How are you? Very well, President. President, welcome back. Good afternoon. How are you, Doctor? How are you? Very well, please.
Well, thank you again. When we started this great project, Doctor, that we have been working on together, well, you know it better than I do, I think, but when we launched it to the population 5 months ago, there was a technology, and it is surprising that, well, now that we are launching this second stage or second phase, the surprising thing is not only that we have incorporated new services, which has been the main objective, but that we found new technology, and technology that did not exist 5 months ago already exists not only now, but it existed at the moment we started developing this second stage, because it was first developed, then tested, then a pilot was launched, then the results were studied, and then now we are launching it to the population. But in those 5 months all that happened, including the invention of the new technology. So, what it tells us is that for the third stage we will probably have new technology already. Let's hope so. And that is the most surprising thing, I think, besides the fact that the plan to launch this second stage was already there, even if we didn't have the new technology that has been developed in these last months, but this new stage of Doctor SB is going to cover a large universe of patients that sincerely excites me a lot because I think we are creating the best health system in the world.
That's right. If you remember, in November we launched Doctor SB, the application, and it was something spectacular because citizens could already have 24-hour, 7-day-a-week access to medical care. And that was for the first instance. And if you remember, at the time something you asked for was a survey at the end to see how the doctors, the laboratories were evaluated, well, in those 5 months we have results.
Yes, because we had a benchmark that was very important, which was the accuracy level. That's right. But there was another benchmark that maybe scientifically is less important, but for public policy is equally important, which is satisfaction, right? Of the people in its use, how they felt, and that is exactly it. So, when all that is available, there are more than 400 pharmacies and laboratories affiliated with the program. There were more than 1.1 million Salvadorans who registered in Doctor SB, and from that we have one and a half million appointments that have already been made. Now, imagine with all that. I don't know if you remember how much the accuracy was before in El Salvador.
Yes, it was around 50%, it was almost like flipping a coin, heads or tails, right? Heads they got it right, tails they didn't. That's right. That's right. And that was exactly the difference. At this moment the results are giving us 93% accuracy in Doctor SB with that program. So that is basically the highest, I think. It's very high. Now, add to that that with the surveys you requested, they are at 97% user satisfaction with that system, with the laboratories and how the doctors treat them. And knowing politics, having 97% approval in a public policy is almost impossible, because you will always have complaints or someone who was treated badly or someone who felt they were treated badly or didn't feel totally satisfied or even dislikes the public policy or has another affinity. We measure other types of policies in education, health, security, public works, economic policies, etc. Almost always achieving 97% in something is breaking the ceiling, very high. So, between this and even internationally, I think, correct, speaking internationally, if you remember recently the CEO of Google even spoke about this program at the World Artificial Intelligence Forum in India. So it is an important reference case. Now with that we go to the next stage. And what is the next stage? With those results, Doctor SB also evolves to treat chronic patients. What is a chronic patient? Someone who has hypertension, diabetes, kidney issues, it can be everything metabolic, it will be treated here and the program will evolve because before it was by incident and now it is to accompany the patient in their treatment and artificial intelligence will be used to the next level with agents that follow up throughout the process, which is the technology that did not exist before.
That's right. To really have a way to follow the patient throughout that path, which normally there isn't the human way to have so many hands to be able to accompany throughout this process. Yes, right now there are 1.1 million users, that's right. I don't know how many there will be in chronic patients, but it will keep growing, well, we can assume there will be hundreds of thousands of chronic patients and it would be impossible to give personalized follow-up to each one, call them by phone or remind them, look, you have to take your pill. And that is what they were explaining to me a little before when they were launching the pilot, they told me that when we launched Doctor, which was relatively recently, it was 5 months, the artificial intelligence that supported the doctor still did not have proactivity because agents like the ones we have now did not exist. That's right. But now artificial intelligence can be proactive and even complete tasks without human intervention, call the patient and remind them that they have to go get a test done or send them a message and tell them, 'Remember to take your medication.' And imagine, today they are handling 18,000 calls a day. When this program expands, remember we have up to 30,000 capacity today. When the program expands, well, it will have this access, but there will be many more, and that is where the intelligent agents support and just as it will be 30,000 because the 30,000 appointments per day are doctors, humans, who are attending to patients with the support of artificial intelligence and the entire system, but they are humans. Correct. But here this will be done by agents separately or in parallel, separately and on top. Uh-huh. Sure. And we are going to see this chronic population that Dr. Edgardo will talk about, how many are in each group and why it is such a large part of the population that is why we have to start covering it in this next way. And there is an important issue, I think you know it, Doctor, and especially here in El Salvador it happens and will happen in many countries, especially developing or underdeveloped countries even more. And it is that there are three issues. First, poor health care, that sometimes I go, they didn't attend to me, I was waiting two, three hours or they attended to me badly. That happens. Then there is also the effect of personal irresponsibility. I didn't take the pill. That happens in the first world and the third world everywhere. Sure. My responsibility. I didn't take the pill. I forgot. I didn't do the diet they told me. I consumed, I am diabetic and I consumed alcohol, sugar, etc. And then there is the other, which is sometimes the lack of medication. We see sometimes there are many complaints on social media that such medication was not available or they didn't give me such medication and it is crucial for a diabetic or hypertensive person. Medications cannot be missing, that is, they have to be there, yes or yes or yes.
