I have the pleasure of introducing Dr. Caroline Davich-Pitts. She's an Associate Professor of Medicine at the Mayo Clinic in Rochester. She is the Director of the Transgender Health Program at Mayo Clinic, and she also is one of the co-directors of the Endocrine Society SIG on transgender health. And so she will be giving a talk on bone health and transgender care. Thanks so much to the organizers for inviting me, and to the chairs as well. It's really great to be here and to speak to you all. I don't have any financial disclosures. I do want to note that I do serve on the Board of Directors for the U.S. Professional Association for Transgender Health, and I also want to note that our hormone therapy that we're going to discuss today is off-label for gender dysphoria and congruence. So here is the QR code to ask questions if you're not planning to come to the mic. And our objectives today include reviewing specific skeletal considerations in transgender and gender-diverse people. We're going to discuss anticipated bone changes associated with initiation of gender-affirming hormone therapy. And then finally, we're going to review indications for assessing bone density in transgender and gender-diverse people, and how to interpret these studies. I'd like to start off with some terminology. So for those who identify as transgender, these individuals have a sex recorded at birth and gender identity that are incongruent. So for example, someone who identifies as a transgender woman will be recorded male at birth, but identifies as female or feminine. Individuals who identify as cisgender have sex recorded at birth and gender identity that are congruent. And those who are gender non-binary identify as neither male nor female, both male and female, or along or outside the gender spectrum. And not everyone has gender dysphoria, but for those who do, have clinical distress associated with this incongruency between sex recorded at birth and gender identity. So it's clear that sex steroids have a significant impact on bone, and this is particularly during puberty where we see an increase in bone density, bone width, and length. And it's also becoming more clear that both sex steroids, estradiol and testosterone, can have effects in both cisgender women and cisgender men. And depending on the dominant sex steroid, we can see that there are effects on the different compartments of bone. So for example, in cisgender women, where estradiol is going to be the predominant sex steroid, that is going to have a positive effect on both cancellous and cortical bone, with testosterone playing probably a lower role in maintenance of bone health. Whereas in transgender men, where testosterone is going to be the predominant sex steroid, this is going to have a positive effect on cancellous bone, which is typically more estradiol resistant. But the high testosterone is going to be able to aromatize to estradiol, which will have a positive effect on cortical bone, which appears to be more estradiol sensitive. So understanding this, let's look a little at the skeletal considerations specific to transgender and gender diverse individuals. So of course, the age that someone medically transitions is going to be important. So whether they initiate hormone therapy during adolescence, or whether they transition around the age of 60. Whether these individuals have a history of gonadectomy, so what would be considered a more high risk bone factor. There are certain medication considerations that we need to take. For example, are the individuals on gonadotropin releasing hormone agonists? Are they on low dose hormone therapy? And this is specific, for example, to our non-binary population. And then of course, many of our patients have medical issues that might lead them to be on adverse bone medication, such as glucocorticoids. It's also important to consider that the hormone therapy that we provide to our patients is going to change their body composition, whether it be a higher lean mass, lower lean mass, and what does that mean to skeletal loading? And then ultimately, ensuring that we are establishing risk factors for adverse skeletal health and optimizing them. And for those of you who might not know this concept, I want to introduce minority stress, in particular gender minority stress, which is not experienced by everyone, but can increase risk for poorer coping and health behaviors. And so minority stress within itself will potentially lead to increased cortisol. We see an increased risk of tobacco, drug use, and alcohol, poor diet, which can include lower calcium, lower vitamin D, and then reduced physical activity. So you can see that many of these aspects can contribute to adverse bone health. And all in all, minority stress can also increase the risk of underemployment, housing insecurity, and healthcare underutilization. So let's look at some data on bone health prior to hormone therapy. And first we're going to look at transfeminine people. So this is a study from Belgium looking at 49 trans women before initiation of gender affirming hormone therapy. And the median age of this group was 30 years. And the comparative group were cisgender men. So compared to cisgender men, this group had a significantly lower mean areal bone density at the spine, hip, and femoral neck. And when looking at QCT data at the radius and the tibia, cortical bone area was smaller, and there was lower trabecular volumetric bone density at the radius. When looking at potential confounding factors, the transgender group, in fact, had a lower serum 25-hydroxyvitamin D level, and a whopping two-thirds less than 20 nanograms per mil. And of course had higher PTH likely because of that. And then when they further looked, for example, at lean mass, the transgender group was found to have significantly lower body lean mass, lower grip strength, and lower muscle mass compared to cisgender men. And generally lower weekly sport activity. Our transmasculine group appears more reassuring. So this was the same group from Belgium who looked at 23 trans men, once again, before hormone therapy. And the comparative group here was cisgender women. So they, in fact, had similar fat, muscle mass, and strength. And similar areal bone density, trabecular and cortical volumetric bone density, and cortical bone size. During the study, they also looked at bone turnover markers, and in fact, they were similar between the trans group and cisgender women. And this was all despite the trans group having a higher rate of smoking compared to the cisgender female control group. So then we may ask, well, how does our hormone therapy influence this baseline data that I've just presented to you? So for those of you who might not be as familiar with the gender-affirming hormone therapy that we use, I thought this was a good opportunity for us to briefly review this. So typically, we will use a combination of anti-androgen and estradiol in our trans-feminine group. And so for the anti-androgen, most commonly here in the US, we use spironolactone at least a dose of 100 milligrams per day. But in Europe, where many of these studies that I'm presenting to you, they will utilize soproterone acetate, which is not available here in the US. You can see there on the list that we do sometimes use GnRH agonists as well. But due to financial barriers, this is not common here in the US. And then our estradiol, you may see very familiar routes of estradiol there if you are in any way involved in sex steroid therapy. But you may notice that our doses of estradiol are quite significantly higher than what are used in the cisgender population to achieve feminization, but also to lead to suppression of the hypothalamic gonadal axis. For our trans-masculine people, I know there's a lot of information here, but really all I'm trying to get across to everyone is that whatever testosterone options you have for your cisgender hypogonadal men, we have those options for our transgender patients as well. And so when we meet with our patients, we go through these in detail, the risks and the benefits, and we will ultimately uptitrate perhaps a little slower than what you would utilize in hypogonadal cis men, but the options are all available to us. I want to take a moment to also mention hormone therapy for transgender youth. And I want to state very clearly that this is an individualized, multidisciplinary approach. And no hormone therapy is initiated prior to puberty. And if gonadectomy is going to occur, this typically occurs over the age of 18. So for those individuals who are going to embark on medical intervention during adolescence, the earliest is going to be at TANF stage two, so right at the start of puberty. And GnRH agonists are the most commonly used puberty blockers for this group. And then ultimately, if sex steroids are going to be initiated in these individuals, it's typically after about two years of the GnRH agonist treatment, and we do not typically start earlier than the age of 14. So we've now reviewed the top part of this figure, but let's now review what the anticipated effects then are of the gender-affirming hormone therapy that I just went through with you. So if you look now at the bottom, so in transgender women, if we are providing high-dose exogenous estradiol, we are hoping that that exogenous estradiol will have positive effects on both the cancellous and cortical bone compartments. And that testosterone, because it's suppressed, is now going to be a lower key factor in bone maintenance. And in our transgender men, where we are providing high-dose exogenous testosterone, that exogenous testosterone is going to have a positive effect at cancellous bone, and then we're going to have aromatization to estradiol, which will maintain the cortical bone. So you can see that, in general, we can anticipate that gender-affirming hormone therapy is going to have a positive effect on bone health. So is that the reality? So let's start looking at some adolescent data. So this is a very interesting study of bone geometry in adolescents from Europe. And so I know this is quite a busy figure, but I will explain what you're seeing. So essentially, this study looked at subperiosteal width as well as endocortical diameter. And you can see the trans girls are on the left, the trans boys are on the right. And essentially, they followed these kids from the start of GnRH agonist therapy, then initiation of the sex steroid therapy, and then followed them two or more years later. And the top two figures are going to be when that GnRH analog was started in early puberty versus when the mid and late puberty it was initiated. And