currently a range of proxy measures are used to gauge advantage and disadvantage.This shows a strong patterns of attainment and progression linked to individual-level factors such as free-school meal status (a measure of poverty), school-level measures such as average attainment (a measure of school quality), fee status of the school, i.e. private or state school (a proxy measure of advantage) and area levels of advantage and disadvantage.A popular measure here is the Index of Multiple deprivation, which is a composite measure including income, employment, health, education, crime, housing and living environment.It is worth noting that, methodologically, using a postcode and an associated index of multiple deprivation is unsatisfying as one is basically attempting to draw an individual-level inference (this person is disadvantaged) from an aggregate measure.This is problematic because we do not know whether this particular person is actually disadvantaged although they live in a community were there are more disadvantaged individuals.The same logical problem holds true for schooling measures.However, despite these shortcomings, currently, these measures are quite universally used in widening participation work because of their relative ease of operationalisation.
Fixing the link between social origin and attainment is not something medical outreach can deliver on its own.However, medical schools can attract greater numbers of disadvantaged applicants and make a difference to who is admitted.For example, US medical schools might use elements of holistic admissions (Bastedo et al. 2018); whilst, in the past, the Netherlands employed a weighted lottery for selection to university (Stegers-Jager 2018).For the UK case, a system of contextual admissions (CA) enables medical schools to asses an applicant's attainment and potential within the context of their individual circumstances, such as their school and neighbourhood (Office for Students 2019a; Mountford-Zimdars and Moore 2020, Mountford-Zimdars et al. 2021).As a result of contextualisation, applicants for medicine might, for example, receive 'reduced offers', meaning that they will be able to attend medical school with lower prior attainment than their more socially advantaged peers (see Figure 1 for more details).The greatest single barrier to successfully applying to medicine for disadvantaged students continues to be their lower attainment in education (UK Government 2022), this is true across OECD (Organisation for Economic Co-operation and Development) countries (OECD 2020).
Contextual admissions is underpinned by research showing that students from traditionally underrepresented groups can excel in medicine: for example, students admitted with lower school attainment than have traditionally been accepted have succeeded on an extended medicine course (Garlick and Brown 2008), whilst those who attend PERSPECTIVES: POLICY AND PRACTICE IN HIGHER EDUCATION state schools are twice as likely to finish in the highest 10% of students at the end of their degree compared to those from fee-paying schools (Kumwenda et al. 2017).
Multiple professional bodies have suggested fair and transparent admissions policies for applicants will enhance diversity (Office for Students 2019a; Centre for Social Mobility 2018; Medical Schools Council Selection Alliance 2018; Supporting Professionalism in Admissions 2010).However, there is no published research into the clarity of contextual admissions information at different UK medical school websites, although we know from research focused on entire universities that these vary greatly in their motivation, understanding and use of contextual admissions (Centre for Social Mobility 2018; Garrud 2014).But is it possible and reasonably straight forward for prospective medical students to understand what type of contextualisation and reduced offers they might be entitled to and realistically assess their chances of obtaining an offer from different medical schools?Auditing current medical school website information is a necessary step for understanding whether diverse candidates can currently make informed choices as to their chances of success in applying to different medical schools and is, as such, important to ultimately improve diversity in medicine.
The present study then investigates the clarity and accessibility of contextual admissions information presented on UK undergraduate medicine course websites from the view of prospective students.

Methodology
Identifying medical courses UK medicine course pages were identified using the Medical Schools Council (MSC) entry requirements website (Medical Schools Council 2018b).Inclusion criteria were: undergraduate medicine course, 2020 entry, and run by a UK state-funded university (excluding one private institution).The entry route was categorised as standard undergraduate (UG -67% of courses found) or widening access (WA -33% of courses).
To replicate an applicant's experience, the medicine course page for each institution was found by using the Universities and Colleges Admissions Service (UCAS) search engine (UCAS 2020) and following the link in the 'course contact details' to the university website and medicine course page.Where a medicine course is jointly awarded by two institutions, the joint medical school website was used.Information was extracted from this webpage and hyperlinked sites.These would be easy for the average applicant to find and required no extra searches into contextual admissions.The data was collected by one researcher in April 2020.

