for the Prevention of Disease in Endocrine Society Clinical Practice Guidelines. There's a couple of announcements that we'd like to make before we get started, so let me just take you through this. So I'll be the moderator, and I'll be talking to you after all these talks, and we hope that there'll be a lively discussion and questions. On our panel, we have Marie DeMay, Tasso Pittis, Deborah Mitchell, Robert Scragg, and Maraid Carley from the University College in Cork. And Chris McCartney from West Virginia University will be the methodologist who will be involved. So the Guideline Development Panel actually included a number of people, and Marie reminded me that it's been at least four years since this started, and you can see the list. I won't read them for you, but this is a multidisciplinary approach that includes not just the methodologic approaches, but also people from all over the world, and they represent a number of different disciplines. So these are the disclosures. I won't actually read them aloud, but you can see what these different conflicts, if there are any, where funding comes from. So I'll just pass this by quickly because we really want to get to the meat of the issue, but here are the final disclosures, and here is the agenda. So first, we'll start with a framework, and I want to point out to you that we'll go through the clinical practice guideline recommendations for pediatrics, adults, pregnancy, and special considerations, particularly prediabetes, daily versus intermittent dosing, and screening. We'll have a summary slide and a path forward, and then the questions. And so to get on to the questions, we may have a long line of people, but there's two ways to do it. One, you can stand up at the microphone and ask your question there, and I'll try to recognize people there. You can also scan the QR code that you see up here to put your question in the cloud, and we will be able to read that aloud. So with that, I'll get started, and Chris will bring us up to date with the clinical problem. Good morning, everyone. Thank you for being here. As Dr. Rosen said, I'm Chris McCartney. I've been the methodologist for this guideline, and I'm going to take you through just a very brief summary of why we developed the guideline and a little bit about how we developed the guideline. I think most in the audience are going to know that overt vitamin D deficiency is characterized by marked clinical findings like rickets, osteomalacia, and hypocalcemia, and that population-based observational studies have consistently disclosed relationships or associations between lower vitamin D levels, 25-hydroxyvitamin D levels, and a number of undesirable outcomes including fractures, infections, cardiac disease, cancer, and mortality. But what's unknown is whether those are causal associations and whether vitamin D replacement or treatment will actually improve those outcomes. So that's been largely unknown. It's led to a lot of controversy. In 2011, the society put out a guideline related to the evaluation, treatment, and prevention of vitamin D deficiency. But as the Clinical Guidelines Committee started seeing more big randomized clinical trials related to vitamin D, it realized that it may be time for an update. So in 2019, the CGC prioritized or put this vitamin D guideline into queue for an update. So a little bit about the process, we have a number of subject matter experts. We have a number of adult endocrinologists, a pediatric endocrinologist, a general, several general internists, a maternal fetal medicine expert, and experts in nutrition and epidemiology. We have two guideline methodologists. I was the one taking the lead on the grade implementation that I'll describe in a minute. And one took lead on the systematic evidence reviews. Importantly, we also had a patient representative to help us, guide us through patient values and give us a patient perspective. Something that's not on the slide that I think is really important is that also at this time, we were working under a new conflict or duality of interest policy for guideline panels. And so the relevant industry relationships in this panel were minimal and possibly even none. So it's in that regard, it's quite, I think, favorable for trustworthiness. I'm going to spend just a little bit of time on this slide because it's really important. Some key points, sort of a foundational beginning for the panel. One, the panel decided that it would really focus on generally healthy populations. We're not talking about people with established osteoporosis with hypocalcemia related to low vitamin D. We're not talking about those folks. We're talking about generally healthy populations. And we started with an a priori assumption that the recommended daily allowance as established by the Institute of Medicine, now the National Academy of Medicine in 2011, that that's advisable for everyone. And as a reminder, that's 600 international units a day for ages 1 to 70, including during pregnancy, and then 800 international units a day for ages over 70. The second really important point is that we all know that observational studies are subject to bias and confounding. And it's very difficult to account for confounding. So the panel decided that whenever we can, if at all possible, we are going to focus on randomized clinical trials because they're the most reliable. In fact, we said we'd only look at observational trials if we didn't have, if we didn't feel we had enough evidence for randomized trials. And then even if we did use observational trials, we would only use large, well-done clinical trials that reasonably, in our view, would account for confounding reasonably. But as I mentioned, it's impossible, well, it's very difficult to really know you're accounting for all important confounding. And then thirdly, whenever possible, we were going to focus on what we call patient-important outcomes instead of surrogates. And I'll describe what I mean by this in just a minute. The process we've been using and we've been using much more faithfully in recent years is the GRADE process. That's Grading of Recommendation, Assessment, Development, and Evaluation. I'm going to outline or summarize just four brief steps here. One is evidence synthesis. And the first step for that is that the panel brainstormed what are the important critical clinical questions that are being asked by patients, by clinicians. And we wanted to prioritize those particular clinical questions. We also wanted to prioritize what are the outcomes that we're most interested in. There are tons of outcomes we could have looked at, but it's difficult to manage in your mind a lot of outcomes. So we selected those that we thought were absolutely the most critical for our decision-making. The panel prioritized three general kinds of questions. The first, should generally healthy populations receive empiric vitamin D supplementation or not? And here's a critical point. What the panel means by empiric vitamin D supplementation is vitamin intake that exceeds the RDA. I've already mentioned the panel assumed that the RDA is advisable for everyone. So should people take more? And by empiric, we mean without vitamin D testing. So not conditioning a recommendation on a vitamin D level. But that leads to our second question, type of question. Should vitamin D supplementation beyond the RDA be limited to those with 25-hydroxyvitamin D levels below a threshold? In other words, you check vitamin D, you offer more vitamin D to those only if their 25-hydroxyvitamin D is less than X. And then third, should certain groups be screened with 25-hydroxyvitamin D tests? So the general population, those with obesity, those with dark complexion, should you screen all members of that group with vitamin D levels? All right. I mentioned patient-important outcomes that we prioritized. And we prioritized those that were critical for decision-making. Here's just an example of some, this was one of the clinical questions we asked. And these were the outcomes, mortality, fractures, cardiovascular disease events, cancer, and then adverse events. Any treatment has the capacity for adverse events. And so we included that in all of our questions. What do I mean by clinic patient-relevant outcomes? Well, you could use a surrogate for fractures, for example. Maybe you don't have evidence related to fractures. You can use a surrogate like bone mineral density. We do that all the time. But anytime you do use a surrogate instead of the actual outcome that we're interested in, you introduce some uncertainty. I'm going to be a little bit provocative here. I would say that unless conditioned to do so, patients don't really care about their BMD. It's asymptomatic. What they really care about is how that translates into a risk for fracture. So we would say fracture is really the patient-important outcome. And you might use a surrogate to try to get at that risk. As a next step, once we formulated those questions and outcomes, we had, we have partners at the