Absolutely. In fact, the country has, according to the statistics we have, about 3 million people who already live today with one or more chronic diseases. And the problem is not only the number of people we have with chronic diseases, but many do not know they have chronic diseases, because they are diseases that evolve silently and when they manifest, they manifest with a serious event that can be a heart attack or the amputation of a foot in a diabetic or a stroke. Therefore, the approach we have with Doctor SB and using precisely the entire system of artificial intelligence agents is first to be able to identify who they are. So the application in a first stage tries to identify each of the patients, because let's keep in mind that only six out of ten people know they are hypertensive or diabetic. Six out of ten hypertensives know, meaning there are four hypertensives who will never find out they are hypertensive and will find out when they had a heart attack. With the issue of high cholesterol, four out of ten know, meaning 60% of the population with high cholesterol doesn't know it. And with kidney disease it is much more serious. There is only one person out of ten who knows they have kidney disease. The rest is evolving and will end up on dialysis. Yes. Nine out of ten don't know. Don't know. Only one person out of ten knows they have it, and that statistic is from here in El Salvador. That statistic is actually from the region. El Salvador doesn't have exactly a precise statistic on that, but in the region, nine out of ten people who have a kidney problem don't know it. They don't know it and they are advancing to need dialysis. Exactly. So that is why the strategy is first to identify who are the people who have a disease. And for that, as I was just saying, artificial intelligence will help us a lot because it will evaluate the risk factors that are currently in the clinical records to know which are these people who have the diseases. And also, a questionnaire has been incorporated into the application that Salvadorans have, so that we can know through filling out that questionnaire what are the risk factors we need to attack. Now, not only are we doing it, but the other principle we are putting into practice is ease. The questionnaire is very simple, it is intuitive. One can fill in the common data, what is the weight, what is the height, what type of diet they have, if they smoke, don't smoke, if they smoke, don't smoke, which are the most important risk factors. Sure. Artificial intelligence takes all this information and is already classifying where the patient's disease would be and puts together a package of tests that it puts in the application automatically, without the patient having to go to the doctor or have to make an appointment for a consultation, not even a teleconsultation, not even a consultation through the application, because the artificial intelligence, based on what the patient or user is putting in the application, is already determining what the package of tests is and already releases an order, and that patient goes to the laboratory with that order to get the tests done, and those tests are what will enable them to have a consultation with a doctor who is trained to treat chronic diseases, after the tests. After the tests. So that instead, let's say, in public health here in El Salvador or even private health, yes, I have to transport myself, whether by bus, car, Uber, I have to get to the doctor, make an appointment, tell him, 'Look, doctor, I have this and this and that.' The doctor is going to ask me some questions and then the doctor is going to tell me, 'Look, we are going to do this blood test, an electrocardiogram, a test, I don't know, all that, right?' Yes. And then I have to go to the laboratory to get the blood drawn. If it's private, well, I have to pay. Sure. Besides paying the doctor, I have to go pay the laboratory. There they will probably give me the results in an hour or a day or the next day. I have to go the next day to get the result. Let's say they can send it to me by email. Then I have to go back to the doctor and say, 'Here are the results.' And wait for a turn too. Ah, no, sure. And in public health sometimes that turn is 6 hours waiting, and then the doctor is going to see and is going to see the results for the first time and is going to analyze them there. Sure. And that's it. And then so he can tell me all that is done beforehand. All that is done beforehand and the artificial intelligence also already suggests to the doctor what the diagnostic classification would be for this patient based on what they filled out in the form and based on the tests they did afterwards. When the doctor sees the patient for the first time in the application, in the application, sure, on the cell phone the doctor is seeing the patient for the first time, but he already has all the answers, the entire questionnaire with all its answers, and has all the laboratory results and has the artificial intelligence suggestions on what the patient has. Exactly. And in that way the patient is already enrolled in the chronic patient program. Now, while the problem is that there are many and the problem is that many don't know, another problem is added that is even more serious and which you just spoke about, which has to do with the fact that these diseases are lifelong and require lifelong treatment and sometimes pills in the morning, afternoon and night and the patient needs to have healthy habits, lose weight in the case of being overweight, or have to do physical activity or change their type of diet. So, it is very common for patients to abandon treatment. It is calculated that adherence to these types of programs in general does not reach 60%, meaning out of 10 diagnosed and treated patients, there are only six who will follow the treatment and the other four, we say, it's as if they fall out of the system. Artificial