Programme types
Thirty-nine UK state-funded universities offer at least one medicine course for 2020 (possibilities being undergraduate, postgraduate and widening access programmes).In total, data was collected for 57 undergraduate programmes for 2020 entry.Two-thirds (38) of the programmes were standard 5-or 6-year undergraduate medicine degrees (UG), the other third ( 19) were classed as widening access programmes (WA).This included 16 courses with a gateway year, 2 courses with a preliminary year and an extended medicine programme (see Figure 1).All but one institution that offered an WA course also offered a standard UG course.

Data collection
Information on the following topics was extracted: how applicants would calculate their eligibility (including the information used to assess this); the allowances made for disadvantaged applicants (using the Medical School Council's framework for medical schools shown in Figure 1 (Medical Schools Council Selection Alliance 2018)); the extra work required by disadvantaged applicants, e.g.eligibility forms; the clarity of the above information and how accessible it was.

Contextual data markers
Data on contextual admissions was collected based on the assessment of the websites as opposed to predefined measures.Contextual markers/measures can be split into three broad groups; area level markers use tools that assess deprivation on a postcode/zip code basis (e.g.POLAR in the UK looks at neighbourhood rates of participation in higher education (Office for Students 2019b)); school level markers (e.g.exam performance compared to the national average); individual-level markers (e.g.time spent in state care).Participation in university outreach schemes is also used, as participants have already met predetermined criteria to be eligible for the programme.
For example, the University of Exeter states on their medicine application website When we review an application, we take into consideration the context in which applicants have achieved their academic qualifications.If candidates meet certain eligibility criteria, we may make an offer which is lower than our typical entry requirements.This is called a contextual offer.(University of Exeter 2022) This is then explained in detail that meeting one of a range of criteria would entitle an applicant to a lower offer, the criteria are listed as postcode data, school data, care experience and participation in an Exeter outreach scheme (University of Exeter 2022).This example would thus be coded as the University taking into account postcode/area, school characteristics, care leaver status and a particular outreach scheme.Because these criteria are formulaic and do not involve discretion, one researcher created the coding from the institutional websites.

Research ethics
This study was granted ethical approval by the University of Exeter's College of Medicine and Health Research Ethics Committee.

Use of diversification policies in admissions
Eight medicine courses had no online information on how they considered diversity and the previous context in their admissions process, this accounted for 13% (5) of UG medicine programmes and 16% (3) of WA courses.

Accessibility of information
Over half of courses have easy-to-access, applicantfacing information on contextual admissions (CA), with 60% (19 UG, 15 WA) of all courses having CA information on the main course page or one 'click' away via a hyperlink.One quarter (13 UG, 1 WA) had this information on a webpage that was two or three 'clicks' away from the main medicine course page.In making this assessment, we took into account explicit mentioning of the term contextual admissions as well as mentioning of specific criteria that we know are indicative of contextual admissions, such as 'area level information'.
Three-quarters of all courses (28 UG, 15 WA) displayed all their CA information in one location, likely making it easier for applicants to find the sections that may be relevant to them.

Assessing eligibilitytools for applicants
Three-quarters of webpages included tools for disadvantaged applicants to establish their own eligibility for concessions in the admissions process, for example, an embedded eligibility calculator (University of Birmingham 2020).The underlying metrics institutions use to design such a calculator include attending a school in the bottom 40% nationally for GCSE exam performance or checking how a postcode links to POLAR (participation in the local areas) quintiles as a measure of area disadvantage (Aston University 2020; University of Exeter 2020).Of the 44 courses (29 UG, 15 WA) that we deemed could have included such a tool to allow applicants to assess their eligibility, 57% (15 UG, 10 WA) included tools for all possible criteria, 18% (6 UG, 2 WA) included tools for some criteria and 25% (8 UG, 3 WA) included no tools.

