 I'm Marta from MSF Switzerland. I'm medical knowledge management advisor, actually only 50% of my time. So I'm presenting a project I've been working on for the past few months. Thank you for inviting me to speak at this great venue. So what is my project about? It's about mapping medical and paramedical knowledge within MSF, so as to improve processes for knowledge sharing across all levels of the organization, but assessed especially in terms of field needs. But before going further, how do we define medical knowledge management? It can be defined loosely according to four ideas. Idea one, get the right information to the right person at the right time. For example, tell a nurse how to manage an adverse event in a patient following a vaccination. Idea two, capitalize on in-house know-how, i.e. know who knows how to do what within MSF. Idea three, retain knowledge from retiring staff, i.e. learn from experience and expertise. And idea four, invest in knowledge acquisition for our staff, i.e. provide learning opportunities so that staff can continue to progress and face new challenges. Thus it is clear that medical knowledge management has a very broad scope and can have a major impact on the effectiveness and safety of MSF projects. So why do we have an issue with medical knowledge management? Broadly speaking for three reasons. One, improved quality of care has emerged as a major goal for MSF Switzerland. Two, our medical and paramedical staff have great demands in terms of training, learning and development. And three, MSF Switzerland recognizes that a lot remains to be done in order to foster our own knowledge sharing culture. Furthermore, 50% of our staff is directly involved and impacted by medical knowledge management processes. That is about 500 medics and around 2,000 paramedics. Yes, okay, sorry. So presentation's a solution. Right, to respond to this problem of at least perceived suboptimal medical knowledge management, we needed evidence. To gather evidence, we decided to carry out a needs assessment. The goal of phase one of the needs assessment was to understand better current practices and challenges in knowledge management within MSF Switzerland. The goal of phase two was to pinpoint problems and possible solutions through group brainstorming in the field and NHQ. The goal of phase three, currently ongoing, so-called majority judgment, is to evaluate a series of concrete proposals by ranking them according to importance. And what I hear you ask is majority judgment. So majority judgment is a mathematical method that asks voters and in this case, all our medical and paramedical staff to evaluate proposals according to a verbal grid. For example, I think this measure would be highly effective, somewhat effective, not effective to be rejected. Why did we choose majority judgment? Because it is probably the best scientific method for understanding where group consensus actually lies. In other words, it comes closest to representativeness. And I will come back to that at the end of my presentation. Phase four will take the form of an in-depth report for management and an implementation plan. So what have our results shown so far? Yes, phase one has showed above all that quality of care is directly impacted by medical knowledge management. Two, that knowledge management processes need to be better adapted to the field. And three, that group consensus is essential for devising a knowledge management strategy across MSF's one solution. Phase two showed above all that we need greater access to training, learning and development for our medical and paramedical staff, greater coherence for our medical documentation policy, including translation of essential texts into local languages, and greater computer access for our staff. Phase three is showing on the basis of 100 results that we need more knowledge to deal with emerging health problems. We need standard procedures for easier sharing, validation and updating of medical and paramedical knowledge. And we need a simplified and user-friendly IT tool. The cost, this was a very, very rough estimation just on basically staff time spent on the project so far and I think it comes well below 40,000 USD. Risks, whether the risks are inherent to this needs assessment. Risk one, those more interested in knowledge management are more likely to have participated in our surveys, which could theoretically bias towards more qualified staff, but not necessarily. And I should just ask that on our first survey we had respondents one third, local staff one third, expat staff and one third HQ staff. Second risk, asking the entire medical and paramedical staff to evaluate a series of proposals can be perceived as risky because we may not be able to satisfy the demands, they express. Risk three, this needs assessment is being carried out by the medical department, but obviously we are not alone within MSF and whatever we implement needs to fit in with overall MSF, Switzerland plan processes, procedures and budget. And risk four, probably the greatest risk, we can create great tools and processes to respond to needs, but if staff don't use them, they will serve no purpose. But there are also risks associated with doing nothing. For example, suboptimal medical knowledge management will necessarily impact adversely on quality of care. Inadequate knowledge management can lead to redundancy and reinventing the wheel, but also to potentially dangerous gaps in clinical and epidemiological knowledge. Finally, we know that every day we are losing valuable know-how from outgoing staff. Okay, so ethical considerations, I just let you read. Implementation, implementation plan. One, highlight priorities. Two, decide on interventions. Three, budget these interventions. Four, monitor step-by-step implementation. Fifth, evaluate interventions. And to finish, the most complicated slide. A final word on majority judgment. Just to say that we rejected yes or no evaluation because it's a very imprecise tool. You don't see what people really think is a good idea when they vote yes or no. But we could have just taken, since we've asked people, is this a high priority, somewhat a priority, et cetera, we could have looked at, well, which is the proposal that gets the most? This is a high priority, plus and or somewhat of a priority. And calculated according to that. Or we could have looked at the proposals that were least often said to be not a priority or to be rejected. And if you see, if we calculated those different percentages on the results we already have, you come up with rank orderings that are extremely different. Why? Because they don't take into account the other intensities of the votes. And why we use majority judgment in the last column in red is because it precisely weighs the different opinions of these votes. So thank you very much. Thank you very much.