 Thank you for the invitation for the Medicine Quality Research Group to share our work with this wider research group. My name is Connie and I'm currently based in Viennjandau, PDR working under Long Rood. I will present the results of a systematic review that our group completed and submitted for publication recently on substandard and falsified antibiotics. It was a joint effort by the whole team, but mostly by Dr Guillermo Zabala and I. So what are substandard and falsified medicines? There were a few terminologies and definitions used to describe categories of medicine quality. These are the WHO 2017 definitions. Substandard medicines authorize medical products that fail to meet either their quality standards or their specifications or both. Falsified medicines are medical products that are deliberately or fraudulently misrepresented in their identity, composition or source. Falsified medicines are also known as fake or counterfeit medicines. In this talk, I will refer to substandard and falsified medicines together as SF medicines or SF antibiotics. Looking back, SF antibiotics have been in circulation since the first productions of antibiotics. These are some notable historical cases relate to SF antibiotics. First in the United States in 1937, within eight months of production of self-anilomide or a liquid, 105 deaths were reported as a result of a toxic-excipient use. Second, falsified penicillin were being traded within five years of first use. And third, in Uganda, a 13-year-old boy died of fatal bacterial meningitis, possibly due to substandard keftriaxone where laboratory analysis of a similar vial given to the boy found to contain less than 50% of the active pharmaceutical ingredients or what we call ATI content. WHO estimates the prevalence of SF medicine is approximately 10% in low and middle income country. In addition, antibiotics is the second most frequently reported class of SF medicines to the WHO global surveillance and monitoring system. After malarial medicine. In regards to the connection between SF antibiotics and antimicrobial resistance, there is no scientific data to backup the suspicion that SF antibiotics contribute to the emergence of AMR. The lack of evidence could be a result of neglect or difficulties in establishing a clear connection due to the many drivers of AMR, which can confound the relationship. However, observational data indicates that high prevalence of substandard plurinphenicol and sulfur amytoxazole and trimester from enema may have contributed to cyphoid antibiotic resistance. And scientifically, SF antibiotics with lower active ingredient content than expected results in sub-reputed concentrations promoting AMR. Currently, there are 736 million people still living in extreme poverty with impaired access to essential medicines, including antibiotics. We expect access of essential medicines to improve and the demand and consumption of antibiotics are projected to double by 2030. The circulation of SF antibiotics will likely increase as the demand and consumption increases. If we fail to tackle quality problems, this will surely lead to high absolute number of SF antibiotics reaching the population, causing direct harm and further contribute to AMR. Our aim for the review is to understand the epidemiology of SF antibiotics globally, discuss the implications of the findings, and highlight some potential impact of SF antibiotics on AMR and potential patient outcomes. Our review included scientific and grey literature. The database searches included publications from conception up to the 31st of December 2020. Only publications written in English, Spanish or French that evaluated or discussed the quality of antibiotics were included. We included different types of studies and publications, but prevalent surveys were most valued. A prevalent survey is a study in which samples were collected within the pharmaceutical supply chain to assess their quality in order to describe the prevalence of circulating SF medicines. We excluded antibiotics primarily used for the treatment of spherculosis. A similar work on TV medicines by our team is currently underway. The following are our key findings. In total, 10,137 publications were screened. 648 publications were included, which 498 were original research publications. Most were studies on laboratory or analytical techniques used to test quality of pharmaceuticals, but these data do not inform us on the epidemiology of SF antibiotics. 117 prevalent surveys were included. As mentioned earlier, the study aims and methods used in prevalent surveys attempt to describe the prevalence of SF antibiotics circulating within selected geographical area. They provide data on the epidemiology of SF antibiotics, therefore our key findings are based on these prevalent surveys. Equivalent studies assess the quality of different brands of antibiotics containing the same excess ingredient in the market, and there are small numbers of notifications or recalls and alerts, seizures and case reports. There is a growing interest in the epidemiology of SF antibiotics. This graph shows a number of prevalent surveys published in a year since 1992, with increasing trend of number of publications per year up to the end of 2020. And half of all these were published between 2010 to 2020. This atlas with color coding of different countries shows where samples were collected for prevalent studies. The darker the blue shade, the higher the number of samples collected in that country, ranging from 17 samples to 1,334 samples. Samples were collected from 67 different countries, but only in certain regions of the world. Majority of the samples are from Africa and Asia, small amounts from South America and very small proportions from Oceania and Europe. The highest number of samples, just over 1,000 samples each, were collected in Laotidia, Cambodia, India, Mongolia and 632 samples in Nigeria. The median number of samples per study was actually quite low at 47 samples. By GDP income category, more than half were from low middle income countries, 20 from low income countries and 8% from upper middle income and high income countries. We combined reported quality results of 13,555 samples from 107 prevalent surveys. The overall failure frequency was 17.4%. We defined failure frequency as a proportion of samples included in a prevalent survey that failed at least one quality test described in the report. These were made up of 1% falsified, 3% substandard and 13% substandard or falsified. The quality categories substandard or falsified are products that failed at least one quality testing, but there is no information on the packaging authenticity. The term is not used by the WHO for quality of medicine, but we think it is important to differentiate these as it is not possible to accurately classify them as