what's quite fascinating about this data is they showed that if GnRH agonist therapy is started in early puberty, the figure with the trans individuals in the blue square tended to follow the curve of their affirmed gender. However, if GnRH agonist therapy was started in mid or late puberty, they actually tended to follow the curve of their sex recorded at birth. And so when we try and look at this with respect to bone density, what's very interesting about this is that the data is actually quite similar to adults. So once again, when we look at Z-scores at the lumbar spine and femoral neck, trans boys actually had a higher baseline Z-score than trans girls. And once again, trans girls tended to sit below the population mean with respect to Z-scores. As anticipated, with GnRH analogs, the Z-scores declined. However, when gender affirming hormone therapy with sex steroids was initiated, bone density actually increased. Although, like we've seen in other studies, the Z-scores for many of the trans girls still stayed below zero. So how about some data in our adult group? Okay, so we're initiating feminizing hormone therapy. How does this do for our bone density? So we have short-term and long-term studies. So our one-year study of gender affirming hormone therapy of 230 trans women, this is really looking positive. So an increased bone density at the lumbar spine, the hip, and the femoral neck. And this is one of our only studies looking at long-term data in trans women, so looking at 10 years. So this is a group in Amsterdam, a smaller group, 102 trans women. But what's reassuring about this data is that at least at the lumbar spine, there was no significant change over the 10 years. So we're not, you know, even our patients who have been treated for many, many years, we're not seeing a decline in their bone density, at least based on this data. However, there was a recent study recently from Australia, which is telling a little bit of a different story. And this is where I think we need more data. So they actually wanted to look at bone microarchitecture, so particularly at the tibia and radius. And this was a small group, 40 trans women, who had been on hormones for at least a year. And what they found in comparison to cis male controls was that these women had total volumetric bone density that was lower, lower cortical volumetric bone density, higher cortical porosity, and reduced trabecular bone volume fraction. And so although our bone density data is looking reassuring, the microarchitecture might be telling a different story. For those of you who might be interested in what is happening with markers of bone turnover, this study looking at after 12 months of hormone therapy in 121 trans women, alkaline phosphatase, CTX, and sclerostin decreased, consistent with estradiol therapy. And it's important to think about estradiol because the 10-year data that I just mentioned actually were able to look at subgroups depending on those individuals' estradiol levels. So those who had an estradiol of 121 picograms per mil, which is right about where we aim for in our trans women, actually had a significant increase in their bone density. Whereas those who had an estradiol level that was actually closer to cis male, so 32 picograms per mil, had a significant decrease in their lumbar spine bone density. So this just really tells us that as part of our work, we need to be ensuring that our patients are in optimal estradiol therapy. And when they looked at actually was there an association between their LH suppression, was there an association between their testosterone suppression, this was not found. Let's look at our transmasculine groups. So after one year in 199 transgender men, this is also some positive bone findings. So significant increases in the total hip bone density. And in fact, what is really interesting is that the lumbar spine had the most pronounced positive effect in those who are older than 50. And that's likely because these individuals had ovarian aging. And so when testosterone was introduced at that time and there was aromatization to estradiol, they had a positive effect on their lumbar spine. And the 10-year data is actually quite similar. You know, the Z-scores increased. And this was in particularly in the older group, once again. And so in this study, that was the individuals who were older than 40. But once again, lower estradiol seemed to benefit the best when testosterone was introduced. And the bone microarchitecture in this group was much more reassuring than what I showed you in our transfeminine group. So these individuals, on one year or more of testosterone therapy, had higher cortical and trabecular thickness and higher cross-sectional area and volumetric bone density compared to cis-female controls. And then when looking at their markers of bone turnover, you can see that P1NP, alkaline phospholate, and sclerosin actually increased in these individuals. It's not really clear at this time why that happened, whether it's a direct androgen effect on bone. It is not quite clear. But interestingly, the individuals who are over 50, who I mentioned seem to have the most benefit from testosterone therapy with respect to bone health, had a profile that looked much more like the estradiol in our transfeminine group. So P1NP, CTX, and sclerosin actually decreased. And we think that that was because of the reintroduction of estradiol to the skeleton in our older group. So you may say, well, what does that mean for fracture risk? And one struggle that we have is we don't have data on how the FRAX calculator is going to benefit our group and help us make decisions about therapy. And so we were pleased to see this study come out in 2020, which utilized an Italian FRAX score called the DeFra score. And they looked at their cohort of 57 trans women, and you can see that they are a higher risk woman because they had a gonadectomy. They were older than the previous studies I mentioned, a mean age of 45. And you can see they had fairly adverse risk factors for bone health. So mostly sedentary, a high risk of smoking, 37%, and 93% were considered to have hypovitaminosis D, and they defined this as less than 30 nanograms per mil. And over half, despite having a previous gonadectomy, reported low compliance with their gender-affirming hormone therapy. And so ultimately, not surprising, 40% had lower bone mass. And those who had lower bone mass tended to be older with lower estradiol levels and lower compliance on their hormone therapy. Fortunately, despite this, no fragility fractures were noted in this group. But when they placed this group into this DeFra tool, they actually found that about 14% would be considered intermediate or high risk for a 10-year risk of fracture. And so the authors conclude that perhaps we really should be thinking about these fracture risk assessment tools so that we can perhaps institute pharmacological therapy earlier, or at the very least, be optimizing their other risk factors for bone health. So what fracture data actually do we have? And I'm afraid to say not much. So this study also, once again, out of Europe, particularly Amsterdam. So they wanted to look at fracture incidents. And how they determined this was through hospital emergency department visits. So you can imagine that this is probably not optimal. We are not capturing all fragility fractures. And there wasn't a specific way that they defined fragility fractures. But what they did was they separated the group into trans women and then trans men. And within the trans women group, those who are under 50 and those who are over 50. And the trans individuals are those in the blue bar. So you can see on the far left side that the trans women under age 50 did not have a statistically significant difference in fractures compared to the cisgender population. However, when you looked at the trans women who are age 50 or older, you can see that their figure is actually very similar to postmenopausal women. And those who had the higher fracture incidents were those who had a lower T-score of their lumbar spine and those who smoked. And I will say that the individuals who are older, majority of them had had gonadectomy. And so that might have been a risk factor as well. But the take-home point is their fracture risks seem to be similar or fracture incidents seem to be similar to postmenopausal women. And then the trans masculine group was more reassuring once again. So lower fracture risk compared to the cisgender population. So I've got a question for you quickly in our last few minutes. So when interpreting bone density results in a 40-year-old transgender woman without gonadectomy, which reference range should be used for the Z-score? A, sex recorded at birth, B, gender identity, or C, I don't know. Maybe, I was going to say it's time to get into that app, I suppose, but with our three responses we have a good grouping of all of them. Okay, so first of all, this is a question we often get, so when should I check a bone density in my transgender patient? And I will say that typically we follow cisgender guidelines for recommendations for screening bone densities. However, if you have a patient who has a number of risk factors at an earlier age, the same way you would in the cisgender group, we check a bone density. But those who you want to keep following are those, for example, who have puberty suppression, so those should be checked at least one or two years. Those particularly who have had gonadectomy and then have low-dose hormone therapy or non-adherence, and then if they have other risk factors for bone loss. And typically we'll check about every one to two years, and this is clearly outlined now in the updated ISCD guidelines from 2019. So let's look at T and Z scores. So T scores, we recommend Uniform Caucasian Female Normative Database for those who are 50 years and older. But Z scores is where it becomes a little more complicated. And so the ISCD guidelines recommended in 2019 that for Z scores we should be using the normative database that matches gender identity. And where does that come from? So it comes from data looking at Z scores, particularly at the hip and spine. And when they actually looked at the bone density compared to cis women, they found that the trans women actually matched cis women more so than cis men. And in trans men it was a little less definitive. Trans men were at least as close with respect to bone density to cis men as to cis women. And so to be affirming to our transmasculine group, it's recommended that the Z scores of the affirmed gender be used. For our non-binary individuals, this