intelligence has allowed us, through this entire process that we are doing with your government and all the agents, to create a whole series of mechanisms where the patient does not fall out. Let's say, when the patient does not go to get their tests done, artificial intelligence warns us that they are not going to get the tests done. That is, it does permanent follow-up of the patient to warn the doctor that they haven't gone, but also in parallel it automatically generates messages that it sends to the patient to tell them, 'You have abandoned your treatment or it's time to get your tests done and 4 days ago you had the appointment, you haven't gone.' And if it doesn't work, well, this message saying this is being generated by artificial intelligence. Absolutely, there is no human intervention in all this. And the, sorry, Doctor, and the, I mean, we are basically talking like every Salvadoran has a person, I'm going to call the artificial intelligence people a person, has a person dedicated 24/7 to be aware of them, of their chronic disease. Exactly. Exactly. Exactly. Look, you have to go tomorrow to get the test done, right? Exactly. What time can you go? No, I'll go in the afternoon. Ah, okay, then. Or remember that you have to take your medication. And it sends you a message once, sends it twice, sends it three times, and if you don't answer or don't go to get the test done or don't go to get the medication, then we call you by phone and we try. Does the artificial intelligence call or does a human still call? At this stage, not yet. The next stage is for it to be called directly by an intelligence. Now a human calls. Now a human is going to call. Well, but for the Salvadoran it's the same. Maybe they even prefer a human to call. Sure. So the artificial intelligence sends you a message. Let's assume it should respond and say, but let's say it's stubborn and doesn't do it. Then it reminds it again, it doesn't do it. It reminds it again, it doesn't do it. The artificial intelligence warns the human doctor and says, 'Look, this person, I already told them three times, they're not going, you call them. Better not to...' And then the human doctor calls them and says, look, we are calling from Doctor SB, please go. Exactly. Yes, that is something not even in private health. Sometimes maybe if you have a doctor who is a very close family friend or something like that, they might worry like that, but not even that, I think, for them to worry about you like that. We, the entire cycle that has been done in artificial intelligence from the first moment to the end is always governing the cycle. It is following, it is saying, 'Now we have to do this, now we have to do the other.' I mean, and it tells the patient what they have to be doing so that they don't fall out of the system. Sure. So this tool is a very powerful tool that is very underdeveloped in other countries. This country... is there any country that has something like what we have, the complete system? No. There are countries that have worked on artificial intelligence for chronic diseases, but they have done what we call vertical lines. For example, an artificial intelligence system that warns the patient and the doctor that their glucose is bad because they have a glucose device, nothing more than for glucose. There are other systems that warn that a patient is going to decompensate with kidney disease because they have all the kidney disease values, but only for kidney disease. A horizontal, longitudinal, end-to-end system that begins with the identification of the patient and then follows the patient for life with adherence, this is the first one in the world. In the world. I always try to relativize when we talk about something we are doing, for example, because if we build a road, for example, beautiful, new, and we publish it, this is the new road and people like it because they relativize it, well, what was there before? What is there now? But I am sure that any other developed country comes, China, the United States, and puts a 32-lane road, right? [laughter] And there, well, here is your road, it's nice, but at a global level, no, it's nothing, right? Yes, no. And generally what one tries to relativize, well, for El Salvador it's a good road, necessary, etc., but there has only been one issue where we have managed to position ourselves globally. We are number one in the entire hemisphere. In the other hemisphere there are some who beat us, well, but we have positioned ourselves globally, which is the security issue. Yes. Correct. With all the confidence in the world. We can say walking in El Salvador is safer than walking in Santiago, Chile, which are relatively safe cities in Latin America, or walking in Montevideo or anywhere else. We can say it in Canada, in Ontario, in Montreal, in Toronto. With confidence, it is the only issue where with confidence a developed country cannot come and tell us, 'We are better,' and at most if they are better, well, there we are competing. But that in health we can say, 'This system no one in the world has,' is another, another big issue. We did a research, we did a research to see how artificial intelligence is being applied in the world, because we always do prior research before, especially in chronic diseases, right? Sure. And this end-to-end, longitudinal system, right? This is the first one. Yes, this is the first one, and we understand that it will be scalable and as new challenges appear in the evolution of everything that is the chronic patient who gets complicated and very often, we will continue with our system, scaling it, because Doctor SB has all the capacity and the possibility to accept and take on those challenges. This is the second stage. It's exactly that, well, the first was 5 months ago, it took us a few weeks to open all age groups. Then it was launched completely, then some mini-stages were launched, well, mental health was