Mayo Clinic led by Dr. Hassan Murad who would do the systematic evidence reviews for us. And each of these clinical questions had its own systematic evidence review. Those were produced by the Mayo team. They also helped us understand what is the certainty of evidence in the evidence that they provide to us. And maybe this is well-known to you all, but you're going to see in the guideline the certainty of evidence designations. It can range from high to very low. And just a reminder of what these things mean. It means something, when we use them, we mean something very specific. High certainty of evidence means that our estimated effect, we're really pretty confident that that's very close to the true effect of the intervention. If we say moderate, we have moderate confidence. That means the true effect is likely to be close to the estimate, but there's a possibility that it's substantially different. If we say low certainty of evidence, it means that our confidence is limited. The true effect may be substantially different. And then very low, if we have very low certainty of evidence, it means we have very little confidence. And it's likely that our estimate is substantially different from the true effect. All right. Then we go through a process of what's called evidence to decision. It's the way we take evidence and then move, translate that into a recommendation. The panel addresses and make judgments about all important evidence to decision factors, and I'll describe these in a minute, and then makes a recommendation, whether to implement the vitamin D, for example, versus not implementing vitamin D. So the evidence to decision factors, we're all used to thinking about health effects. So that can be the desirable effects of the intervention, the undesirable effects, and what's the balance of those effects. And we also consider what's our certainty of evidence that we know what those desirable and undesirable effects are, but there are also some really important contextual factors. What's the cost of the intervention? What researchers are needed to implement that intervention? What's the cost effectiveness? Do we know it? What's the impact of the intervention on health equity, which is something that we paid attention to? Is the intervention acceptable to key stakeholders? And also, is it a feasible intervention to implement? These are all important factors to consider when we make a recommendation. And so at the end of the day, we have desirable consequences of an intervention and undesirable consequences as compared to the comparator, and the panel has to weigh those, make judgments about the weighting, whether there's a net benefit or not. So here's an example where, just an example where maybe you have a situation where the desirable consequences clearly outweigh the undesirable consequences. So there's clearly a net benefit of an intervention. So in that case, we may issue a strong recommendation, meaning that really most people should get this intervention as an example. In contrast, maybe you have a situation where either the desirable consequences or in this case, the undesirable consequences probably outweigh the other. And in that case, when you think you have a probable weighting one side or another, but you're not, it's not clear, then in that case, we may issue what we call a conditional recommendation. So what do we talk about when we talk about a strong recommendation? If we were to issue a strong recommendation, as I said before, it means that we've judged that most, if not everyone, should get this intervention, most, if not all the time. To let the cat out of the bag, we, all of our recommendations were conditional. So what do we mean by that? It means that we've judged that a majority of people in this situation would want the recommended course of action, but a substantive proportion would not. And practically speaking, it means that clinicians need to understand the data, understand the benefits and harms of an intervention, counsel their patients, and enter into the shared decision-making process to help the patient make a decision based on their values. And with that, I'm going to turn it over to Dr. Deb Mitchell from MGH to talk about children. Thank you. So for each of our recommendations, we'll be starting with a clinical vignette. So for case one, we have a seven-year-old boy who has come to his pediatrician's office for a well-child check. There are no significant medical concerns endorsed by the family. He's growing well at the 25th percentile. He is eating what is unfortunately a fairly typical diet for patients in my clinic, nuggets, burgers, pizza, mac and cheese. He's eating limited fish, limited fortified dairy intake. And so poll question is, would you recommend vitamin D supplementation for this child? And the possible answers are yes, no, or we don't yet have enough information. Wow, a lot of different practices. Okay. So the recommendation from the panel is that in children and adolescents ages 1 to 18 years, we suggest empiric vitamin D supplementation to prevent nutritional rickets and potentially lower the risk of respiratory tract infections. So as mentioned earlier, this is a grade two or conditional recommendation with two out of four pluses, so low certainty of evidence. And a couple notes about this recommendation. First is that we were only looking at children ages 1 and older. There are recommendations from other organizations about vitamin D supplementation in infancy. So we were only looking at children ages 1 year and older. And in terms of what we mean by empiric vitamin D supplementation, this can take many forms. So this can be daily intake of fortified foods. And different countries have different practices about food fortification. This can be vitamin formulations that contain vitamin D or a specific vitamin D supplement. So in terms of the outcomes that the panel examined for pediatrics, we looked at symptomatic rickets, respiratory infection, fractures in adulthood, autoimmune disease including but not limited to type 1 diabetes, inflammatory bowel disease and multiple sclerosis, type 2 diabetes and or pre-diabetes, allergic disease including asthma and atopic dermatitis, and adverse events. So looking first at rickets and talking about the scope of the problem. So while many people think of rickets as really just a historical disease, in fact, we see it quite frequently. There's a rising incidence in high income countries, particularly in groups such as immigrants and racial and ethnic minority groups. In low and middle income countries, the burden is much higher with estimates of prevalence of anywhere between 1 and 24% based on the region of the world. And on data I'm showing on the right, this is a cohort study out of Turkey looking at the age of diagnosis of rickets from a hospital system. And while, yes, the bulk of patients were diagnosed in the first two years of life, there is ongoing incidence throughout childhood. So we do need to worry about the risk of rickets in anyone who has open epiphyses. So what is the evidence for vitamin D and prevention of rickets? So the panel is not able to find any randomized placebo controlled trials supporting vitamin D for the prevention of rickets. And this is likely because this was an established clinical practice before the development of modern clinical trial methodology. However, there are numerous pragmatic trials from the first half of the 20th century which established the importance of vitamin D, commonly as cod liver oil, the active ingredient of which is vitamin D, in the prevention of rickets. And supporting this is in 2005, there was an initiation of a program in Turkey with free distribution of vitamin D to infants, and this was associated with a decrease in prevalence from 6.1% to 0.1%, so certainly environmental and epidemiologic evidence of efficacy. Turning to respiratory tract infections, again, an enormous problem. Pneumonia is the leading cause of death globally in children ages one month through five years, with an estimated 762,000 deaths per year. The etiologies of this are varied viral infections, and particularly RSV is the most common cause of mortality. But additional causes, including bacterial and mycobacterial and fungal diseases are common as well. So the panel was able to find 12 trials, including 12,951 participants with using vitamin D placebo-controlled trials in children. These trials had very variable dosing regimens. So some trials were small daily doses, some trials had moderate weekly doses, and some trials, two trials were one single high dose of vitamin D. Weighted average was approximately 1,200 units per day across the trials, but again, with high variability. The relative risk of respiratory, the number of participants who had a respiratory tract infection was 0.94, with a 95% confidence interval of 0.87 to 1.02, so it crossed one, and this would translate into an absolute number of 43 fewer infections per thousand with a range of 93 fewer to 14 more. And the age