is a little less clear because these individuals might have a lower sex steroid regimen compared to our more binary trans individuals. And so it's recommended by the ISCD guidelines to use sex recorded at birth. So what you'll see is there really is more work to be done in this area. So it would be ideal if, you know, talking about consortiums, you know, getting together and finding appropriate Z scores that can be used for our transgender population without having to rely on the cisgender group. We need more data on bone microarchitecture. And at Mayo we are in the process of now analyzing some prospective data on this, and we hope to present that to you soon. But clearly there is something going on when we look at the transfeminine microarchitecture data that might not be as reassuring as what we're seeing with the bone density data. Fracture risk assessment is going to be important because right now, you know, we have adolescents that are going through puberty suppression and then subsequent sex steroid therapy. What does a long-term fracture risk mean for those individuals? And then of course for our non-binary individuals or those who might have individualized treatment plans, we don't have as clear data for that group, and it needs further investigation. So in summary, our gender-affirming hormone therapy aligns secondary sex characteristics with the affirmed gender. Multiple studies have shown that transgender women have low baseline bone density prior to hormone therapy. Hormone therapy in general seems to have positive effects on bone health in both transgender men and women. However, bone microarchitecture and fracture risk requires further study. Bone density is recommended when risk factors for bone loss are present, and when interpreting Z-scores, it is recommended to use the affirmed gender unless the person is non-binary. Non-binary people might have lower hormone-level goals, which could influence bone health considerations. So thanks very much, and we are open for questions. The podium's open for questions. Hi, nice talk. Eric Immel, Indiana University. If I was understanding, I'm sort of extrapolating a little bit, the fracture risk data, the trans women had similar, over age 50, had similar, and those were actual fractures, not a calculated fracture. Correct. Actual fractures, yeah. Similar fractures to the women who were cis women over age 50. Yes. But the women over age 50 are hormone deplete. They're not seeing estrogen, whereas your trans women probably are still getting estrogen. So does that mean that the risk is actually higher for their level of sex steroid exposure? Potentially, yes. And I think what I'm thinking is going to be happening in our trans feminine group is that there are so many confounding factors that might not be addressed in studies, but if you looked almost in all the studies I presented, the high rate of, for example, things like hypervitaminosis D. So if that was replete in all our group, would that change those fractures? The other thing is the high rate of gonadectomy in that group and potentially lower compliance with hormones. And so is that why we're seeing such a big difference, for example, compared to cis men. And the problem with that study is it's kind of all coma fractures, and so we're not looking specifically at fragility, well-defined fragility fractures. So it just really makes quite muddy, yeah, with sorting some of those things out. So I actually have a follow-up question, because in terms of hormone replacement in transgender women, does that ever change with time? Because whenever we talk about treatment guidelines, we're thinking about postmenopausal women and older men, and so I'm not sure how that's being addressed. Yeah, good point. And this is one of the big questions that we have at many of our trans health conferences is, what do you do with hormone therapy in the aging transfeminine population, knowing what potential effects estrogen can have on vasculature, et cetera? I would say it depends when that individual has transitioned. So for example, if I have someone who transitioned in their 20s and has been on estrogen therapy for 20 years, probably around the time of a natural age of menopause, I will try and mimic to some degree their sex steroid therapy. Now the problem with that can be is that if they have their gonads in place, if you drop their estrogen too low, they're going to get a rebound in their testosterone. And that can be quite non-affirming and developing kind of masculine symptoms or secondary sex characteristics again. And the other thing is, once again, if you go too low, and someone with a gonadectomy, you can risk things like further bone loss, et cetera. So it's actually one of our big question marks, is what to do in our trans aging population. But I would say probably mimicking as much as we can the cis population is where we're heading. Thank you. Can I ask a follow-up? Carolyn, great talk. You showed us data that trans women start off with lower bone density even before starting sex steroid hormones. And you mentioned gender minority stress. What stress factors do you think are most important to cause the lower bone mass before starting hormones? Yeah, my biggest concern is how I showed you the lower weekly sport activity, the lower muscle mass. You know, many of our patients can be socially isolated. So this is, you know, we're not going to get political, but this is why it's so important for our trans individuals to be involved in sport, because we want to maintain health in our trans individuals like we do in our cisgender individuals. So maintaining good bone strength and muscle strength to be able to get them out and, you know, be part of the community. So we know when people do not feel part of the community, you know, that's when they will retreat inside. And all the things that I went through, it can directly translate to the skeleton, right? So not enough sunlight, not enough exercise. And so, yeah, I think just we have to be so mindful with how we are ensuring we're integrating our trans individuals into our community, in our diverse community. Yeah. Great. Thank you. Thank you. Great presentation. So give us her talk. Thank you. Great. Thank you. Thank you to the organizers for inviting me to speak here today. It's an honor to present with fellow investigators and to the chairs. So we'll transition a bit to talking about sleep and bone health in women. I have no financial relationships to disclose, and this is the QR code for you to ask questions in the app. So we'll start today by just talking about normal sleep in circadian physiology, in part to introduce some concepts and terms that we'll need to understand the sleep-bone relationship, both the physiologic basis for a potential relationship between sleep and bone, and then mostly human observational and interventional research studies that suggest that sleep could affect bone and potentially the other way around as well. If there's time at the end, we'll briefly discuss the role of chronotherapy and osteoporosis therapeutics. So circadian, the term circadian originates from the Latin term circadian, which means about a day. And when placed in constant darkness, species show 24-hour periodicity in behavior and physiology, which is a real hallmark of the endogenous circadian rhythms. In contrast, many daily rhythms do not have 24-hour periodicity under constant conditions, and this includes metabolic rhythms that shift with changes in food intake or body temperature. So we all tell time with the master circadian clock located in the hypothalamic suprachiasmatic nucleus. Now the master circadian clock receives light and dark input to synchronize its own behavior and orchestrate cellular rhythms optimally across the day and night. So now the, let me see if my mouse pointer works, let me try this one, how about this one? Well, okay, so I'll just walk you through it. So the pacemaker, central pacemaker in the suprachiasmatic nucleus receives light input in order to entrain itself to the 24-hour rotation of the earth. It then coordinates and communicates with clock genes located in peripheral tissues via direct neural connections, the sympathetic nervous system, and hormonal signals like cortisol and melatonin, in addition to body temperature, in order to synchronize and optimize cellular processes. Now ideally, the central pacemaker located in the suprachiasmatic nucleus is aligned with the clocks in our peripheral tissues and with the external environment. However, misalignment can occur. So when we refer to external circadian alignment, that's in reference to optimal timing between the organism and its external environment. So common examples of external circadian misalignment are things like jet lag, which some of you may be experiencing, and night shift work, which others of you may have also experienced, in which you're attempting to sleep during the biological day and are awake during the biological night. External circadian alignment refers to appropriate timing between the central clock and clocks located in peripheral tissues. Because the clocks located in the peripheral tissues have different susceptibilities to environmental inputs, like food intake, they can resynchronize at a different rate than the central clock when perturbations are experienced. So this can, for example, lead to a mismatch between protein production and receptor expression. Now ideally, all of these tissues are in alignment, and that's important for overall health in order to optimize metabolic processes and, for example, match energy supply with activity demands. So the molecular circuitry of the circadian clock is encoded by an auto-regulatory 24-hour transcription loop. Clock and BMAL dimerize. They activate the transcription and translation of period and cryptochrome, which then dimerize themselves, translocate back into the nucleus, where they inhibit the clock-BMAL complex until they themselves are degraded. And this process takes approximately 24 hours. And it turns out that disorders in sleep and wakefulness and circadian disorders are common and are becoming more common. 