Assessing eligibilitymarkers of disadvantage
Not all courses included tools for applicants to assess their eligibility, making it more challenging, or impossible, for applicants to correctly assess their eligibility and potential for success in an application.For example, applicants would have to research the CA markers used by the institution and how they sit in relation to them.Of the 43 courses that stated their CA measures, 72% (22 UG, 9 WA) published their cutoffs used, whilst 14% (2 UG, 4 WA) only included some of the cut-offs, thus making a fully informed judgement for applicants impossible.
Overall, we found 26 different markers of disadvantage were used in 42 different combinations across all medicine courses.The measures used are detailed in Tables 1 and 2.Even when courses appeared to use the same measures, there was great variation in the definitions of eligibility.Of the 12 courses using free school meals (government-funded meals for those whose parents/carers receive government benefits) as a measure, there were 7 different definitions of what would qualify for the contextual marker according to how recently and how long for the student had received them (University of Aberdeen 2020; University of Leeds 2020; Lancaster University Medical School 2020).Similar disparities in cut-offs were found in 9 other measures.

How contextual data is used
In standard UG courses a reduction in entry grade requirement (25 courses) and increased initial consideration or invitation to interview for eligible applicants (19 courses) were the most frequent methods in which disadvantaged applicants were given extra consideration during the selection process (see Figure 1 for all possible methods).Fifteen of 33 UG courses with built-in considerations for disadvantaged applicants used contextual data once in the selection process.A further 15 courses used contextual data at more than one point, with four-fifths of these combining increased initial consideration/invitation to interview together with a reduction in entry grade requirement.Almost 10% (3) of UG courses with CA policy were unclear on the extra considerations made for disadvantaged applicants.
The majority of WA courses (16 of 19) used contextual data in the eligibility criteria for applicants applying to the programme.

Extra steps for disadvantaged candidates
Twenty-five courses (17 UG, 5 WA) required disadvantaged candidates to undertake additional steps before they were eligible to receive the consideration, or the reduced attainment offer.Of these, almost three-quarters explicitly stated an applicant must fill out an additional questionnaire or submit evidence.A further one-quarter stated they required a candidate to make the course at that university their first choice or sit an online module to receive the reduced offer.

Clarity of information
Almost half of courses (17 UG, 11 WA) did not give applicants clear and understandable information about how their contextual admissions policy worked in practice and/or did not allow prospective students to assess their own eligibility where possible.Figure 2 details the different methods in which this occurred.At every stage of the information-gathering process, there were websites that failed to give clear    information to prospective students across all the criteria reviewed in this study.