subdivided at less than or greater to five years, and the dose of vitamin D did not modify the outcomes. When we looked at these data, looking at not the number of participants who suffered an infection, but the number of infections, there were three trials that reported this as an outcome. And when we looked at these three trials, the incidence rate ratio of infection was 0.64, with a 95% confidence interval of 0.51 to 0.82, so it did not cross one. Again, vitamin D dose did not modify outcomes. And in support of this finding, there was one trial which did not specify the number of infections, but did stratify into three or more infections over six months. And in the vitamin D treatment arm, 7.7% of participants had three or more infections versus 32% in the placebo arm. Looking at other outcomes, there were six trials that reported on lower respiratory tract infections, with approximately 10,000 participants. The risk ratio was 0.93, and the 95% confidence interval crossed one. And similarly, with tuberculosis, we found two trials, again, roughly 10,000 participants. The risk ratio was 0.67, with a very wide confidence interval. We found insufficient evidence to make a recommendation regarding asthma, and we did not find any randomized placebo-controlled trials looking at the risk of fractures in adulthood, autoimmune disease, prediabetes, or type 2 diabetes. We did not find any evidence for harm of vitamin D treatment. So putting the evidence together, again, this was a conditional recommendation. And in the justification for the decision, the panel felt that the importance of Ricketts prevention and the importance of the prevention of recurrent respiratory infection was quite high. These are both diseases that cause significant morbidity. And in the case of respiratory infections, significant mortality. Economic modeling, a couple studies suggest cost savings or cost effectiveness in the prevention of Ricketts. And given that the outcomes of Ricketts and death from respiratory infections are more common in populations that have adverse social determinants of health, the panel felt that this recommendation would likely increase health equity. We did have uncertainty about the optimal regimen. There's no evidence basis for recommending any particular dosing strategy to these guidelines. Good morning, everyone. My name is Robert Scragg from the University of Auckland in New Zealand, and I will be presenting the recommendations for adults, studies around adults. There are three age groups here, adults age less than 50, those age 50 to 74, and those age 75 and above. So, firstly, looking at the adults age less than 50, the two recommendations there relating to supplementation and whether we should actually screen for deficiency based on the 25-hydroxyvitamin D test. Firstly, in the general adult population younger than 50, we suggest against empiric vitamin D supplementation. And you can see that there's one cross there in the four circles. This is a recommendation made with very low certainty. And similarly, in the second recommendation, in this population age less than 50, we suggest against routine vitamin D testing for vitamin D deficiency. And again, this recommendation is made with very low certainty. The three outcomes that we looked at for this age group were bone density, which I'll show you in a moment, but the other outcomes were respiratory tract infection and amelioration of fatigue. For respiratory tract infection, there were two studies that had over 17,000 participants, follow-up of six to 18 months. Mean baseline level in one of the studies was 29 nanograms per milliliter. And the vitamin D dose given in these two studies was 100,000 IU in one study per month, and 400 IU daily in the other. For the amelioration of fatigue, there was only one study that could be found, excuse me, with 120 participants. This follow-up was a month only. The mean baseline level was 13 nanograms per milliliter, and they were given a single vitamin D dose of 100,000 IU. So looking at the respiratory infection outcome, you can see the risk ratio there is only 1.02, very close to 1 and the confidence intervals are bound 1, so it's not a significant finding. And similarly with fatigue amelioration, the risk ratio is slightly elevated at 1.49. You can see that confidence intervals are both above 1, but the numbers are very small. And also, if you look at the change in the mean score of fatigue assessment, where the maximum score is 50, vitamin D decreased by only about three units, and for the placebo group, about one unit. So these moderate changes, the panel felt, were not convincing in terms to make a recommendation. Excuse me. Looking at bone mineral density, the range in these studies was 400 IU per day, up to 40,000 IU per week, and the follow-up in some of these trials was less than one year. So that is, I think, a weakness in some of these studies. There were four parts of the body that were scanned, the lumbar spine, total hip, femoral neck, and also the tibia. You can see the number of studies is quite small and the number of participants is quite small. And all of the results where the main difference comparing the vitamin D with placebo, all of them are not significant, except for total hip. You can see a p-value there of less than .01, but it's opposite to what you would expect. The vitamin D status, sorry, the bone mineral density in those who were given vitamin D is actually lower than in the placebo group. In terms of adverse events, there was no evidence of increased nephrolithiasis, symptomatic hypercalcemia, or kidney disease. So the justification for our decision, the panel recommended against routine vitamin D supplementation above what is required to meet dietary reference guidelines because the identified trials did not show a benefit from vitamin D supplementation. It also recommended against routine testing for 25-hydroxyvitamin D because of the absence of large trials. Also, the variable feasibility to test could actually decrease health equity with some parts of the population having access and others not to testing. And, however, the panel accepted that testing to identify people with low vitamin D status may be required in special populations such as those with malabsorption, increased vitamin D catabolism, or increased renal loss. Turning to the second age group in adults, so those aged 50 to 74 years, again, there were two recommendations here. In this age group, we suggest against routine vitamin D supplementation, and we make this with moderate certainty. So you can see the three crosses there in the four circles. And in terms of screening in the general population ages 50 to 74, we suggest against routine vitamin D testing, 25-hydroxyvitamin D testing. This was made with very low certainty, and so there's only one cross there in those circles. The outcomes that was examined in this age group were all-cause mortality, fractures, cancer, and cardiovascular disease. In terms of average study baseline, the median level was 24 nanograms per milliliter in these studies. The range was from 13 up to 33 nanograms per milliliter. In terms of the vitamin D dose, the median dose was 2,000 IU per day. The range was from 300 to 4,800. And participants in these studies on top of that were allowed to take 4 to 800 IU per day in terms of the studies. This slide summarizes all the outcomes that were actually looked at for this age group. You can see there's quite a substantial number of trials for this age group. And for most of the outcomes, there are more than 80,000 participants. And so that provided very high certainty for all of the outcomes that were looked at in this age group. And if you look down the column with risk ratio, you can see most of the risk ratios are very close to one. And for all of them, the 95% confidence intervals encompass the reference value of one. So none of these results are significant. And this includes mortality, fractures, cancer, cardiovascular disease, and also the adverse events were not significantly elevated. In terms of those studies where there was evidence of participants receiving vitamin D below a threshold, the threshold in these studies is 20 to 24 nanograms per milliliter. And you can see the number of trials is actually very small. Although some of them have quite a number of large number of participants. Again, when you look down the column for risk ratio, all of the results there are not significant. The confidence intervals include the reference value for one. So in terms of the decision, the justification for it by the panel had judged that the anticipated desirable and undesirable effects from vitamin D are likely to be trivial in healthy populations age 50 to 74 years. Most of the trials were completed in populations that did not have a low vitamin D status at baseline, which is a limitation. But the nutritional supplements may be required in some individuals to meet the recommended daily allowance of 600 IU per day. Also, in terms of the recommendation for not