 So preventing, detecting and responding are key WHO strategies to fight the issue of substandard and falsified medicines. In this talk, I'm going to focus on detecting substandard and falsified medicines, which is of interest in our group, using especially portable screening devices. So in many countries, medicines inspectors, police, and customs officers only rely on their own eyes to identify it, substandard and falsified medicines. But as Paul illustrated on his slides, it's not always very easy, as it is on this picture, where there is a misspelling of a word on the package. And so it's harder and harder to identify, especially, good copies of medicines. They are getting very sophisticated copies and very difficult to identify. And also, substandard medicines, by definition, are not easy to identify. A quick look online, and you would find that there are dozens and dozens of portable screening devices that could be potentially useful for the detection of SF medicines. These ones are of many shapes, sizes, portability, technologies, and costs. Going from lab on the chip, it's like a paper card size card with chemicals on it that will react with some medicines to identify active ingredients. It goes to very fancy gun-shaped-style spectral photometers like this one that can be very costly to more bench-top-sized devices. But what is the scientific evidence regarding these devices, how they perform to identify SF medicines? We performed a systematic review of the literature, scientific literature, a few years ago. I won't go into too much details, but what we found is that there were 41 devices, mainly spectral photometers, like two-third of them were spectral photometers. But I'll come back to that later for those who are not familiar with what a spectral photometer is. What is published in the BMG Global Health. So if you're interested, please have a look. One of the main findings we found is that there is at least quite a lot of gaps of evidence regarding many different aspects of these screening devices. And I'll come back to that also later. So what we did is that two years ago, we did a study, a pluriddisciplinary study, multi-phase study in Laos, not only in Laos. It started with an inception phase where we selected devices to test. These devices were tested in a laboratory in Georgia Tech to test their performance in a laboratory setting. And the six most field-suitable devices were then shipped to Laos in our laboratory for a field-phase evaluation of their utility, usability, and user satisfaction. And this was done as in the picture here in what we call a simulated pharmacy. That's a fake pharmacy in our laboratory. But we invited inspectors from the Laos government to test the devices. Joel Lubel, who is a next speaker, will discuss the cost-effectiveness analysis of this work. And at the end of that work, we also organized stakeholder meetings where we discussed the findings of our study. We had hands-on sessions with the devices. And these stakeholder meetings was with many regulators from Southeast Asia and also other institutions like the USP, the UNDP, the WHO to discuss the main bias challenges and recommendations for implementation of screening devices. So in this talk, I'm going to utilize that research and also the available literature to discuss the main challenges for policymakers, NGOs, wholesalers, or hospital pharmacists considering implementing screening devices in post-market surveillance. So sorry, this work is published in open access online. So you're very welcome to have a look. There's a lot of results. I'm not going to throw all of the results in 10 minutes. That's just impossible. But what I wanted to highlight is that one of the main challenges for policymakers is choosing the right device for their objective, for their setting. And so in one of the paper I mentioned, we have this illustration, this diagram. It shows different aspects of how to choose and implement devices in big fonts are the main aspects. And then they are described in subdivisions. But I'm not going to go through all of them. I selected a few. What I wanted to mention first is that the regulatory framework where the device would be implemented is quite important. Indeed, a fast identification of SF medicines will have very limited impact. If quarantine or recall of the suspicious batches cannot be acted upon quickly, if there is no legal or regulatory considerations of screening devices results, which happens. So next important aspect for choosing the right device is choosing the right device for the right objective. And you can see the prevalence of SF is one key thing. Indeed, the devices we tested in our study were generally very accurate to identify medicines or fake medicines containing no or wrong active ingredients. But none of them, when used out of the box and as many spectrophotometers were able to accurately identify substandard medicines containing incorrect active ingredient amounts. Also, the ingredients and the formulation of the medicines you want to target for this testing are very important because we know that some active ingredients or even excipients can be problematic with some technologies. For example, ingredients which are fluorescent will interfere with the spectrophotometers using Raman technology. Cost and logistics. UL will discuss cost and cost effectiveness. So I will leave that for now. But logistics, I wanted to mention a few important things to think about. Procurement of consumables is not always easy in some countries. Device maintenance, there was quoted by regulators as one of the main buyer because of the lack of in-country customer services. Device implementation is quite important. I will mention the standard operating procedures in the absence of manufacturers of the screening devices protocols. There will need to be a standard operating procedures in place. I think of one thing that was a barrier in our difficulty in our work was that there was no guidance on how many tests to perform on the same sample with the screening device and how to interpret when there was discordant results to optimize the accuracy of the device and avoid false negative or false positive. So I'll stay here. But you're more than welcome to look at the paper. There's more details in there. One other main information from our work was that there is a lot of gaps of evidence, of scientifically sound evidence, of too informed policy, in fact, to choose the right device. I listed a few here. There is very limited information on what's the best device to screen substandard medicine containing, for example, incorrect active ingredients, as I mentioned earlier. To detect degraded medicines, medicines to support the solution. There is also not much evidence on how coating of tablets will interfere with the screening devices signals, especially spectrophotometers. There is a lot of information on tablets, but not on other formulations like capsules, vaccines, like all injectables. Testing strategy in the field, how many times did you test the sample? I already mentioned this earlier. Reference library, I'll come back to that, but there is not much evidence on the number of batches required to take into account the inter batch viability. Some medicines contain very low concentration of APIs, active ingredients, as dioxins. And we know that some devices already cannot detect those. And how do for devices that can screen through packaging, how do different plastics and glasses would affect the rate penetration? Other types of gaps that we identify are more like for infrastructure support or implementation of the devices. And I mentioned reference library. So for those who are not familiar with the spectrophotometers, the way they work is that, in fact, when you use, you will scan the sample and get a signal. And this signal is called a spectrum. And this spectrum will be compared to what we call the reference library, often called a reference library. And this will be compared within the device. And then there will be a match or not match or fatal pass results in that data will be a falsified sample. But how do we construct the reference library? So it needs to be that of the genuine sample produced by the manufacturer. And that needs to be done individually for each product brand and for each formulation of that brand. So that's quite a lot of work when you think about the number of different brands and products and formulations available in the market. So a lot of discussions in our stakeholder meetings was about who will be responsible for creating the reference library and for updating them because as soon as there is one modification of an excipient of process in the manufacturing process of the medicine, then there need to be a new reference library. So how will the libraries be stored? And also how do we ensure that the medicines we used for creating the reference library are actually of good quality and not substandard? But that would bias the result of the device. There was also discussions and there is not much evidence on how much training is needed and like regular training or not and also a past effectiveness analysis are lacking but UL is going to tell you more about the work is performed. One also key point is what are there going to be for devices that can only detect falsified medicines with no or wrong agreement or that are very accurate for those but less for substandard medicines? What is going to be the impact of false reassurance of the medicine quality in the context? Last one is there is not a lot of evidence is what are the desirable qualities at different level of the supply chain in different environments? You know already probably that slide from the WHO GSMS report from 2017. Globally the pharmaceutical supply chains become very, very complicated. So which device do we use at which level? That's a key gap. So in conclusion, even though screening devices look pretty promising to empower MIs and medicines regulators and medicines inspectors in particular, there doesn't seem to be one device that will be able to test to be used effectively to monitor the quality of medicines in the whole pharmaceutical supply chain but more research and pluridisciplinary research in particular is of great importance to cover the gaps of evidence regarding screening devices. Thank you, a special thanks to the team who performed that in this study. Thanks.