50 to 70 million U.S. adults suffer from disorders of sleep and wakefulness, so that's about one in five people. The CDC highlighted insufficient sleep as a public health epidemic in 2014, with over a third of U.S. adults reporting sleeping less than seven hours per night, which I'm sure is not a problem in this room. And the global prevalence of sleep problems during the COVID pandemic got up to over a third of the population. Night shift work is also exceedingly common. One in four American workers work non-daytime shifts, which inherently alters sleep timing and duration. And there are a lot of changes in sleep timing and duration that occur with the aging process. This is a map describing, or a graph describing, how preferred chronotype changes with advancing age. So on the y-axis, you see late and early. That refers to people who commonly describe themselves as either night owls or early birds. And you can see that in both women in the dark black line and in men in the gray line, there's a preference with advancing age towards that early bird chronotype, with a preference for going to bed earlier and waking up earlier. There are a lot of other sleep changes across the lifespan globally, and without the pointer, it's hard to point this out, but you can appreciate that along the right side of this slide, with advancing age, there's lower overall total sleep time. There's also less slow-wave sleep with advancing age. That's that deep restorative sleep. There's increased in sleep latency, or how long it takes to fall asleep. And there's increases in wakenings after sleep onset. And women in particular are susceptible to changes, detrimental changes in sleep across the lifespan. Overall, women are more likely than men to report poor sleep. And the vulnerability to sleep disturbance in women seems to be exacerbated in certain times of the reproductive lifespan, particularly during pregnancy and menopause. Approximately 50% of menopausal women report sleep disruption related to frequent awakenings, nocturia, and vasomotor symptoms. I think it's interesting to observe how the incidence and prevalence of sleep disturbance around the menopausal transition also correlates with the time period of accelerated bone turnover and accelerated bone loss that we see in postmenopausal women with the loss of estrogen. I think it's also interesting to speculate how sleep and circadian disturbance on the left side of this graph in earlier years, particularly with the increased prevalence of sleep disorders in adolescents and youth, in part related to device usage and in part related to additional stressors at that time, and also the incidence of shift work in earlier stages of career, how these exposures could potentially affect, early in life, could potentially affect attainment of peak bone mass and subsequent fracture risk. So what do sleep and bone have to do with each other? Well, it turns out that, as many of you probably know, bone turnover has a diurnal rhythm. This 24-hour rhythm is stronger for markers of bone resorption, depicted in green as CTX, than it is for markers of bone formation, depicted in orange as P1NP. And although this diurnal rhythm is influenced by food, there have been a couple of studies to suggest that it is an endogenous, centrally controlled circadian rhythm. And it's because of this variation that it's recommended that if you're checking and monitoring bone turnover markers, that they be obtained in the morning while fasting to help control for some of this normal variation. Rhythmicity is likely, this rhythmicity is likely important for normal bone remodeling and suggests that sleep and circadian disturbance could directly affect bone physiology. Those clock genes that I mentioned earlier, the molecular circuitry that drives the circadian clock, it turns out that clock genes have been identified in all of our bone cells, including osteoblasts, osteoclasts, and osteocytes. And although this is a busy slide, it's not intended for you to read it, but it's to highlight how many studies have been performed looking at differences in bone phenotype when clock genes are knocked out in animal models. And this is probably already outdated because it was published in 2021, but as an example of some of the findings from these studies, in a publication by SAMHSA et al., mice who were deficient in BMAL had lower bone density phenotypes at both femoral and, not pictured here, but tibial sites in both cortical and trabecular bone compartments. And although these studies are not really feasible to do in humans. The epidemiological literature suggests that there could also be an association between alterations in sleep time and duration and bone health, namely fracture risk. So the nurse's health study looked at over 38,000 postmenopausal women and reported that those who had a 20 plus year history of night shift work had an increased risk of hip and wrist fractures. That association was particularly strong in those with normal BMI and no history of using hormone replacement therapy. And although this is an association, not a causation, they did adjust for typical fracture risk factors and things that may differ between night shift workers and daytime workers, including alcohol use and smoking history. The Women's Health Initiative also supports an association between shortened sleep and an increased risk for fracture. So they looked at data from over 157,000 postmenopausal women with self-reported sleep duration and found that those with short sleep duration, which was defined as up to five hours of sleep per night, had an association with lower bone density, higher risk of osteoporosis, an increased risk of falls, and an increased risk of fracture out to five years of follow-up. This association with fracture was independent of established risk factors, including the increased risk of falls. So in order to better understand these relationships identified in the epidemiological literature, we looked at bone turnover markers in stored serum from a previously performed study to understand how bone metabolism may be affected by sleep restriction and circadian disruption. The study that was previously performed had a built-in age component to it. So the serum that was available for this study included data from six younger men and five younger women, and from four older men and four older women. This cohort was predominantly Caucasian, non-Hispanic, very healthy with a normal BMI. This study was pretty labor-intensive for the participants and the people overseeing the study. And so I'll just take you through what really these participants went through in order to obtain these data. For three weeks prior to admission, they were required to adhere to a 10-hour sleep opportunity per night. Those horizontal gray and on subsequent slides, black bars represent the timing of that sleep opportunity. This was verified by wrist actigraphy, time-stamped call-ins, and sleep diary. Upon admission to the inpatient unit, they had 16 hours of sleep opportunity per day, which probably sounds pretty amazing to some people. This was broken up into 12 hours of sleep opportunity at night and a four-hour nap opportunity in the afternoon. And this was designed to ensure complete sleep satiation so that there was no pre-existing sleep debt that could have been affecting the participants coming in. They then returned to their 10-hour sleep opportunity per night, during which the first of two 24-hour serum profiles was obtained. Data from this time point will be displayed in subsequent slides in light blue and referred to as baseline. They then entered a forced-to-synchrony protocol, which is akin to the stresses endured during rotating shift work. And it's meant to induce both cumulative sleep restriction and also concurrent circadian disruption. They achieved this by having individuals live on a 28-hour day instead of the typical 24-hour day. So this would be the equivalent of flying four time zones west every day for about three weeks. And they had the equivalent of 5 1⁄2 hours of sleep opportunity per 24-hour period. At the end, and there were 24-7 supervision to ensure the protocol was adhered to. At the end of this forced-to-synchrony protocol, when the participants were back at their normal circadian phase, meaning asleep at night and awake during the day, the second 24-hour serum profile was obtained. And data from this time point will be displayed in purple and referred to as post-intervention. So we measured bone turnover markers on the stored serum every two hours on these two 24-hour serum profiles, which were very carefully aligned, such that each participant was in their same circadian phase for both collections. So that means that any difference we see between these two time points is either due to the cumulative sleep restriction, the history of circadian misalignment, since they weren't acutely misaligned at these time points, or any residual dyssynchrony between those central and peripheral clocks as they attempt to catch up from what they just went through. So I'll take a second to orient you to these graphs, which I'll use to display data on this in subsequent slides. We'll go over data from the men first and then the women. Data from younger individuals will be displayed on the left, older individuals on the right. The x-axis is number of hours into the sampling profile, and the bone biomarker of interest is on the y-axis. Mealtimes are represented by those upside-down triangles on the x-axis. Each graph will have bars along the top representing the sleep opportunities at baseline and post-intervention. Each graph will also have a dotted light blue line representing the fitted curve for the individuals displayed on that graph at baseline, and a solid purple line displaying the fitted curve for the individuals post-intervention. Raw average values for those individuals are displayed as open circles, blue circles at baseline, and open purple squares post-intervention with standard deviation bars. So baseline is always in blue, post-intervention is always in purple. So as you can see displayed on the graphs, men had a significant decline in P1NP levels after exposure to the sleep and circadian intervention. The younger men also had higher bone turnover markers at baseline, so it wasn't entirely clear at this point if the statistically significantly greater declines in P1NP that the younger men experienced was a result of age or baseline bone turnover marker levels. Yeah, these declines in P1NP occurred despite no change in the bone resorption marker, CTX. And as you can see, there was maintenance of the circadian rhythm, the diurnal rhythm of CTX under both conditions. So when we looked at results from women, we saw similar themes in that there was a significant decline in the bone formation marker P1NP in both younger and older women, but the decline was again statistically significantly greater in the younger women who had similar levels of bone turnover markers at baseline compared to older women. So these data would suggest that it's actually younger age that may be a risk factor for the greatest impairments in bone metabolism