Discussion
In the context of widening participation and seeking to diversify the medical workforce, this study investigated how clear and accessible the applicant-facing contextual admissions (CA) information was on UK public medical school websites.The premise is that easily accessible information enables disadvantaged applicants to make informed choices about their chances of success in applying to different medical schools.Overall, our findings show that applicants for UK medicine courses need to work through a wide range of contextual markers and combinations of contextual information.In short, applicants cannot come to an easy assessment of their overall eligibility and entitlement to CA and often have to become experts in CA and measures of disadvantage and their own background in order to work out their entitlement.This challenge in accessing information is a barrier to increasing diversity in medicine in itself.Also, CA measures do not consider international applicants and their contexts.
In detail, findings show that eight undergraduate medicine courses lacked any online information on CA.Just over half (51%) of courses gave applicants clear information on how they used CA and, where possible, allowed applicants to self-assess their own eligibility for widening participation concessions at that institution.61% of medical courses provided easy-to-find CA information in one location, either on the main medicine course page or one 'click' away via a hyperlink.
This study identified 47 different ways that UK medical courses allocate and use contextual markers, with 42 different combinations.This, in particular, resonates with previous reports identifying the variation between CA policies across HE institutes (Centre for Social Mobility 2018; Garrud 2014).Cleland et al. identified that applicants from underrepresented groups could be disadvantaged by the use of different contextual markers in different ways and the lack of clarity surrounding their use (Cleland et al. 2015).With such great variety in the use of CA, institutions have a responsibility to prospective students to provide clear and accessible information.
In addition, we also identified 26 different contextual markers used for undergraduate medicine courses for 2020 entry.Previous research has looked at the reliability and accuracy of currently used contextual measures (Boliver et al. 2017;Gorard et al. 2019).The Fair Education Alliance highlighted individuallevel measures as the most important type of contextual data (Fair Education Alliance 2018).The current study finds that one-fifth of medicine courses did not use any individual-level data.Furthermore, free school meals and the UCAS MEM-quintile (multiple equality measure quintile) have both been cited as alternative and more accurate measures (Fair Education Alliance 2018).However, we found that only 21% of courses consider free school meals and only one course uses the UCAS MEM data.
The above findings are in direct contrast to multiple recommendations, stating the need for fair and transparent admissions systems (Schwartz 2004; Office for Students 2019a; Centre for Social Mobility 2018; Medical Schools Council Selection Alliance 2018; Supporting Professionalism in Admissions 2010).Online course information is essential for many prospective students (Simões and Soares 2010).Therefore, this study offers a unique insight and adds to the existing literature on applicant-facing medical school information, which has looked at the widening access discourse, pictures in prospectuses and on websites (Alexander et al. 2017;Macarthur, Eaton, and Mattick 2019;Chowdhury, Chowdhury, and Sandars 2009).We add that institutions need to ensure they provide clear and transparent CA information.A lack of transparency is in itself a barrier to increasing diversity, as potentially eligible applicants may be unaware of how they can benefit from contextual admissions policies.Clarity and transparency will likely allow students to easily assess if they would be flagged as an underrepresented applicant and how they would be treated in the selection process, thus allowing them to make informed decisions and apply smartly.

The wider context
This study identified a number of noteworthy points when looking at the current contextual admission (CA) system in the UK.Firstly, applicants from underrepresented backgrounds applying to medicine are different to the majority of disadvantaged students, as they are likely to be achieving significantly higher grades.It may be that outreach and CA need to be different for this group of students, as there remains a disparity between disadvantaged groups in medicine and higher education as a whole (Medical Schools Council 2018c).
Secondly, the increasing diversity medicine courses appear to be quite limited in their reach, with the nine of these programmes accounting for less than 5% of all UK medical student entrants in 2017 (General Medical Council 2020).The current study highlights there are now 19 such courses available in 2020, but it is likely that the proportion of overall medical students they enrol will continue to be small when compared to standard UG programmes.This raises the possibility that medical schools could use their widening access programmes to in effect let their much larger standard UG programme 'off the hook' when it comes to increasing diversity and not make changes to their admissions practice for those standard courses.This is beyond the scope of the present article but would be of interest for further investigation.
Thirdly, we question if contextual offers are being used as a recruitment tool rather than an initiative to increase diversity.This study identified six medical courses that only provided candidates with a reduced attainment offer if they place the university as their first choice at the offer acceptance stage, which may be limiting prospective students' choices (Office for Students 2019a).We are concerned that this practice is ethically questionable and may contradict the purpose of contextual admissions with the aim of such a policy primarily considering institutional needs rather than the needs of the applicants.
Finally, the study revealed 26 different contextual markers used, and there are documented concerns about the validity of some, especially the more areabased rather than individual-based measures (Fair Education Alliance 2018).The quality of measures used in policy making and regulation need to be readdressed and based on firmer evidence of what are valid and reliable measures of disadvantage.Simultaneously, being serious about CA also means allowing qualitative information about particular experiences not captured by quantitative measures to be considered, this might, for example, include bereavement.
We believe that CA has a role in addressing the diversity problem in medicine in the UK and elsewhere in the world but it is not the only answer.It will work best when used alongside other initiatives that support increasing diversity, such as outreach schemes increasing aspirations amongst under-represented groups and recognition of and support with the challenges disadvantaged students face when at medical school.