doing routine 25-hydroxyvitamin D testing, there is insufficient evidence of a benefit limited to those with very low levels below this threshold of 20 nanograms per milliliter. And also, there was concern that variable feasibility could decrease health equity in the population. Now, this is the second case that we're asking you to check on your phones there. This is a 78-year-old man who attends his doctor for a checkup following a recent hospital discharge after a myocardial infarction. He has a past history of elevated lipids and blood pressure. Both were managed by appropriate medications. He has a family history of heart disease, so he's at high risk of having a heart attack, and that's motivated him to stop smoking about 20 years ago. He's very fit. He exercises regularly, and he's very careful about his diet. However, since his heart attack, he searched the internet and found substantial evidence that vitamin D can protect against cardiovascular disease. And this patient would like your advice about whether he should take vitamin D supplements. So what would you advise or do for him? And so the options are recommend vitamin D supplementation, measure 25-hydroxyvitamin D, and then decide whether to supplement, advise increased but safe sun exposure, and last option, D, none of the above. So what's it looking like? All right, okay. Well, that's very interesting. Okay, we'll move on to the next slide. So here we do have a conditional recommendation. So this is looking at the age group 75 years and above. In the general population in this age group, we suggest empiric vitamin D supplementation because of the potential to lower the risk of mortality. And you can see this recommendation was made with moderate certainty with the three pluses there. And in terms of screening, we suggest against routine testing for 25-hydroxyvitamin D, and that's mainly because there is very limited evidence, very limited certainty about this recommendation. So focusing on all-cause mortality where there is a benefit, there were 25 trials with nearly 50,000 participants. These trials were carried out in community settings, nursing homes, and also in hospital clinics. In terms of the vitamin D supplement that was given in these trials, it was mainly D3. Thirteen trials gave a daily dose, plus or minus calcium. The dose range was from 400 to 3,333 IU, daily equivalent. And the weighted average dose was about 900 international units per day. Participants were allowed to remain on their existing supplements. And the follow-up of these studies varied from 12 weeks to seven years with a median of two years. So looking at all-cause mortality, I know you won't be able to see the individual studies in this forest plot, but I'm just showing it so that you see the depth of the evidence here. The summary value is down at the bottom where the red arrow is. So I've taken that up to the top to show you on the top right there. The risk ratio for all-cause mortality was .96, so a 4% reduction in mortality in those who received vitamin D supplementation. And you can see that the confidence intervals just touch one, the upper limit. In terms of an absolute effect, it was six fewer deaths per 1,000 treated. And the range here was from 11 fewer per 1,000 up to no more, zero more. And this effect was not modified by the risk of bias in the studies, the gender of the participants, calcium co-administration, vitamin D dose or trial setting. Looking at the other outcomes, there was no evidence of a benefit. So you can see here with fractures and falls, there were reasonable numbers of participants in these trials of 43,000 and 12,000 respectively. And you can see the risk ratios are not significantly different from one. And respiratory infections, we only found one trial that had 821 participants. Again, the risk ratio is not different to one. But there's low certainty because of the small number of participants. In terms of adverse events, the risk ratios are slightly reduced, but not significantly so. The key thing in terms of harm is that they were not increased. So looking at studies where there was evidence of the effect of vitamin D below a threshold, here the threshold was 20 nanograms per milliliter. There were three trials with all-cause mortality, but only 589 participants. The risk ratio was not significantly reduced. With falls, there were two trials of 276 participants. Again, the risk ratio was not significant. And adverse events, only one trial, 196 participants, and no increase in the risk of those adverse events. So this is my last slide. The justification for our decision was prioritized on the mortality outcome. The evidence for vitamin D supplementation suggests a small but important mortality benefit, and minimal to no harms. Vitamin D supplementation is inexpensive, may be cost-effective, and increase health equity. In terms of routine 25-hydroxyvitamin D testing, this was not recommended. And the last two bullet points apply to testing. There was insufficient evidence of a benefit, and the extra cost from testing could decrease feasibility and also decrease health equity. Thank you. I'm going to present from University College Cork in Ireland, and I'm going to present the pregnant. It's just loaded there, okay, excellent. So we've got another poll for you. It's a 35-year-old patient attending for her antenatal checkup at about 16 weeks of gestation. She has normal 12-week ultrasound bloods and urinalysis. Her BMI is 28. Her blood pressure is 125 over 83, and she reports stopping her prenatal folic acid when she reached 12 weeks of pregnancy. Would you recommend vitamin D supplementation for this patient? Very interesting. Thank you. These polls are great. So our recommendations are as follows. We suggest empiric vitamin D supplementation during pregnancy given its potential to lower risk of preeclampsia, intrauterine mortality, preterm birth, small for gestational age, birth and neonatal mortality. We're giving this recommendation with a low certainty of evidence. During pregnancy we suggest against routine 25-hydroxy D testing with very low certainty of evidence. So these are conditional recommendations. So the outcomes that we examined were preeclampsia, all types, offspring mortality, which would be intrauterine mortality, past 20 weeks, neonatal and infant mortality, preterm birth, small for gestational age, and maternal adverse events. So the scope of the problem is that the fetal dependence on maternal vitamin D status has been reported with the last 40 years. And this little graph shows you core blood 25-hydroxy D, maternal 25-hydroxy D, and where this is data from a trial that we did. And I've put the line here where you can see where mothers had 25-hydroxy D above a certain threshold. Their babies were born with 25-hydroxy D above a certain threshold. There's a high prevalence of low vitamin D status and we haven't been defining thresholds in this guideline as you can see. So when I'm talking about low vitamin D status from a global perspective, I'm talking about serum 25-hydroxy D below 12 to 20 nanograms per ml among women and girls throughout the world and particularly during pregnancy. Maternal 25-hydroxy D below 10 to 12 nanograms per ml is associated with neonatal hypocalcemic seizures, cardiomyopathy, and neonatal rickets. And this is happening around the world. And maternal 25-hydroxy D below in and around 20 nanograms per ml is associated with hypertensive disorders of pregnancy, fetal growth restriction, SGA, and preterm birth. There are a couple of comments on the evidence base required. These recommendations apply to generally healthy individuals and we restricted the PICO to or the question that we asked where we are comparing vitamin D with of any dose versus no vitamin D. And this is really important from the pregnancy perspective because it has excluded a huge number of trials because any studies that provided any vitamin D to the control group were ineligible. The doses in the studies that we encountered ranged from 600 IUs to 5,000 IUs daily and the estimated weighted daily average was 2,500 IUs so in and around 63 micrograms per day. Of all of the studies that we evaluated, 10 RCTs met the inclusion criteria. And as you can see most of these studies bar Roth from 2018, and I think I need to move my cursor over here for to emphasize certain things, bar Dan Roth's study in Bangladesh in 2018, most of these studies are very small sample sizes. His was the largest sample size at almost 1,300 individuals and of these studies not one has been carried out in the United States. So I think it's very important that the evidence base here is a little bit different to what you might expect. If we look at the outcomes in preeclampsia and gestational hypertension, we had a sample size of 2,674 to consider. The effect, again, as with many of the others that you've seen this morning, the relative risk crossed one and the absolute point estimates were 23 fewer per thousand, 46 fewer to 13 more. The certainty here is low. Moving on