Digital learning, face-to-face learning and climate changeDavid Liam Davies, AbdulAzeez Lawal, E. Orji, Chloe Tytherleigh, Kieran Walsh∗BMA House, Tavistock Square, London WC1H 9JR UK

Debates about digital learning, face-to-face learning and blended learning often focus on their effectiveness inDigital learningachieving a few core educational outcomes. The cost or convenience of using different methods to achieve certainMedical educationoutcomes have increasingly come into the educational framework over the past two decades. However, only rarelydo educators or learners consider the climate footprint of their various activities. This is an importantshortcoming,as all learning activities can contribute to our overall climate footprint. Providers of education should do theirbest to minimise the carbon footprint associated with their learning. But learners also have responsibility toensure that how they access learning is also associated with minimal environmental cost. Both providers andlearners should focus on activities that are likely to have the greatest impact. This is relevant both to face-to-faceeducation and digital learning.Debates about digital learning, face-to-face learning, and blendedE-mail address: kmwalsh@bmj.com (K. Walsh).professional behaviours. However, travel is often required for face-toface learning.3 This can involve driving and/or using public transport,learning often focus on their effectiveness in achieving a few core educational outcomes. These include applied knowledge, problem-solving orwhich can include air travel to events such as international conferences.procedural skills, and professional behaviours. The cost or convenienceTravel to some in-person conferences can produce around 1,000 timesof using different methods to achieve certain outcomes have increasmore metric tonnes of CO2 than a virtual conference, with over 90% ofingly come into the educational framework over the past two decades.1these emissions potentially being attributed to travel.4,5 It is possibleThis is important as it enables learners and providers of education tofor delegates or organisers to calculate the carbon footprint of differentaccess and deliver education that is ideally high quality for the lowestcomponents of face-to-face learning; however, Leddin et al suggest thatpossible cost. However, only rarely do educators or learners consider theproviders should us a simplified model to do this ’using flying distanceclimate footprint of their various activities.2 This is an important shortonly, to estimate travel-related emissions’.6 Using public transport is thecoming, as all learning activities can contribute to our overall climatemost climate-friendly means of travelling. Hotel accommodation is alsofootprint. The ongoing climate crisis means that we should be doingrequired if people are traveling for a meeting that lasts longer than aall that we can – however large or small – to reduce that footprint, esday. This will have its own environmental effects – such as the use ofpecially given the wider global healthcare implications. More extremedetergents for laundry of beddings and towels. Physical events are ofweather, increasingly poorer air quality, and threats to food securityten associated with usage of implements such as single-use plastics andare just a few problems that affect patients. In the following perspecpolystyrene food packages. These are not a necessity, but they are ative paper, we will outline a number of considerations that should becommon accompaniment of face-to-face learning events. This is not ataken into account when linking together learning activities, learningcomprehensive account of the carbon footprint of face-to-face learning,outcomes, and carbon footprint.but it should encapsulate most of the activities that can contribute –namely travel, accommodation and subsistence.7Face-to-face learning is the traditional means of delivering medical education activities, and it has much to commend it. Students andDigital learning can also enable multiple useful learning outcomes. Ittrainees can learn knowledge, skills and behaviours in face-to-face setcan enable learning knowledge and the learning of certain skills and betings and also have the opportunity to network and engage in inforhaviours. Technological advances mean that certain learning outcomesmal learning with educators and peers. The face-to-face environmentthat until recently would have only been possible with face-to-face learncan especially help with learning practical or procedural skills anding (such as communication skills training and improvement) are nowThis article reflects the opinions of the author(s) and should not be taken to represent the policy of the Royal College of Physicians unless specifically