in response to sleep and circadian disruption. We also saw as a secondary check of bone formation significant declines in osteocalcin. And interestingly, when we looked at CTX, we saw that in older women results paralleled what we saw in men where there was no change in CTX. But in younger women, the declines in bone formation markers occurred despite a significant increase in CTX levels, such that it seems that younger and in particular women may be most susceptible to detrimental changes in bone metabolism in response to sleep and circadian disruption, kind of a double hit. They appear to be forming less bone, but also resorbing more based on serum marker data. Although we don't have long-term data from humans showing that these changes in bone metabolism translate into decrements in bone density and microarchitecture, these biochemical changes do closely parallel results from animal studies in which we do have that longer-term data. So as an example, this is a study from Carol Everson's group in which 18 male rats were repeatedly sleep-restricted over 72 days, which is about 10 to 15% of the adult rat lifespan. And you can see these are representative micrographs of the metaphysiofemur. The ambulation controls, the arrows indicate really plump, active osteoblasts laying down the new osteoid bone stained in blue. And this is in contrast to the sleep-restricted rats that have a flat quiescent osteoblast with no evidence of new bone formation. In the middle graph, you can see that dramatic difference in osteoblast activity in the sleep-restricted rats compared to controls. And that significant decrease in bone formation with no statistically significant change in the bone resorption surface did translate into lower bone mineral density in the sleep-restricted rats in gray compared to control animals. In another study, 24 five-month-old female rats were sleep-restricted to six hours a day for three months. And they had lower P1 and P levels found after as little as one month. And you can see in these representative micro-CT images of the fourth lumbar vertebrae, more disconnections in the trabeculae compared to control animals. And the trabeculae that are there are thinner, more disorganized, and fewer in number compared to the control animals. So it's no secret, particularly at an endocrine meeting, that sleep and circadian disruption has numerous health consequences, particularly metabolic and cardiovascular. And so as we develop a conceptual framework by which altered sleep timing and duration could lead to impaired bone health, it's entirely possible that some of these known decrement, some of these known poor health consequences of sleep and circadian disruption could play a role. But there could also be other direct mechanisms, including increased sympathetic tone, and indirect mechanisms, including medications that people use for their sleep and circadian disorders, or an increased risk of falls. I also wanna highlight that there are data to suggest that this could be a bidirectional relationship. At a minimum, fractures resulting from osteoporosis could contribute to altered sleep timing and duration. So to begin to investigate these mechanisms, we performed, we looked at bone turnover markers in two sleep restriction studies, and I'll take you through these fairly quickly. But in this study of 12 healthy men, 21 to 40 years old, they habitually slept seven to nine hours per night, and they were recruited for an inpatient sleep restriction study. They slept five hours a night for six nights, and we, again, obtained 24-hour serum profiles at baseline and on the last night of sleep restriction. Posture, diet, and physical activity were controlled in this study. And as opposed to prior studies I've shown you of both humans and animals, we actually increased diet caloric intake by 7% during the sleep restriction days to compensate for the increased energy expenditure that occurs with sleep restriction. By design, participants slept longer at baseline than they did on sleep restriction nights. So we observed no change in P1NP or any bone formation marker or bone resorption marker in response to sleep restriction when diet, physical activity, and posture were controlled. Body weight was stable by design during the inpatient study, and so it's possible that changes in body weight or energy expenditure or physical activity are required to see the detrimental changes in bone turnover markers that were observed in prior studies. To further evaluate some of these similar themes that we're seeing in the literature, I'll take you through the final study we did looking at bone turnover marker response to sleep restriction and weekend recovery sleep in 20 healthy young adults who were randomized to one of three groups. The first was a control group that had nine hours of sleep opportunity per night, and then two sleep restriction groups. The WR group on the right stands for weekend recovery. So they were sleep restricted during a typical work week and then had an opportunity to catch up on sleep on the weekends, mimicking what some people may do with usual work weekend differences in their sleep durations. The SR group in the middle was sleep restricted the entire time. So we found no statistically significant change in P1NP, oh, and food intake was balanced at baseline, but then ad-libbed thereafter, resulting in some people overeating and gaining some weight. So there was no statistically significant change in any bone turnover marker, and there was no difference between these two groups when we adjusted for age and sex. But interestingly, when we look at younger women, we again see that there was a clinical but not statistically significant decline in P1NP after just those first a couple initial days of exposure to sleep restriction. Now, this is a really small group, just three women, so we can't really draw any absolute conclusions. We may have been underpowered to detect a relationship, but it's interesting to see that, again, we see younger women may be most susceptible to changes in sleep timing and duration. Another similar theme we saw was changes in bone turnover markers correlated with changes in weight, such that those who gained weight had increases or smaller declines in bone formation markers compared to those who did not. So again, it may be that weight gain or avoidance of weight loss with sleep restriction may protect against detrimental changes in bone metabolism. And the final similar theme that we saw in these data was the correlation between changes in bone turnover markers and changes in morning circadian misalignment. So the change in DILMOF 25% phase angle, that's a marker of morning circadian misalignment, meaning longer melatonin levels after wake. And this phase angle predicted changes in both P1 and P and CTX, and was correlated significantly with all the markers that we looked at. Perhaps most importantly, the directionality of these correlations matches prior literature, and that greater morning circadian misalignment was associated with lower bone formation and higher bone resorption, such that, at least during sleep restriction, greater circadian misalignment, particularly in the morning, could affect bone turnover. Bone turnover markers do peak in the morning, and so that may be why the morning circadian misalignment may play more of a potentially mechanistic role. So to summarize these sleep restriction studies in humans, weight gain or perhaps avoidance of weight loss with sleep restriction may protect against detrimental changes in bone metabolism that were previously observed in sleep and circadian disruption studies in humans, and bone metabolism may be more disrupted by insufficient sleep when it's severe enough to cause significant circadian misalignment, particularly in the morning. Finally, I'll just touch on the concept of chronotherapy to introduce how that may influence some of our prescription choices or guidance in the clinic. So chronotherapy is the concept that the time a medication is administered affects its pharmacokinetics, efficacy, and safety profiles, and as an example of benefits of chronotherapy, we looked at the MAPEC study. Well, I didn't, but just to highlight it. So evening administration, I had no role in that study. Evening administration of antihypertensive medications resulted in better blood pressure control and lower risk of cardiovascular events compared to morning administration. There are similar positive outcomes in the oncology field, and it may play a role in at least some of our menopausal, postmenopausal osteoporosis therapeutics. So this is a small study of just 14 women with postmenopausal osteoporosis who are randomized to administer their teriparatide either in the morning before breakfast or in the evening after dinner for six months. And you can see that there was significantly greater daily mean values in serum CTX in those who administered their teriparatide in the evening as opposed to the morning. When 50 women with postmenopausal osteoporosis were followed out to 12 months, those who took the teriparatide in the morning had statistically significantly greater gains in bone mineral density at the lumbar spine compared to those who took it in the evening. Larger studies are likely needed. However, I think that this is an interesting concept to consider, particularly because teriparatide and other PTH analogs can sometimes make our older women with postmenopausal osteoporosis dizzy and put them at risk for falls. And so it's not uncommon that, particularly at initiation, I advise them to maybe take it immediately before bed so that they sleep through potentially then and avoid a fall. But these data may suggest that morning administration would make our meds most effective. And it's also possible that chronotherapy has less of a role for our long-acting medications. So in summary, but to my knowledge, that has not been formally tested. So in summary, night shift work and short sleep duration are associated with an increased risk of fracture in postmenopausal women. Small interventional human and animal studies suggest that chronic sleep restriction, particularly when combined with concurrent circadian misalignment, impairs bone formation. Sex and age maybe affect modifiers with younger women being most susceptible to the greatest effects. And mechanisms for this relationship may involve food intake and weight regulation. And finally, chronotherapy may play a role for some osteoporosis medications. Thank you, and I'm happy to answer any questions. All right, so before we have someone coming to the podium, but I do have a question that just came