to interuterine mortality, we had still a reasonably substantial sample size where you have, again, a relative risk of 0.7 but crossing one, 6 fewer per thousand, 13 fewer to 9 more and the certainty is moderate. Neonatal mortality is quite similar where you have 8 fewer per thousand, 14 fewer to 9 more and the certainty is moderate. And preterm birth with over 2,000 to consider, 28 fewer per thousand, 62 fewer to 37 more. Certainty was low. Small for gestational age, the certainty was low. And so you can see that there is low certainty of evidence and this is absolutely consistent but that the direction of the evidence is also consistent. Adverse effects were rare but we have to say that adverse effects and serious adverse effects were usually not collected prospectively with the exception of the Roth 2018 study in Bangladesh. Our pre-specified sensitivity analyses, such as risk or bias or dose, made little difference to these outcomes. Of the seven trials that reported baseline 25-hydroxydeconcentrations, it's very important to note that mean values were below 12 nanograms per ml or 13 nanomoles per liter in four of those. The others simply did not analyze 25-hydroxide. Doses for the outcomes were in and around 65 to 85 micrograms or 2600 to 3400 IUs per day. Only one out of the ten trials were what we might consider to be adequately powered. The specification of the clinical outcomes was not standardized and this was quite difficult because there was such variety in the specification of the outcomes and 25-hydroxide analysis was either absent entirely or poorly conducted and described. Even saying that though, it was our objective to derive some conclusions and the systematic reviews delivered point estimates that favored some benefit with empiric vitamin D for all selected outcomes and low to moderate certainty of evidence. But it is important to note that the 95% confidence intervals did include the potential for harm. But with little evidence for harm and some evidence of benefit, particularly for very important outcomes, the panel has issued this conditional recommendation favoring empiric vitamin D supplementation during pregnancy. The research recommendation is to focus on high-risk pregnancies and to focus on women living with obesity, particularly with dark complexion. There is a contrast between this guideline which recommends vitamin D supplementation and the WHO 2016 and revised 2020 guidelines which do not. And it's important to note this and to consider the reasons why. One of the reasons, although the WHO 2020 guideline is largely based on the updated systematic review by Palacios et al. in 2019, there are some differences, points of difference between the systematic reviews conducted here and Palacios as well as Deregil. One of those important differences was that Palacios included studies that use co-supplementation of vitamin D with other nutrients. The other important point of difference was that although Palacios was able to reference Roth et al. 2018 in the document, they were not able to include those data in the systematic review synthesis. So they didn't have access to that particular study which contributed at least 45% of our samples. Thank you. All right, just a few more recommendations. We left the best for last. Actually, I'm just kidding. The best part of this presentation will be the Q&A and you don't want to miss that, so don't go anywhere, okay? All right, I'm going to go over the last 98 slides again, just kidding. Okay, so recommendations for special considerations, and we're going to start with a case. So John is a 56-year-old black man with a hemoglobin A1C of 6.2 and a fasting glucose of 118. Two years ago, his numbers were just a little bit lower, but it seems like his prediabetes is progressing. He follows the Mediterranean diet and exercises for at least 150 minutes every week. He takes a multivitamin supplement that includes 400 units of vitamin D3. His BMI of 27 has remained unchanged for years. Okay, so now you're going to look at your phone, but don't look at the screen until you answer the question. The choices are encourage him to reduce his caloric intake and exercise more. You recommend vitamin D at doses higher than the RDA. You measure the vitamin D level and then decide based on result. And of course, if everything else fails, you start Ozempic. All right, thank you. Okay, so this recommendation refers to adults with pre-diabetes. So we suggest that in addition to lifestyle modification, we suggest empiric vitamin D supplementation to reduce the risk of progression from pre-diabetes to diabetes. This is a recommendation in addition, not in lieu of lifestyle modification, so I may have to run this question again. All right, for this recommendation, the key outcome was, of course, new onset diabetes, but we also examined additional non-diabetes, well not instant diabetes related outcomes, respiratory infections, fractures, all-cause mortality, cardiovascular disease, and in addition to the pre-specified adverse events that we examined for all outcomes. In terms of the non-diabetes related outcomes, we looked at respiratory infections, fractures, all-cause mortality, and cardiovascular events, and there was very limited evidence to support vitamin D to reduce the risk of these conditions in this particular population. The evidence, as you can see here, it was either very low or low. In terms of the pre-specified adverse events, we did not find any evidence that vitamin D increases the, again, the pre-specified adverse events. In the D2D study, which was the largest vitamin D and diabetes prevention trial, adverse events overall were less frequent with vitamin D than placebo. In this case, 4,000 units versus placebo. The commissioned systematic review identified 10 trials that used vitamin D2 or D3 and reported instant diabetes as the primary outcome. The evidence synthesis showed that vitamin D reduces the risk of progression to diabetes by 10% and the certainty of evidence was moderate. There are now three other systematic reviews that have confirmed this finding. The panel, in addition to the data, the evidence synthesis generated by the commissioned systematic review, we were also allowed to look at other systematic reviews that may add additional insight. And so we also examined an individual participant data meta-analysis that combined participant data from the three diabetes vitamin D trials that were specifically designed for prevention of diabetes. This was published last year, and depending on the type of analysis, the summary result was that vitamin D reduces the risk of diabetes by 12 to 17%, again, depending on the type of analysis. So although the effect of 10 to 17% may appear modest, the panel prioritized the key outcome in this question, and even though, again, the effect may be modest, they judged this to be significant if applied to the very large population with prediabetes. We also judged that, you know, in contrast to other interventions for diabetes prevention, which may be costly or burdensome, we judged that the vitamin D will, given its low-cost acceptability to populations, will likely favorable health equity. The panel also recommended some future research, and there were two points here. One is a trial that would evaluate a treat to target strategy to define the level that optimally reduces the risk of new onset diabetes and also increases time spent in normal glycemia. And also there is some observational data. We didn't cover this in our guideline, but in that suggests that vitamin D may have a role in type 1 diabetes, and so we recommend studies in these trials in this arena. Okay, so the next recommendation refers to dosing of vitamin D. So if vitamin D is indicated, then we suggest daily lower dose of vitamin D instead of non-daily higher dose of vitamin D. And this was the evidence of certainty here was low. We examined fractures, false respiratory infections, and all-cause mortality in addition to the pre-specified adverse events to answer this question. Initially, our inclusion criteria were very tight, and we only found two trials that compare high dose with kind of regular dose or lower dose. And eventually, we expanded the inclusion criteria, and then we found 15 studies and 19 publications that included high dose supplementation. So we found that studies that gave higher than 100,000 units per day, there may be an increased risk of fracture compared to those that gave a lower dose. And we also found a trend towards higher risk of falls for studies that gave, again, doses higher than 100,000. And we also found some evidence that dosing more frequent or less frequent than 12 weeks may also increase the risk of falls. There were five trials that reported on respiratory infections, and there were no difference between high dose and lower dose. We did not find any data on mortality, and we also did not find any evidence that a higher dose versus lower dose will increase the risk of, again, the pre-specified adverse events. So we judged