included, another series of little things were included, mini-stages, let's say, how is it? 1A, 1B, 1C, but this is the second stage which is for patients with chronic diseases and that we are already talking about a totally different level from simply basic consultation or a cold, an allergy, a migraine pain, right? Here we are already at another level, we are talking about diabetes, hypertension, kidney diseases, etc. But then looking at the third stage, fourth stage, at some point we will be treating cancer, we will be doing surgeries, that's where we are going. Sure, in other stages. The other thing it has that is interesting is that the protocols we are using for the treatment of chronic diseases are clinical protocols made in the context of El Salvador, with the reality of El Salvador and with the tools that the laboratory has and that it has in imaging and the possibilities that El Salvador has. Which means this is an absolutely realistic system, it is a system that is applicable and that is adapted to the possibilities our country has. Within those protocols, what we have are the addition of new medications like insulin, for example. We are going to treat, we are going to have insulin to treat diabetics, we are going to have antihypertensives, the medications for pressure, for patients who have arterial hypertension, and also complemented with new functionalities of the network. Let's remember, as Gay was just saying, we have more than 400 providers in a private care network where we have laboratories, imaging, pharmacies. Well, now we are adding in-person care for services of a bit more complexity that chronic patients need. So, I mean, the doctor in person, right? The doctor in person, I mean, the doctor and other professions that are not medical, but that in chronic diseases are crucial. For example, a podiatrist, the diabetic foot and the amputation of the diabetic foot produce a lot of impact and in reality if one takes care of a diabetic person's foot in time and prevents it from getting infected, it won't end in an amputation. So, we have cardiologists, we are going to have podiatrists, we are going to have specialized nurses for in person. In person, the doctor is going to send them to go see the office of... Yes, because this is something important you mention, because there are people who say, 'Doctor, that is a telemedicine system,' but it is not a telemedicine system. It has telemedicine, but in the same way it has laboratory tests, which are telemedicine, it has medications which are telemedicine, it has imaging which is telemedicine, it has artificial intelligence which can go very well with medicine, with telemedicine, but is not necessarily part of telemedicine, or is it? Correct. And now it will even have in some cases health personnel, doctors or other types of personnel in person that are also not in telemedicine. So, it is becoming a complete health system. Absolutely. So there we will be complementing with the specialties we need so that the service we provide is a complete service. And what is necessary to keep in mind is the economic impact that complications of chronic diseases have, not only on the health system, from the accident, the event itself like the heart attack or the stroke, but everything that costs afterwards the rehabilitation practices of these patients who have been left with half their body paralyzed or who have one less foot. So, all the economic and human effort made on these patients has a great economic impact on the system, but it is not only on the system. You think that a family that has at home a young person because they didn't know they had the disease and who debuted at 55 with a stroke and was left with half their body paralyzed. From there, the economic burden for that family, beyond the health service systems, the burden, there always has to be someone taking care of them, there always has to be someone doing tasks for them, there always has to be someone who mobilizes them, I mean, the economic impact that... and the person who cannot work. And the person who cannot work. Exactly. So, the economic impact that chronic diseases have on society is very high in the pocket of the patient themselves for issues that go beyond health. So, that is why, a little to conclude, what we say is that the spirit of this program is that no Salvadoran who has a chronic disease is left without knowing it and that no one who knows it is left without treatment. That is the spirit we have within this program and artificial intelligence has been of great help and of great value so that we can do this and here we are on our way and we are also doing it with a lot of research support, which Dr. Estela will talk about later because we are doing it with scientific foundation. There is an important issue to mention because in the traditional system, let's say, it also always exists and there it is, well, people can use it, people who like traditional systems and there they are, but in the traditional system that you mentioned, the problems already exist, right? Only one in ten knows they have a kidney problem, six out of ten, I think, in the case of diabetes, four out of ten in the case of cholesterol, cholesterol, memory, and but then that percentage that is small and or minimal in the case of kidney issues, of that percentage, then those who receive, those who adhere to the treatment is already a much smaller percentage. I mean, still of that small percentage there is a large part that does not adhere, does not adhere, whether for any reason. And then there is another problem that happens, especially in developing countries like ours, that sometimes it is not the patient's fault, but the health system's fault, that there is no medication, that there is, but they didn't give it to me on time, they gave it to me late. All that disappears here. Absolutely. Here there is never a lack of medication, never a lack of a test, the