substandard or falsified without packaging analysis. The methods used to collect samples for the studies were mostly convenient sampling results at 66% and randomization of outlets of sample at 28%. This atlas follows on from the previous atlas the number of samples collected for country. This one is of quality failure frequency of samples assessed. Again, the darker the shade of brown, the higher the failure frequency for samples collected in that country. The failure frequency of samples range from 0% to 71%. Per region, the highest failure frequency was in Africa at 28%, Asia at 15%, Oceania at 15%, America at 13% and Europe at 8%. The source where samples were collected were often reported. Most studies collected samples from different outlets, but results for half of the samples were not broken down by outlet types, instead were aggregated as a combination of outlets. We also reported the results by outlet type, private pharmacies were the most commonly sampled at 33%. Only 5% were collected from unregistered or unlicensed outlets, looking at the quality of the samples by different outlets, samples collected in unregistered or unlicensed outlets had the highest failure frequency at 34%. Samples were also broken down according to the WHO aware classification of APIs to align with global work to optimize antibiotic usage, reduce AMR and improve access to quality, safe and affordable medicines. 21 out of 48 antibiotics in the access group were included in the prevalent studies, 16 out of 110 antibiotics in the watch group, none of 22 antibiotics in the reserve group were studied. The failure frequency for the access group was 20% and the watch group 14%. The top five APIs most studied were Cyprophyloxacillin, Amoxicillin, Sulfo-Methyloxazole Trimessoprims, Tetracycline and Amphicillin. Cyprophyloxacillin from the watch group make up 20% of all samples and had a failure frequency of 10%. Quatermoxazole was third highest proportion of sample studies and had the highest failure frequency at 26%. When assessing the quality of products, most samples use more than one quality test on a sample, a median of three analysis techniques per sample. The quality defect that has potential impact on AMR are API content, dissolution and impurities or contaminants. The failure frequency for API content was 7 to 10% where 7% of the samples had lower API content than pharmacopoeia reference. 2% was higher API content, 7% with lower and higher API content between units within the same sample, and 1% with no API. 9% failed due to impaired dissolution rate and 3% contained impurities or contaminants. So to summarize our key findings, there were 107 prevalent surveys from 67 different countries with 13,555 samples in total and an overall failure frequency of 17.4%. Of the failed samples, there was a high proportion of the standard of falsified antibiotics without authenticity verification. Limited analysis techniques were used for quality assessment, majority tested for API content, and to lesser extent dissolution and packaging analysis. The main quality failure was due to increase API content or slow dissolution rate, which affects bioavailability of the product compared to pharma-compared standards. There are gaps in the data if most samples were from low income and low middle income countries in Africa and Asia. Majority are antibiotics in the access group, small proportion in the watch group and none from the reserve group. There were limited data for many antibiotics of importance for public health and AMR. Survey methodologies and reportings were of low quality of small sample size and two-thirds used convenient sampling methods, which can introduce bias to the results. The findings are further limited as only English, French, and Spanish publications were included. We are also limited to the data sets available in the public domain and do not capture data from medicine regulatory authorities in the pharmaceutical industry that are usually unavailable to the public. The quality of the samples were also determined by authors' conclusions and we do not perform any independent analysis. Due to these limitations, our findings are not generalizable. The overall failure frequency of 17.4% does not indicate that the proportion of ECF antibiotics from the global market is 17.4%, but this data show reasons for global concern. ECF antibiotics can lead to treatment failure as a result of subtherapeutic concentration due to lower API content or delayed bioavailability, as we suspected for the 13-year-old who died from bacterial meningitis. A mathematical modeling also indicated this led to 72,430 childhood pneumonia deaths per year. There is a risk of toxicity as a result of high API content, then pharmacopoeia standard, and presence of cryptic contaminants. These could lead to decreased public confidence in pharmaceutical products, health workers, and health care systems. And currently, there is no scientific evidence on the relationship between ECF antibiotics and AMR. So what can be done? Poor antibiotic quality appears to affect all classes, especially those which are most consumed and countries facing greater socio-economic challenges, where health systems are weak or non-existent. To reduce circulation of ECF antibiotics, more could be done to strengthen medicine regulators with skilled staff, use of advanced technologies to test for costs, improve manual picturing process, storage logistics, and sales practices to ensure qualities throughout the supply chain. This should include cooperation at international level, increase access to safe sources of antibiotics, especially to people living in socio-economic challenging regions, unlicensed outlets, which are a common source of medicines in some low and middle income countries, seem to be particularly affected by this public health threat. Work could be done with unlicensed sellers and informal vendors to ensure safe source of antibiotics. By definition, the root cause and motivation for the production and circulation of ECF medicine are different, and therefore actions to address them need to be different. To better inform the epidemiology and type of quality issues of ECF antibiotics, and therefore appropriate actions taken. So future prevent studies should aim to perform authenticity analysis impossible. Research to assess the link between ECF antibiotics and AMIs needed, such as marrying antibiotic consumption data with the prevalence of ECF antibiotics could be a novel and meaningful method to assess the population risk of ECF antibiotics, and with a high antibiotic consumption and quality issues related to AMR. And finally, we hope that the finding of this review will highlight to the scientists of communities the needs of said local researchers in this area. Thank you for the opportunity to share this research with you all.