that any, you know, desirable effects of intermittent high dose, for example, you know, the convenience of taking something once a week or once a month or once a year, are likely outweighed by the potentially undesirable effects that are also, you know, small. And we didn't find any studies on cost-effectiveness, acceptability, feasibility, or effect on equity. All right, the last three recommendations refer to screening with 25-hydroxyvitamin D. And we suggest against routine screening for 25-hydroxyvitamin D in healthy adults, and that includes adults with dark complexion and adults with obesity. Here, I'd like to make the point that this is for healthy populations, again, not in individuals that would otherwise testing would be, you know, indicated, as was noted earlier. For these outcomes, our questions, we looked at the following outcomes for healthy adults, dark complexion, and obesity, and I will not describe, I will not go all of them, but again, similar outcomes to what we've looked at in the previous questions. For healthy, there were, for these three questions, the screening questions, we did not find any trials that test screening versus no screening. And, you know, we're not, we were not surprised. We looked for it. We didn't find any. So what we did was we addressed the screening questions following a published framework that considers several important factors when making a decision about screening. So, you know, cost, feasibility, acceptability, burden, things like that. The question that relates to screening in healthy adults, we used the data that Dr. Scruggs presented, the questions three, five, and seven that address the benefits and harms of vitamin D in adults with a 25-hydroxyconcentration below a threshold. And as, you know, was described before, the clinical trial data were insufficient to assess whether the benefit of vitamin D varies according to someone's baseline 25-hydroxy D. And here I'd like to make a couple of kind of conceptual points. So they're actually very, most of this evidence came from subgroup analysis where the large trials will give you subgroup analysis by level. But there were very few of those. And subgroup analysis are never powered to really show anything. But when we did see that evidence, we tried to combine, as long as it made sense, to see if there is any benefit. We also refrained from combining aggregate data from trials that may have had a higher or lower mean baseline vitamin D level because these are types of analysis are subject to ecological fallacy. And at the end of the day, they don't really make any sense. In terms of screening with 25-hydroxy D in people with dark complexion, we found very, very few subgroup analysis. And in terms of fractures, all-cause mortality, cancer, cardiovascular events, and respiratory infections, we found either no data or very few trials. And there was no clear difference in terms of the effect of vitamin D in people with dark complexion or not. Similarly, we found very little evidence that vitamin D has a differential effect in people with obesity versus people without obesity. Again, this is specific to the outcomes we examined. We cannot examine all outcomes. So in short, in terms of the screening, the three screening questions, the evidence did not suggest a different benefit-risk profile with vitamin D based on a threshold 25-hydroxy D and also no clear effect modification by dark skin, dark complexion, or obesity. And even though testing is acceptable and easy, relatively easy, it's a blood test, we consider cost and burden in making our decision. And also, we judged that such a recommendation would have a negative effect on equity. So in the absence of a clear benefit-risk ratio, we suggested against screening. Great, thank you. So I'm going to briefly talk about the limitations and the path forward. So here is a summary. Empiric vitamin D supplementation in the pediatric age group and those over 75, as well as in pregnancy and prediabetes. Now, there were a lot of limitations that we've encountered. One of the major limitations is paucity of randomized clinical trials addressing the benefit and safety of vitamin D supplementation in populations with low baseline 25-D levels. Many of the recent large clinical trials had baseline levels of 30 or 25, and we don't know whether that is sufficient for the outcome or not. The inability to design trials with a true placebo group. This is not like we're giving an antibiotic. So everybody has some exposure to vitamin D, so we can't have a zero exposure group. And the question is, how much vitamin D do people need to be exposed to to prevent certain outcomes? Many of the trials that we looked at had available to us had very inconsistent doses or administration schedules, which will be summarized in the pocket guide that you should have grabbed on the way in. Some of them were 800 units a day. Some of them were 100,000 units a month. Many of the trials, as was alluded to by Dr. Scragg, were of insufficient duration to address outcomes. So for changes in bone density, trials that are really less than a year may not be long enough. For respiratory infections, certainly it's an acute condition, acute onset, acute resolution. For malignancies and autoimmune diseases, exposure may have to occur over a period of decades. So the trials really need to be designed, thinking of the natural history of the disease outcome, empowering them to really examine that. A lot of the trials, all the trials were performed in healthy populations. So these trials really focus on disease prevention, not treatment outcomes. Most of the trials were in those of European ancestry. The few trials that looked at dark complexion were, the dark complexion was designed as black or African American. Benefits may actually defend on other factors, such as calcium for skeletal outcomes. And the target 25D level for skeletal outcomes, while it depends on calcium, may be very different than the target 25 degree, 25D level that is needed to be achieved for perhaps autoimmune disease prevention or other outcomes. And then not outcomes could be examined in all populations, basically because of the limitations of resources of the systematic review. So where do we go from here? The panel felt that we needed adequately powered trials with appropriate controls and of sufficient duration, considering the natural history of the outcomes being examined. The trials should be performed in populations with low baseline D levels, maybe a variable, a number of low baseline D levels, targeting specific D levels so that we can understand if there's a different optimal target D level for different outcomes that we're interested in. And we desperately need clinical trials in those of different racial and ethnic groups, people with different body composition and different environmental lifestyle and dietary factors. Thank you for your attention, and we'll open it to questions. Thank you. Great. So if you have questions, please come to the microphones that are placed in the aisle. We have two questions, one on sarcoid that we're not going to answer because it's an unhealthy population, and we really wanted to focus on healthy populations. The second question has to do with the potential mechanism of vitamin D supplementation to improve the risk of type 2 diabetes. So Pataso, if you'd like to comment on that. Yeah, the mechanism for this is most likely at the pancreatic beta cell level. There's quite, quite, quite a lot of basic science for the last 20 or 30 years that vitamin D has a healthy effect on the beta cell. The beta cell has vitamin D receptors. The beta cell has the ability to convert 25 to 125 by drugs of vitamin D, so it doesn't depend on the periphery. So, yeah, so the effect is thought to be at the pancreatic beta cell. Great. Thank you. Go ahead, microphone one. Simeon Taylor, University of Maryland School of Medicine. One of the questions I have is, while I do agree with you that measuring 25-hydroxyvitamin D may not be all that useful most of the time, I wonder whether there are other things one could measure that would be more useful. Back when I was a medical student about 50 years ago, people focused with thyroid hormone on measuring total levels of thyroxin, and people eventually became aware of the fact that variations among individuals with respect to their levels of thyroid binding globulin made those data almost uninterpretable. And this approach was eventually supplanted first by measuring free levels of thyroxin and eventually measuring TSH levels as a reflection of whether the wisdom of the body thought that there was a deficiency or too much thyroid hormone. The literature has long drawn analogies between vitamin D and thyroxin with respect to the impact of vitamin D binding protein and albumin binding, and there's also very substantial, perhaps threefold variation in the levels of vitamin D binding protein. So I find myself wondering whether it might eventually be more useful to have indices of vitamin D status