in. Really interesting talk, thank you. Have you looked at changes in sleep eating of people observing a month of fasting in like Ramadan? Yeah, we have not. That would be, there are a lot of natural populations that I think would be really interesting to study this in, and that could be one example of that, but we have not looked at that. Waymeyer University of Florida. So very intriguing study, thank you for presenting that. How did you control for 25-hydroxyvitamin D during the course of the study? Yeah, so in the very first study that I showed where we were using stored serum from a previously done study, 25-hydroxy D was not assessed in that. In the studies that I've done, they have to have sufficient vitamin D in order to be eligible for the study. But the first study was not designed to answer a bone question, so I actually don't have any data on their 25-hydroxyvitamin D. And the study where you did check a baseline, did you check it again at the end to see if there was any change? We didn't, but that was a much shorter inpatient study, six days, so I would expect their 25-hydroxyvitamin D to be relatively stable, but I do think it's a consideration for the other study that we did. They were inpatient for three to six weeks, which is a long time to not see the sun, yeah. So before, I'm gonna ask a question. Great talk, it was very interesting and enlightening. So regarding the teriparatide, that was kind of my question. So part of the theory as to why it might be more efficacious in the morning is because of these bone markers being higher in the morning. So do you think in people who have, they work night shifts, at least based on the paper that you showed, I don't know if you can go back to the first one, the timing that you had in terms of zero, six, 12, no, earlier, or the study where people were, they had a 28-hour days instead of 24 hours. I just wanted to look at the x-axis and ask you what you meant by the timing. Is that? The 24 hours into the sampling protocol? Yeah. Yeah, so is the zero, six, is that from when they woke up or is that from the time, like the actual midnight? Yeah, it's an excellent question. Those are not clock times. So because all of these, because we all go to sleep at different times, some people that's 10 p.m., some people that's one in the morning, all of these individuals were allowed to pick their 10-hour sleep opportunity and then the 24-hour serum profiles were aligned to the midpoint of sleep. So this, and we started, so that's 24 hours into the 24-hour sampling protocol, that's not clock time. So that's interesting because then that means that the rise in the C-telepeptide might be different for different people based on their sleeping patterns. And so maybe the terepeptide administration should also be switched around based on night shift workers or something like that. It could be. The other thing with night shift workers that I think could also play a mechanistic role is that, so obviously there are people who self-select to stay with night shift work over prolonged periods of time. Probably night owls are more apt to select staying with that. But people tend to alternate when they're on, so they're on a night shift schedule and then on their off days they tend to go back to a normal sleep-wake pattern. And so that introduces, although that's probably better for a lot of social and other reasons, that introduces a lot more circadian disruption than if they were constantly on the night shift work schedule. So to your point, they might have to, and this obviously has not been studied, but you might have to change the administration time according to are they on their night shift work or are they on a normal schedule or in how much, yeah, it's, yeah, complicated. I don't have an answer to that. Thank you, Chipkin University of Massachusetts. A wonderful set of work and obviously a huge amount, so congratulations on having done that. And thank you for showing the weekend recovery of sleep, actually, for some of us who live our lives that way. Yeah. Not terrible. Yeah. But now you've really focused on obviously kind of timing and duration of sleep, but is there any evidence or do you know of any work doing, looking at the quality of sleep? Have any studies been done looking at bone markers in obstructive sleep apnea or any of those sorts of circumstances? And have you even adjusted for that in the people that you've studied so far? Yeah, excellent question. So in the intervention studies, people with sleep apnea are typically excluded from participating in the studies. There is a whole nother body of literature on sleep apnea and bone. That's actually where I started, but it's a complicated phenotype because there are so many comorbidities with that. And I forgot your first question with regards to, but yes, there, oh, sleep quality, yes. So there was one study, again, epidemiological, cross-sectional seeing an increased risk of osteoporosis in those with poor sleep quality or altered sleep timing. But yes, sleep apnea, there are associations going every direction and I think it's probably because of its complex phenotype. Thank you. Our next speaker is Dr. Nicole Wright. Dr. Wright is an associate professor in the Department of Epidemiology at the University of Alabama at Birmingham. She earned her doctorate degree in epidemiology and master's of public health from the University of Arizona. Her research focuses on musculoskeletal epidemiology, osteoporosis outcomes, and racial disparities in osteoporosis management. And that will be the focus of our talk. All right, thank you for the organizers for allowing me to present and thank you to my two previous speakers who gave me some interesting thoughts with respect to the racial and ethnic disparities in osteoporosis management. So I do not have any financial disclosures. I recently was a consultant in the Radius Advisory Board meeting and some of the work being presented was funded by my K01 as well as Amgen. So for this talk, we're gonna talk, present the racial and ethnic disparities in osteoporosis management and then discuss how that impacts the state of race and ethnicity in the United States with respect to osteoporosis and fractures. And then at the end, sort of highlight some strategies that we can use to address these gaps. So I don't have to go through this. Osteoporosis, over 54 million Americans have osteoporosis or low bone mass and osteoporosis increases your susceptibility for fragility fractures. Based on the most recent NHANES data, the prevalence of osteoporosis by race and ethnicity was reported by Ann Looker and colleagues and having a T-score of minus 2.5 at the femoral neck or lumbar spine, we see that the highest prevalence of osteoporosis is in the Asian population. The next highest is in the Hispanic population followed by white and then non-Hispanic black. However, when you start to dive a little bit deeper and as you've heard in many of the conversations around racial equity, that these are social constructs that do not really represent individuals groups. And so within each race and ethnicity there, or within each race, there are ethnic groups that could have different bone phenotypes. So for example, in the Asian population, that can consist of Southeast Asian, like Chinese, Korean, whereas say, you could also be thinking about East Asians of Indian and Pakistani, which may be more similar to Caucasian counterparts. The same is true within the Hispanic ethnicity in that the primarily of the studies have been in the Mexican American population, but in the United States, we have a growing population of Cuban, Puerto Rican, Central American and South American populations, which again, based on admixture data could have different bone phenotypes. So then when we dive a little bit deeper and looking at the prevalence of osteoporosis within some of these racial and ethnic groups, you can see some differences. And so this was work by my colleague, Sabrina Noel at the University of Massachusetts Lowell, who using the Boston Puerto Rican osteoporosis cohort and compared the prevalence of osteoporosis in men and women in their Puerto Rican population compared to what was reported in Haines. And so as you can see in the women, the Puerto Rican women actually had lower prevalence of osteoporosis than their Mexican American counterparts and more on par with the non-Hispanic white group, whereas in Puerto Rican men, you're seeing a higher prevalence of osteoporosis than the Mexican American non-Hispanic white or non-Hispanic black men. This is true again in the Asian population. So this was work done by Joan Lo that looked at osteoporosis in white South Asian, again, the Indian, Pakistani, Bangladeshi population, as well as a Chinese group. And you could see sort of, although may not be statistically significant, definitely seeing differences in the prevalence of osteoporosis in these ethnic groups. So, and going back to defining osteoporosis, we first need to have, be screened for osteoporosis via a DEXA scan. And previous speakers have referred to the national guidelines around DEXA screening and all of the guidelines that are out there, whether it be from the ASBMR, the NOF, or ISCD, generally recommend DEXA screening in all women, 65 years or older. And then in those who have had hip or spine fractures, and then any younger women who have risk factors that could increase their risk for fracture. And then based off of your BMD, you then fall into these three buckets of normal, low bone mass, or osteoporosis. And those with low bone mass and risk factors, as well as those with osteoporosis are recommended for treatment. Well, within the racial disparities in osteoporosis, we see that there are some differences in sort of following these guidelines by broad race categories. So this was a great study using commercial claims data that had employer-based commercial claims, as well as some Medicare Advantage, and looked at the prevalence of DEXA screening in the race and ethnic groups compared to non-Hispanic white, and found that in all of the age groups evaluated, non-Hispanic black had lower odds of receiving a DEXA scan, ranging from 18 to 8%, whereas non-Hispanic, I mean, non-Hispanic Asian and our Hispanic counterparts had higher odds of DEXA scanning compared to their white counterpart. And so this may align with sort of, we thinking about the prevalence of osteoporosis higher in the Asian and Hispanic populations. So is there more attuned to bone health in these communities, whereas not so much in the black community. And looking at high risk populations, we're still seeing this disparity. So this work