stated. ∗ Corresponding author.https://doi.org/10.1016/j.fhj.2024.1001562514-6645/© 2024 The Author(s). Published by Elsevier Ltd on behalf of Royal College of Physicians. This is an open access article under the CC BY license(http://creativecommons.org/licenses/by/4.0/)D.L. Davies, A. Lawal, A.E. Orji et al. Future Healthcare Journal 11 (2024) 100156quite feasible with e-learning. And digital learning is usually more conboth providers and learners share a responsibility that they get the mostout of any form of learning that they take part in. Sometimes this mightvenient than having to travel. Digital learning also has advantages inmean blending face-to-face and digital learning – there is some evidencethat this may help with the transfer of learning to practice.11 It wouldterms of its carbon footprint over face-to-face learning. Digital learningdoes have a carbon footprint, but it is generally less than the carbonalso be good practice for providers of all forms of learning to publish thefootprint involved in travel.8 The carbon footprint associated with digicarbon footprint of their programmes and efforts that they are makingtal learning involves electricity, hardware and software – to create andor have made to reduce this.access learning materials. Electronic devices have varying degrees ofclimate-friendliness. Some devices consume more electricity than othersDeclaration of competing interest– generally, desktop computers set up for advanced gaming or simulation activities consume the most.KW works for BMJ which produces digital learning and clinical deSo how best to balance all of this? There is a clear opportunity to getcision support resources such as BMJ Learning and BMJ Best Practice.the best from digital and face-to-face learning and to minimise carbonfootprint at the same time.Source(s) of supportIt is possible to reduce the amount of face-to-face learning that is required to achieve needed learning outcomes and to replace unnecessaryNone.face-to-face learning with digital learning. One example is that lecturescan be recorded or delivered live online. If travel is necessary, flyingAcknowledgementsshould be avoided apart from when there is no alternative. Public transport should be the preferred means of travel. Physical events providersNone.should take all means possible to reduce their carbon footprint, fromavoiding single-use plastics to providing environmentally friendly mealsEthical approvaland staying in accommodation close to the conference.If digital learning is to be employed, learners should utilise devicesNot applicable.that are the most environmentally friendly. These might be devices thatReferencessatisfy energy efficiency standards and that use less power. It might alsobe preferable to use devices with a longer life or constructed with en1. Sandars J, Walsh K. A consumer guide to the world of elearning. BMJ Career Focus.vironmentally friendly material or that can be recycled. Another con2005;330:96–97.sideration is the use of artificial intelligence in digital learning. Many2. Gherhes V, Stoian CE, Farcas MA, Stanici M. E-learning vs. face-to-face learning: analyzing students’ preferences and behaviors. Sustainability. 2021;13:4381.artificial intelligence (AI) programmes are based on machine learning,3. Walsh K. E-learning in medical education: the potential environmental impact. Educand this can have a significant environmental impact.9 This is becausePrim Care. 2018;29(2):104–106.of the energy requirement associated with this form of computing. There4. McClintic SM, Stashevsky AG. Assessing strategies to reduce the carbon footprint ofare ways to reduce this carbon footprint, such as by reducing the computhe annual meeting of the American academy of ophthalmology. JAMA Ophthalmol.2023;141(9):862–869. doi:10.1001/jamaophthalmol.2023.3516.tational demands of machine learning, using renewable energy sources5. Gattrell WT, Barraux A, Comley S, Whaley M, Lander N. The carbon costsand/or thinking through the lifecycle of machine learning programmesof in-person versus virtual medical conferences for the pharmaceutical indusfrom deployment to maintenance and eventually to replacement. AI protry: lessons from the coronavirus pandemic. Pharmaceut Med. 2022;36(2):131–142.doi:10.1007/s40290-022-00421-3.grammes can enable more personalised learning, but it is a matter of bal6. Leddin D, Galts C, McRobert E, Igoe J, Singh H, Sinclair P. The carbon cost ofancing this against increased carbon footprint. There is some evidencetravel to a medical conference: modelling the annual meeting of the Canadianthat younger doctors have a greater preference for e-learning than olderassociation of gastroenterology. J Canadian Assoc Gastroenterol. 2022;5(2):52–58.doi:10.1093/jcag/gwab021.
x