care is always timely, fast, on time and adding that you have an agent reminding you and trying to support you in the individual responsibility you have over your discipline in adhering to your treatment. That's right. That's right. That is why we think this program is going to have a very high impact on the general health of the country. Chronic diseases represent the most important disease burden the country has at this moment. And basically this program can treat 100% of that. All those who adhere, all those who enter the application and fill out the form will be there. But also those who don't fill out the form and who have had a consultation for something else and in their tests, in their exams, have some variation in their glucose, for example, very above what is expected, we are going to capture those through artificial intelligence, reviewing clinical records and seeing in the clinical records who are at risk. And those we are going to contact through artificial intelligence. It's the two things, right? Either they come because they themselves filled out the questionnaire, did the exam, or consulted for a headache and they found they had a cholesterol issue. Exactly. They asked for tests and let's say they asked for tests, they have high cholesterol and the patient didn't come back anymore, the headache went away and they didn't come back anymore. Yes, the acute pain went away and they didn't care anymore. So, they didn't go back to see their cholesterol. Artificial intelligence will realize they had high cholesterol. Yes. And achieving that personalized care with human health personnel is humanly impossible, literally. Yes, of course. It would require tens of thousands of people just to be aware of whether they arrived on time, if they took the medication, if they did the test, if they came back for a consultation, if they went to get the electrocardiogram, how it turned out, if it reached the doctor. It would be practically impossible to achieve that. Correct. That is why chronic diseases are not only, but continue to be the most serious problem of countries. Today when one looks at the health systems of countries and what is the disease burden that has the most impact on countries, they continue to be chronic diseases because they are very frequent and because it is very complicated to manage them. What we are putting together at this moment, what we have developed, is a whole system of humans and artificial intelligence that makes this management automated and that we can really manage this. We believe that this care model is going to be a regional model that will be seen as the path to follow to treat chronic diseases. That is what we are almost sure is going to happen.
Excellent. Excellent. Thank you for having me here again today. Thank you for coming. Happy to have achieved so many things in a very short period of time and we actually launch this new stage that is going to cover a whole lot of new diseases and problems. We're advancing really fast. Well, I think you have a lot to be happy about and we can talk a little bit more about the global relevance of what's happening in El Salvador and I'm very happy to tell you that Doctor is already part of the international conversation on healthcare and integrating AI into healthcare. As you know, our first commentary on Doctor has been published in Lancet Regional Health and that means that the program is being studied and discussed beyond the borders of El Salvador right now. So that's great news and there's a lot of interest because as my colleagues said, this is a unique program in the world. So a lot of countries are looking for their own novel ways to treat the growing burden of chronic disease. And so they're looking to El Salvador. But anytime we introduce something novel, something innovative into healthcare, we have to be careful and we have to ask, well, does it actually help people or does it harm them, right? We say in medicine, do no harm. That's an important tenet. So to answer that question, we're currently conducting a clinical trial and we chose that study design because it generates the best quality evidence for whether or not something works. So this randomized design allows us for a fair comparison and it allows us to actually determine what is the effect of AI support on the doctor's diagnostic accuracy. And that kind of evaluation is especially important now because other countries are also experimenting with AI and healthcare, but not all of these experiments, unfortunately, are being carefully evaluated and the results are not always being published. So by doing this trial, El Salvador is helping establish a higher standard for evidence in a clinical trial. It's a full clinical trial. It's a full clinical trial. That's right. It's in progress right now. Normally those are not done here. Usually you will read about those done elsewhere. Well, you can read about it right now. It's already been registered on clinicaltrials.gov, which is a website for global clinical trials. So, we're doing it using the most rigorous methods using approval from the ethics committee here in El Salvador. So, we're quite excited about the upcoming results and we really think that again this goes beyond just having a strong national program. It can be a reference point for how digital health can be implemented and also evaluated and eventually scaled. And I'm very happy to see regional institutions like CF are following us closely and...
Because they're considering Doctor SB as a model for other countries as well. It's the first time El Salvador is going to be discussed at a global scale in something that's not security, in a positive way. We do a lot of things in terms of infrastructure or economics, but it's never at the top of the world. In this case, I really think we can achieve results that are not only way better than what we had before or even the region, but even developed countries.