that are independent of these inter-individual variations in vitamin D binding protein. So again, the question is, there are some patients, and I think you implicitly acknowledge it, or even explicitly, such as patients with chronic kidney disease or malabsorption, in whom it would be useful to be able to assess vitamin D status. And I do think the recommendations would be strengthened if at least there was some reference to the fact that there could be better ways of measuring this, even if the data to know how to interpret them is not yet available. Thank you for that question. There are research availability of free vitamin D levels that hasn't been found to be tremendously clinically useful in the general healthy population. Our recommendations don't really apply to those with nephrotic syndrome or chronic kidney disease. The one thing that you mentioned is the TSH, which is really the body's responsiveness. We do have a biomarker like that for intestinal calcium absorption in secondary hyperparathyroidism. There is no biomarker for perhaps the effect of vitamin D on prevention of autoimmune disease. So that's something that really needs to be developed in association with large-scale clinical trials. The other thing to emphasize is even if we have an accurate assessment of a vitamin D level, we do not know the optimal target level for skeletal health versus autoimmune disease, cardiovascular disease, et cetera. Thank you. Sameer. Next. Dr. Abdulaziz, Debut Hospital Qatar. So my question is, those recommendations, has they been looked at the PTH or any bone biomarkers, for example, because we know for people with subclinical hypothyroidism, we know if there is a cutoff of a TSH level above a certain level, then there is a clear recommendation that we need to treat. So we have a lot of our patients that usually have their vitamin D checked, and it's a little bit in the lower side, and if the PTH is high, is there still a recommendation that there is no need to replace the vitamin D in those situation? So Chris. Yeah, thank you for that question. This is an example of the difference between a surrogate marker and a patient important outcome. So for example, if PTH is high and you can lower it with vitamin D, how does that translate into fracture risk reduction, which patients really care about? Well, that's much more uncertain. You've got this hypothetical causal chain, but the farther you are from the outcome of real importance to patients, the more uncertainty you have. So that's point number one. I think the other thing, just as an example to highlight this for, in particular with the idea of subclinical hypothyroidism, just because the TSH is high does not mean they will benefit materially from thyroid hormone administration. So that's another example how simply because you have a compensatory response doesn't mean treatment of that's going to help people in a meaningful way. Yeah, and I think the PTH levels vary by a lot of different factors as well. So it makes it even more difficult. Next. Hi. Thank you. Raquel Farage from Mexico. I have a question regarding the obesity population. Many of the obese patients will have prediabetes, and those are the ones that will benefit from the vitamin D treatment to prevent diabetes. So why is it that you, I understand there's no data, but doesn't it seem like the obese people should be screened for prediabetes and then screened for vitamin D? Why don't do both things at the same time since it's harder to get blood draws many different times? Yeah. So thank you for that question that makes me repeat the key point that the recommendations depend on specific populations and specific outcomes. So if you have a patient with, or a person with a BMI of 31, you don't need to check the vitamin D level because there's no evidence that just doing that and replacing will, but if you have a patient with prediabetes, regardless of whether they're overweight or not, then the evidence is in favor of, so the population and the outcome is specific. Now, if you want to check the vitamin D, define the glycemic status and check the level at the same time, that's more like a convenience clinical approach, but we recommend first and then the second. Just a comment. Go ahead. Can you comment on the difference between the ECE guidelines and your guidelines? Because they seem to be very different, I think. Which guidelines? Which guidelines? The European Society of Endocrinology, ESC. Oh, okay. Which, I mean, I don't know, I know that they were presented last year in Turkey. I don't know if they were guidelines or just a talk, but they were very, very, very different to what you're saying here about treatment and screening and stuff. I would like to say I think it's based on the methodology that's used, that this is really based on systematic reviews of randomized clinical trials, and to make it as rigorous as possible, and really to define, yes, we know this, there's evidence for this, or there's no evidence for this, so we can't recommend it. Okay. Thank you. Thank you. Next. Hi. Marcy Shafetz here from Atrius Health in Boston. I have a few questions. My first question was if you could comment on the finding of benefit in terms of respiratory infection prevention in pediatric population, and acknowledging the paucity of data in adults, why there may not have been the finding of a similar result in the adult population. So who wants to take the for upper respiratory infection step? Yeah, so, I mean, we don't have evidence from the systematic review, and there have been other systematic reviews, there have been other meta-analyses, so notably the Joliff et al. meta-analysis from a few years ago, which also demonstrated that overall, children seem to benefit more from vitamin D supplementation in terms of prevention of respiratory infection than from adults, than adults did, and it may have to do with something about the training of the immune system, and particularly those immune cells that are antigen-presenting cells that are thought to be vitamin D responsive, but I don't think we fully know the mechanism of that. Do you anticipate with future studies that it's possible a benefit might be uncovered, or do you think that's unlikely? We have to wait for those trials. I'm just curious. Okay, thanks. The other question was as a corollary to someone's prior question about the benefit for prediabetes, what about the mortality benefit for the older population? What might the mechanism be for that? Great question. Bob? Robert, yeah. Hi. I'm really not sure what the explanation might be. I mean, the effect is so small, and it might be that there are small cumulative effects occurring across a range of diseases that can't be picked up individually for those diseases, but when you combine everything together, you see a small benefit, but that's a good point, why it just shows up mortality, but not some other chronic diseases that are quite common. That's the evidence that we found, and I've certainly been in the school pushing for vitamin D supplementation, and I've had to rethink things, having been part of this process. Thank you, Robert. An additional question is about the autoimmune disease prevention in light of some, like the vital trial findings. I'm just curious about any comments about that. So, in the vital trial, it was a secondary analysis. The vital trial had people with levels, baseline levels of 30, and I believe there was a more recent paper by Kost and Bader's group showing that vitamin D after two years, after vital did not, was the omega-3. So what this really emphasizes is that the secondary analyses are often not powered to give us, they're hypothesis generating, let's say that, rather than the randomized clinical trials. Okay. Thank you. That's it. Okay. Okay. Suzanne, you're next. No, not by the way. I'm Suzanne Jondeber from Charlottesville. Thank you to the committee for this herculean effort. I know it's really a big process, so thank you. I just wanted to clarify that people with osteoporosis or people with fragility fractures are not included in these recommendations. Is that true? That's true. Correct. Great. And I know this is going on a limb, so you can kick me off with a cane if you want, but are there guidelines for people with osteoporosis or fragility fractures that you use or you would recommend about vitamin D supplementation? So there are no current guidelines based on randomized clinical trials, and they would need to clearly involve calcium recommendations as well, because as you know, part of the main problem, main issue with vitamin D in skeletal health is providing an environment for mineralization and calcium. Thank you all. Thank you. Next. Hi. Jackie Bates from the University of Toronto. So from the expert panel, and these are recommendations, guidelines that clinicians will be looking towards to really advise their patients, there's a great deal of vagueness around the dosing. Everybody in the room would love to know what you mean by low-dose vitamin D