Built for the trainee without a skills lab.
SimulationM8 gathers the published evidence on simulation in surgical and dental training, most of it scattered across specialty journals, into one place. It covers every cost band and highlights what is genuinely frugal.
Share the build, not just the finding.
Simulation is well evidenced. It improves surgical practice, and safer surgical practice saves lives.
What is not shared well is how to build the thing. Simulation technology is advancing quickly and a great deal of it is expensive, which is exactly why frugal innovation matters, and why it matters most when it is shared rather than kept behind a paywall.
SimulationM8 is an open library built so that this can be collaborated on internationally. Realistically, it will not close the gap between a well funded skills lab and a hospital without one. But we aim to make the basic models accessible to the people who need them, with the cost, the materials and the strength of the evidence stated plainly on every entry. That is worth doing on its own terms.
If you have built something, taught with something, or have an idea for how this should work, we want it in here. The library gets better by being added to.
Join the projectWho is behind this
SimulationM8 was created by a UK surgical registrar who wanted to improve how simulation is taught.
It started as a search rather than a project. Building SkillsM8 meant looking for simulation models to teach procedures with, which meant reading what other people had already built. That turned out to be the hard part. The models are out there and many of them are good, but they sit in specialty journals, get described halfway through a paper about something else, and are often behind a paywall. There was nowhere to browse by specialty and see what already existed for the thing you were about to teach.
That is the gap this fills. Someone preparing a teaching session should be able to open one page, pick a specialty, and see what has been built for it, what it costs, what it is made from and how far the evidence goes.
What we are aiming at
Most simulation innovation is never published. Somebody builds a trainer for their department, teaches with it for years, and it never leaves the building. A library of published papers catches only a fraction of what exists, which is why this one accepts models with no paper behind them and labels them as unpublished rather than turning them away.
The aim goes further than cataloguing. We want a video library showing how models are actually built, because a methods section and watching somebody make the thing are not the same, and collaboration is easier when you can see the build.
Neither has happened yet. Today the library holds 305 published models, no unpublished submissions and no videos. We would rather say that than imply otherwise.
If you have built something, or taught with something, it belongs here, whether or not it was ever written up.
Submit a modelQuestions, corrections and takedown requests: simulationm8@gmail.com
The global surgery lens
The Lancet Commission on Global Surgery estimated that 5 billion people lack access to safe, timely and affordable surgical and anaesthesia care, and set a benchmark of 20 surgeons, anaesthetists and obstetricians per 100,000 population. Median density is 0.7 per 100,000 in low income countries and 5.5 in lower middle income countries. Workforce training was named as one of the central levers to close that gap.
Meara JG, Leather AJM, Hagander L, et al. Global Surgery 2030: evidence and solutions for achieving health, welfare, and economic development. The Lancet 2015;386(9993):569–624. Read the Commission Workforce densities are from Holmer H, Lantz A, Kunjumen T, et al. Global distribution of surgeons, anaesthesiologists, and obstetricians. The Lancet Global Health 2015;3(Suppl 2):S9–11. Read the workforce data
A prospective cohort study published in The Lancet Global Health in 2026 followed 52,187 cholecystectomies across 1,218 hospitals in 109 countries. Country income level was not independently associated with complication rates. The presence of local simulation based training facilities was, with an adjusted odds ratio of 0.78. The benefit was largest in hospitals with limited infrastructure and an inexperienced workforce, where 21 patients needed to be treated to prevent one procedure specific complication. Simulation is not a training nicety in that context. It is patient safety infrastructure that most of the world cannot currently afford.
National Surgical, Obstetric and Anaesthesia Plans are the policy mechanism most low and middle income governments now use to close this gap, built around infrastructure, workforce, service delivery, information management and financing. Simulation based capacity building, frugal and commercial alike, sits directly inside the workforce and training pillar of that plan, which is why SimulationM8 catalogues evidence across the full cost range rather than only the cheapest end of it.
“Safe implementation of minimally invasive surgery varies widely worldwide but is not defined by national income level; differences in outcomes reflect the ability of health systems to adopt and safely deploy new surgical techniques. We identified for the first time that the presence of local simulation-based training facilities is independently associated with improved patient outcomes.”
How we grade evidence
Most published simulator studies still use outdated validity language. We try not to.
Surgical education research has largely moved on from talking about face, content and construct validity as separate, fixed properties of a model. The current standard, Messick’s unified framework, treats validity as a single concept built from 5 sources of evidence: content, response process, internal structure, relationships to other variables, and consequences of use. A 2018 systematic review found that only 6.6% of published surgical simulation validity studies actually used this framework, most still relying on the older, discouraged terminology.
Borgersen NJ, Naur TMH, Sørensen SMD, Bjerrum F, Konge L, Subhi Y, Thomsen ASS. Gathering Validity Evidence for Surgical Simulation: A Systematic Review. Annals of Surgery 2018;267(6):1063–1068. 498 studies, 18,312 participants. Read it
Every model in the SimulationM8 library is labelled with what evidence actually exists for it, not what the marketing implies. A model with a single trainee satisfaction survey is labelled as such. A model with a published construct validity study distinguishing novices from experts is labelled as such. We would rather under claim than over claim.
In practice, each appraised paper on the site carries 4 things. The closest fitting CASP checklist for its actual study design, since most of this literature is a before-after study or a technical note rather than a randomised trial, so the checklist is applied where it genuinely fits and the mismatch is stated where it does not. An Oxford Centre for Evidence-Based Medicine style level of evidence, from Level 5 for a build method with no outcome data through to Level 3 or better for a comparative or construct validity study. A Kirkpatrick level, since a paper can be methodologically sound but still only measure reaction or learning rather than behaviour on a real patient. And a two line, plain language summary next to a link to the actual source, so a trainee never has to take our word for it.
There is a fifth thing, and it is the one most libraries leave out. Every appraisal states what it was based on. A library of this size cannot be built by reading every paper end to end, so most entries are screened and appraised from the abstract. That is enough to decide whether a model belongs here and to describe what was measured, but it is a genuine limitation and it appears as one on the card rather than being quietly omitted. Where we have read the full text, the card says that instead. Papers found but not yet appraised are shown as pending or in progress rather than left out, so the queue itself is visible.
Simulation based tests with solid evidence of validity and defensible pass or fail scores are a prerequisite for mastery learning. It is well past time surgical education replaced the outdated language of face and construct validity with a contemporary, unified framework.
Nobody should be excluded by what a model is made of
Gelatin based and animal tissue models exclude trainees on religious, ethical and allergy grounds. Where a listed model has a synthetic or plant derived equivalent, the library says so.
Halal, kosher, latex free and vegan safe variants
Gelatin based and animal tissue models exclude trainees on religious, ethical and allergy grounds. Where possible, SimulationM8 defaults to platinum cure silicone or plant derived materials such as konjac glucomannan gel, both of which sidestep the issue by construction rather than needing separate certification. Where an animal derived material is genuinely the best option for a given skill, it will be labelled clearly and a synthetic alternative offered alongside it.
Submit a model
If you have built, taught with, or published a simulation model that is not yet in the library, we want to hear about it, whatever it costs. Trainee submitted models are especially valuable from settings that the published literature under represents, and genuinely frugal submissions will carry the highlighted tag once appraised.
To be listed, a submission needs a materials list, an approximate cost, a short build method, and whichever evidence exists for it, even if that evidence is informal trainee feedback rather than a published study. We will label it honestly either way.
If effort only goes into resources that need serious funding, the result is an educational divide. Low cost interventions should be shared and properly evaluated, not kept inside the institution that built them.