Limitations
This study did not look at other sources of information that applicants have access to, such as online prospectuses, higher education fairs and information, advice and guidance from parents, professionals, peers and university students.We appreciate that contextual admissions is only one factor in the journey to becoming a medical student.

Recommendations
As a result of our research, we would recommend that medicine course pages should: (1) Clearly state the changes in the admissions policy for eligible applicants from disadvantaged backgrounds (e.g.increased initial consideration or reduced attainment offers).
(2) Explain to prospective students who will be eligible for contextual admissions and, where possible, state the metric cut-offs used (e.g.living in a POLAR quintile 1 area).
(3) Allow applicants to easily assess their own eligibility for contextual admissions allowances through built-in eligibility calculators.( 4) Ensure all the above information is clear and provide contact details in case clarification is needed.(5) Not link the availability of contextual offers to applicants accepting the institution as a first choice.(6) Consider coming to an evidence-based agreement across medicine courses which measures to use to enable greater simplicity of the CA process.

Conclusion
With medicine continuing to be highly selective across the globe, contextual admission policies may be an important tool to help increase diversity.This study highlights, however, that despite repeated calls for fair and transparent admissions policies almost half of all UK undergraduate medicine courses lack clear information on policies for disadvantaged applicants, including not allowing candidates to assess their eligibility at that institution.UK medical schools should be mindful of the information they present to prospective disadvantaged students and whether it allows them to make an informed decision when choosing where to apply.Future research should explore: . CA information available to candidates in other forms, such as prospectuses and information, advice and guidance. .The effect of clear and transparent contextual admissions information on an applicant's decision when choosing which medicine courses to apply to. .Applicant perceptions of tools that allow them to assess their own eligibility for CA, such as online offer calculators. .The evidence base of using different measures for assessing eligibility for CA with a view of proposing a single approach across medical schools.

Patient and public involvement statement
It was not appropriate or possible to involve patients or the public in the design, or conduct, or reporting, or dissemination plans of our research.However, we have undertaken dissemination work with the Association for Medical Education to reach those responsible for admission to medical education across the UK.We have worked with several admissions departments individually to provide them with feedback on their individual contextual admissions policy and suggested enhancements.

Figure 1 .
Figure 1.Methods of using contextual data within the selection process.Adapted from Indicators of good practice in contextual admissions (Medical Schools Council Selection Alliance 2018).
three most frequently used measures have been highlighted in bold font.a Participation of local areas [POLAR] (Office for Students 2019b).b Index of Multiple Deprivation [IMD] (Ministry of Housing C&LG 2019), Scottish IMD [SIMD] (Scottish Government 2020), Welsh IMD [WIMD] (Stats Wales 2019), Northern Ireland IMD [NIMD] (Northern Ireland Statistics and Research Agency 2017).c ACORN is a segmentation tool for postcodes by occupant characteristics (ACORN 2020).dPercentage of pupils entitled to free school meals at school/college, school/college with few applicants to Oxbridge.e This varied by institution but included applicants from Roma, Gypsy and travelling communities, children from military families, transgender individuals and mature students.

Figure 2 .
Figure 2. Chart showing the percentage of all medicine courses that failed to give applicants clear information on contextual admissions and how.

Table 1 .
A breakdown of the different types of measures used for flagging widening participation candidates within undergraduate (UG) and widening participation (WA) courses at all UK public medical schools.a a Five UG courses and three ID courses had no visible CA policy and therefore did 'not use' any measures.

Table 2 .
A breakdown of the percentage of all UK state-funded medical schools stating they use distinct markers of disadvantage to assess applicants across undergraduate (UG) and Widening Access (WA) courses.