supplementation. And obviously, based on the evidence, the ranges of doses are huge. So left to the clinician to decide when the patient is standing in front of them, doctor, if you suggest that I take a vitamin D supplement, how much? It would be very helpful for the wisdom of the panel. Thanks. Yes. So two-part answer to this question. One is the recommendation is against high intermittent doses, and high was defined by anything over 50,000. The second part to your important question is the reason why we didn't give a specific recommended intake is because there is none. In each recommendation in the document, we have the range of the vitamin D doses that were used for each trial, and then we also provide a mean weighted intake. So for example, I have here, because I knew this question was coming, for pre-diabetes, it was 800 to 7,000, the range, and the weighted mean average was 3,500. So if you want an answer, you know, 3,500 might be your answer. But we were not committed to a specific number because of the range. Great. Thanks. Yeah, that's good because obviously it's one thing to say don't test, but people are getting tested. So if you do have a low number, what are we shooting for, 20, 30? The other question is the RD requirements, I see 600 to 800. In my experience, they're not useful for raising that vitamin D level to the normal range, which I say 30. And also the third is recommending everyone to take a supplement for vitamin D. It's not inexpensive, in my opinion, and maybe for the ones that are fish-based, but like if I'm vegetarian, it costs me $20 for a month supply to get a vegetarian vitamin D supplement. So again, those are not practical either. Next question. Great. Excuse me. Okay. We deliberately avoided the question of thresholds because there is not data sufficient to address that question in the research base that we had available to us. And the other thing that's really important is that we're referring to supplementation over and above the RDA, in addition to the RDA. So the RDA is established on the basis of achieving 20 nanograms per mil, and those were the dose response data that were available at that time, and that was 2010 when that modeling was being done. So you're saying 600 plus 600 to 1,200 is what you're saying? So what we're saying is that the doses are specified in the guideline document on the basis of the research that was available for specific outcomes. So for instance, for the pregnancy outcome, the range was 2,600 to 3,400. So if you wanted to go with something like 3,000 as a midpoint of the range for pregnancy outcomes, so long as the individual patient is staying within the tolerable upper intake level, which is 4,000 IUs per day. So I think what we've been is very clear around no screening, treat where the treatment is indicated, use the doses for specific outcomes where these have been provided, and an acknowledgement that the supplementation is continuous and that we're not recommending high dose single use doses. Okay. Great. That's all clear. Thank you. Yes. Yes, sir. Hi. Thank you very much for the recommendations. I'm Juan Pablo from the University of Chile. I have two questions, actually. The first one regarding the- We'll have to limit you to one. We have five minutes left. I will be very quick in my- Okay. Regarding the pediatrics recommendation, in the guideline it says that it's from one to 18 years, but it seems like most of the evidence comes from early infant studies. So I was wondering why did you choose to extend the empirical universal treatment within considering that there is more evidence for the early infants? And just the second one is, is there any outcome considering mental health as an outcome that were studied by the commission? I guess there's low evidence, but it was just a curiosity. Thank you. Okay. Who wants to take the dab? Oh, good. Yeah. So in terms, I do agree that in terms of the outcomes of both rickets and respiratory infections, the majority of evidence was in the younger children. In terms of the respiratory infection outcome, we did look whether there was variation in risk when we subdivided the group into those who are younger than five and those who are five and older. We didn't see any variation, and that was the basis for our recommendation of just universal supplementation in that age group. And we just, we didn't have the evidence to basis to break it down into smaller subgroups. And nothing on mental health? Yes. About that, no. Okay. Thank you. Thank you. Yes, sir. Ranganath Muniappa from the NIH. Thank you all for your efforts. I have an important question. I know there is an accompanying article with these guidelines, but I want to put it in the forum here. The guideline panel recognized that the self-identified race is an inaccurate and otherwise problematic proxy for dark complexion. So you chose black race as a surrogate for it when you asked for it. If it's problematic, what was the compelling reason that you all had to ask for this particular PICO? I'm just reading from what you guys said. And I know there is an accompanying along with JCNM, so why don't we discuss it? So Chris. Thank you. Can I do it? Chris. Yep. And I, thank you, Ranganath. I really want to point everyone to this commentary. One of the audience, Dr. Nike Singh Aspina, put together a beautiful description of the problem with conflating broadly race with biology and specifically dark complexion with a given race. For example, self-identified black race. That is very common in the field, that race is used as a proxy for skin complexion. One of the big issues is that it's terribly inaccurate. The skin complexion across those who self-identify as black, it's tremendously variable. So it introduces a degree of, we talked a little bit about the ecological fallacy and the potential for misclassification bias. It's a real problem. So we initially started down that path, but with help from our other methodologists, recognized this as a problem. And so we kept that secondary analysis of self-identified black race in the document for a couple of reasons. One, because it commonly comes up. And so people, I think, our assumption is that readers would value knowing what that is. The other thing that's important is that the 2011 guideline, among their criteria for 25-hydroxyvitamin screening, was those who self-identify as black or Hispanic Latino. And which, of course, that's a tremendous number of people across the globe. So I think that information that we uncovered, that no, we don't have evidence that that kind of screening actually helps or provides a net benefit, was important for people to know. And then the last thing, the reason we kept it, is that we wanted to be transparent about exactly how we did it. And we wanted to be able to point to Dr. Espina's really important paper that I encourage you all to read. Thank you. Thank you. Thank you for bringing it up, too. Yes, ma'am. Hi. My name is Dalal from McMaster University. I think my question is quite similar to the previous person. My only comment on the recommendation for 50 to 70 years old, when we say we don't suggest routine screening for those people. And I know it is very low certainty evidence, but we may miss some people who are healthy in this age group, especially post-menopausal women with no complaints, but they come with a bone density upon their routine measurements. And we find very low vitamin D levels and high alkaline phosphatase, so osteomalacia. Are we, like, in the trials that gave this recommendation, are there specific ethnic groups that were studied in this studies that we can say these apply to this ethnic group and so that we don't miss other ethnic backgrounds with lower vitamin D levels? So easy answer to your second question. There are no trials that are based on ethnic groups. And I think that that's a huge problem, and that's something that we need going forward. Similar to the pre-diabetes question with obesity, I think it is, there are recommendations to do bone densities in post-menopausal women, and once something is found, then that person is not a healthy individual, they're at risk of fractures, and therefore need a workup and may need additional vitamin D and additional investigation. Just like an obese person who doesn't have pre-diabetes, we don't have evidence that they would benefit more than a person who's not living with obesity. So that's the issue. So if someone has an underlying condition, for sure, our guidelines don't cover them. These are really healthy individuals with no underlying conditions that are impacted by vitamin D. Thank you. And thank you to the panel for their great work over the four years. Just make sure you get your little cards, and we also provide autographs on the cards.

Prevention of Disease

Endocrine Society Clinical Practice Guidelines

Vitamin D supplementation

Population groups

Randomized clinical trials

GRADE process

Evidence-based recommendations

Observational studies

Age groups

Health equity

Cost-effectiveness

Pregnancy