was using Medicare, looked at black and white women that had, or I should say black and non-Hispanic white women who had a hip fracture and looked at their DEXA scanning before they had the fracture, as well as after, and compared to non-Hispanic white women, black women had half the rate of DEXA scanning before their fracture, and then even after their fracture, the black women had lower rates of DEXA screening. Moving on to treatment, black women are less likely to receive osteoporosis treatment. In a stroke study that is housed out of UAB, we looked at all of the medications that were reported by the women enrolled in the study and saw that among women who were, say, at high risk of fracture based off of FRAC scores, we're seeing low rates of osteoporosis medication use, but that use was lower in non-Hispanic black populations at 8.4% compared to 13.6%. And again, in sort of this high risk population of a fracture group, in Medicare data, they showed that non-Hispanic black women were 18% less likely to receive medication after their fracture in comparison to their white counterparts. Unfortunately, we have limited data on osteoporosis treatment and in the other racial and ethnic groups. This was a fantastic study using the Kaiser Permanente Southern California data in which in the early 2000s, they developed sort of this healthy bone initiative and optimized sort of patients for osteoporosis therapy where it was warranted. And so in women, when they evaluated among those women by race and ethnic group, who treatment was warranted, they saw no racial and ethnic disparities in the overall treatment rates. However, when they dove a little bit deeper and looked at the hip fracture population, so again, this high risk group, despite all of the initiatives they had during this program, they were still seeing racial and ethnic disparities in osteoporosis treatment and in hip fracture populations, particularly low rates of osteoporosis treatment in African-American and Asian women. Same is true for when they evaluated this in men. You can see a few things overall, the use of osteoporosis medications is lower in men than in women. But when they evaluated again overall, they saw no racial and ethnic disparities in osteoporosis treatment in men. However, in men who sustained a hip fracture, they saw significant disparities again with lower rates in African-American and Asian men at the 11, 13% compared to 35% in white men. So the next part of this is sort of getting into like, why is this important? And there are a few reasons. One, the population growth. And so I think it's well known that we're having sort of this silver tsunami of the increase in the aging population and using census data back in 2018, they compared sort of the distribution of the increase in the 65 plus population from 2010 and saw this dramatic increase, which was more than any of the other age categories, as well as we're seeing an increase in communities of color. And so the blue line is, or the blue sort of section, dark blue section is the Hispanic population. So almost a doubling of 2.5 fold increase in the black population, again, seeing significant increases as well as in the Asian community. And so having growth in these sections can mean more money is being spent by the healthcare system on fragility fractures and communities that were in the past, maybe not as big of a problem. And so the previous speaker showed this great slide, which is important to sort of realize sort of the bone cycle across the lifespan, where we're having this exponential growth in our younger years and reaching peak bone mass. And you can see the lines of this sort of highlights what happens if you sort of reach your peak bone mass and men and women, as well as if there's suboptimal lifestyle and I'll add comorbidities in that list particularly in those middle years where bone formation and bone resorption are coupled. And so thinking about clinically during this time, that's when a lot of comorbidities come up, particularly the endocrine related comorbidities. And so one, if you don't reach your peak bone mass because of the suboptimal lifestyles, and then you have an increased burden of comorbidities, then are you sort of following this natural pattern of sort of decreasing bone with age, or are you experiencing sort of different trends and rates of decline overall? And this we don't know. And thinking on the racial and ethnic side of it, that we do know that racial and ethnic minorities are more burdened with endocrine and cardiometabolic conditions. And so this could play a role in why we're seeing some of the outcomes that we're seeing. So when we think about the next big thing with the increasing in the older population could be increase in fractures. And we know right now annually there's about two to three million fractures with hip being the most serious, spine being the most common in older women, but wrist is the most common in younger women. Fractures have a significantly significant medical expenditures with, when this paper was first published in 2007 saying, you know, in 2005, costs related to osteoporosis and fractures was about the 14 to $20 billion. And then by 2025, it was gonna increase to $25 billion. Well, 2025 is not so far away anymore. And that is probably an underestimate of the true burden of osteoporosis and fractures. And one study sort of broke it down that annually fractures leads to, you know, over 400,000 hospitalizations, 2.5 million medical office visits, which is probably primarily primary care, but in some of our other specialties, endocrinology is one that takes cares of osteoporosis, and then 180,000 skilled nursing facility admissions. And then with respect to the serious outcomes of fracture, mortality is the greatest with an excess one-year mortality in eight to 36%, depending on the fracture site. And we'll go talk about a few other post-fracture outcomes, particularly where we're seeing racial and ethnic disparities. So this was work in Medicare data by our UAB team that showed sort of the racial and ethnic differences in hip fracture rates in women. And to highlight of looking at that blue line where non-Hispanic white women have the highest incidence of hip fractures, and then the red line at the bottom where non-Hispanic blacks have the lowest incidence of hip fracture. Asian and Hispanic individuals are somewhere in between. And one thing also of note that despite Asians having the highest prevalence of osteoporosis, they have a lower sort of incidence of hip fracture. And diving a little bit deeper about looking at fracture burden within racial and ethnic groups, this was work done in the Nora study that looked at fractures within the Hispanic subgroups. So Mexican-American, Cuban, and Puerto Rican. And again, showing that not all groups are the same and seeing that there were more fractures in Mexican-American males and on par with white males and then lower fractures in Cuban and even lower in Puerto Rican males. You can see that similar pattern in females. However, white females still have the highest burden followed by Mexican-American, Cuban, and Puerto Rican-American women with a nice linear decline in the fracture incidence based on Hispanic origin. Within the Asian community, again, work in Joan Lowe looking at hip, wrist, and humerus fractures. You saw higher incidence rates of hip fractures in the Chinese women compared to South Asian. However, you saw a higher incidence of wrist fractures in South Asian community than the Chinese community. And thinking of the post-fracture outcomes, again, mortality is the one that has been most studied and early work in Medicare data showed, predicted, plotted the observed and predicted survival after a hip fracture of black and white men and women. And so for men, you saw really no racial differences in the probability of survival after a hip fracture and that's probability of survival was lower than women. But in women, you saw racial differences in this probability of survival with black women having a lower probability of survival than white women. Again, work in our Medicare population of remembering with the incidence, white women had higher incidence of hip fractures and black women had the lowest incidence of hip fractures. But among women who sustain a hip fracture, black women have the highest incidence of one year mortality following a hip fracture followed by white, Hispanic, and Asian. More recent work in our Medicare data, we evaluated differences in these post-fracture outcomes between black and white women who sustained a fracture. So we identified fractures in our claims data and we had about 400, 500,000 women with the predominant being non-Hispanic white, but we had a nice size of black women and looked at three different outcomes. So the first was death, which is in the green squares. And on average, in Medicare 2015 to 2019 data, we saw that the one year mortality after fracture was about 25% with higher mortality rates in the more serious fractures like femur and hip and lower mortality rates in say, clinical vertebral fractures. Our next outcome in the red circles was debility, and this was defined as a long-term stay in a nursing home. And so after you sustain your hip fracture, maybe you go to acute rehab, but those who lose sort of their mobility and other issues where now long-term nursing home admission is required, so longer than a year, could then lead to other serious outcomes that's associated with nursing home residency. And in this fracture, we saw that black women had higher incidence of debility, particularly in the femur and hip fractures than their white counterparts. The last outcome in the blue triangles was destitution. And we called it this, although it's not necessarily intuitive, but to have a nice ring with the three Ds, that this was basically losing your financial resources that you've gone from Medicare-only coverage to Medicare and Medicaid dual eligibility. And to get sort of Medicaid, particularly in the older populations, that's a significant decline in financial assets and salary. And so for black women, irrespective, so we excluded those who were already dual eligible at the time of their fracture, and that those who had fractures were two to three-fold higher, had higher rates of destitution than their white counterparts. Other post-fracture outcomes are things related to sort of quality of life. And so this great work in the New York State Hip Fracture Database looked at things that you could say could be related to quality of life, like times of admission, so greater or less than two days, readmissions rates, complications, or needing a reoperation within a year of the initial fracture. And with respect to, say, time to surgery, we saw that all racial and ethnic minority groups had a higher odds of having a hip fracture repair greater than two days compared to their white counterparts. And then other indicators of quality of life, we were seeing higher odds in the black community compared to the white community, whereas lower equal odds in other racial and ethnic groups. Again, in the use of physical therapy, thinking about how that relates to long-term mobility, early work at Medicare showed that black individuals after a hip fracture had higher odds of not having physical therapy than their white counterparts. So what can we do? And this is more from a research standpoint and thinking about a clinical side, we can have, I'll entertain questions and thoughts about that, but number one is to increase bone health knowledge in both patients and providers. So in these type of seminars are things that the National Osteoporosis Foundation is doing, or I should say, the Bone Health and Osteoporosis Foundation of increasing that information, that one, there are racial differences in bone, but as well as disparities in the management of osteoporosis. With respect to our patients of increasing their knowledge about bone health. So myself and Dr. Noel have done mixed methods work in her population of Dominican Puerto Rican men and women in my population, black women in the South have shown that these minority populations just have an overall lower knowledge about bone health, lower knowledge about osteoporosis, which then ultimately leads to maybe less compliance with physician orders because they just not understanding. And so we can increase education about bone health and example program was one that I partnered with the American Bone Health Foundation to sort of increase why healthy bones matter in black women. With respect to research, we need to increase information on racial and ethnic differences in bone loss. So one, most of the longitudinal osteoporosis cohorts are about 95% white. And so do we really know what are risk factors for bone loss in communities of color? I have a feeling age and BMT are there, but some of these other comorbidities that may be diabetes has a greater impact on bone in the black community versus other populations. Also one interesting thing in the research world is around this appropriate reference population. And this could be based off of, we are using the 20 to 29 year old non-Hispanic white as the reference group for calculating T-scores. If we change that, would it factor into the prevalence of osteoporosis and therefore treatment decisions with that? More information on fractures. Again, all of the longitudinal studies are mostly in non-Hispanic white women. And then also primarily look at hip fractures. And so we need more information on other fracture types, other outcomes, other than mortality, and then more information around pharmacologic therapies by race and ethnicity. So the receipt of treatment versus what actually happens with treatment with retaining to adherence or different treatment regimens. There's little work in that space. Lastly, I wanted to highlight that we need more research on patient provider and system level barriers to osteoporosis management. So again, myself and Dr. Noel have done a good job on the patient level barriers, but we need more work on those provider level barriers with respect to clinic operations, workflow, who's ordering the dexa, who's in charge of the patient with osteoporosis. Is it their primary care? Is it their endocrinologist? And so then also thinking at the systems level globally, should we do some bundling of preventative services so that we're reducing that disparity by race and ethnicity? So just to end that these are our sort of faces of osteoporosis. So when you Google osteoporosis and images, this is what you're seeing. Whereas a lot of people, this is the face of osteoporosis to them, and this is actually my grandmother. So I'd like to thank my colleagues and we'll now entertain questions. Thank you. talk, one group of people that I think you left out that I always have a hard time with. By the way, I'm sorry, I'm Allison Myers from Montefiore Einstein in the Bronx, are people from the Caribbean. So the Caribbean is a nice hodgepodge of people who may be from Africa diaspora, may be from Asia, some from Europe. And I'm never really sure if I have this woman from Guyana, do I treat her as a black American? Do I treat her as an Asian? I really don't have an idea because there's really not great information in the literature about people from the Caribbean. I mean, you touched a little bit on Puerto Rico, Cuba, but what about people from other parts of the Americas that are not North America? Right, and so as you said, there is little data on that. I think within the major Hispanic sort of subgroups, particularly in the Caribbean, Dominican, Puerto Rican, Cuban, those have a little work need more, but other sort of Caribbean West Indies folks, I think the Trinidad and Tobago osteoporosis studies, which were a long time ago, were the most recent information on that. So we definitely need more work. And then just one other thing I would say in terms of, on an institutional level, access to care is a big issue. So for my patients to have to get to the other side of the Bronx to get an infusion, it can be really troublesome. So I think that a lot of it too has to do with access to care if we have limited places to go. I mean, it's nice that during pandemic, you can actually get the Nalcimab from your pharmacy and a family member can give it, but will this stay as a permanent thing or is this just going to be temporary? But for a lot of people who are low SES, like a lot of my patients, this access and having to get to get another appointment when they're already working two jobs or have limited access to the care, it makes it really difficult. Right, and that's where I think, on again, on the system side of maybe thinking of what else could we be capturing about our patients and some of those social determinants of health that those were systematically captured, then maybe you can alter your treatment regimens based off of some of those things. So low access, maybe financial instability, well, maybe Alendronate may be the best option because it's generic and all of these things versus having Teriparatide every day as an injection or infusions where would mean other things. So I think that's some of the things that we can do more in the field to sort of better assist with reducing those disparities that race may be capturing as a proxy, but actually getting more at the core of why we're seeing these disparities. So I have another question. Great talk. And obviously it's very hard to answer any of these questions, but when it comes to FRAX, right, we're trying to, especially in the people who have osteopenia. So you briefly touched on the T-score calculation and the normative database that we use, which could underestimate osteoporosis. But then looking at FRAX as well, entering which continent they're from, which is sometimes very confusing and then entering kind of the race ethnicity. And when you're entering black, you're gonna get a lower risk of fracture. So are we under treating patients in that sense? So I worked with Mike Lewicki and Andrea Singer on sort of an opinion piece about this. And so on one hand, should we correct for race? And if you look at the longitudinal data then you're like people who are of African origin, identify as black, have higher bone mass. And so therefore it seems like you should correct for that when you're trying to create sort of fracture risk calculators. Now, where I think FRAX is limited in sort of what is actually in the FRAX score. So those 10 variables that you're putting in, are those the appropriate variables that we should be thinking about when you're looking at fracture risk assessment for people of color? And that's where they may fail. So maybe you have this correction based on bone and bone mineral density is assessed by DEXA. So we're not even getting into the quality issues and micro architecture, but do we need to add variables say diabetes, which they say to use the rheumatoid arthritis factor that's about an equal calculator. Things about other cardiometabolic conditions that really have an impact on bone that are not being captured that are more prevalent in communities of color. So for now, do you recommend continuing to use the race adjusting query? I know there's some institutions I've heard where they've completely gone away with FRAX. I've talked to some, and particularly in the endocrinology space of, they're like, well, if you were white, this is what it would be, but for your race, this is what it is. So maybe we'll say it's something in the middle. And then basically for me, as a non-clinician of saying, treat the person who's in front of you based off of their risk factor profiles and not using FRAX per se, but at a population level, I think that FRAX still has some benefits with caveats, obviously. Great, thanks. Joan Lowe, Kaiser Permanente Northern California to this fantastic presentation. You alluded to, and it sort of came up a little bit in the Q&A about the reference population. And yet every time this question comes up, the response is everyone should be compared to the young adult, non-Hispanic white female, even in the previous talk with the transgenders and also transgender population and also men. And how do you feel about the different race ethnicities? Because we've looked at this in Asians, and when you give a Chinese American woman a Chinese American T-score, it does change the definition, classification. So I'm aware of maybe one or two studies that have looked at changing the, when they've used, say, a non-Hispanic black, non-Hispanic white, or non-Hispanic black for black women or for everyone, that it didn't change sort of the categorization of people, right? But to me, personally, it seems like you are not comparing apples to apples when you're putting everybody at a 20 to 29-year-old white reference. So I would challenge the field to maybe have more research on those reference databases. I completely agree, and we need to have that kind of research supported. I think that's the other big challenge. Well, thank you all for staying all the way to the end. Thank you.

Dr. Caroline Davich-Pitts

sleep

bone health

women

circadian rhythms

fractures

bone formation markers

racial disparities

osteoporosis management

mortality rates

hip fracture

knowledge

barriers