skip to Main Content

My mom is an Asian woman in her 70s with osteoporosis.  She tried an oral bisphosphonate and had horrible esophagitis.  She said never again, though she eventually tried an IV bisphosphonate.  She had terrible flu-like symptoms.  She said never again.  But based on reports that symptoms are worse the first time, she tried the IV again the next year and fortunately experienced no symptoms.  (Story used with permission, thanks mom!).  

I tell this story because these issues don’t typically register as more than a nuisance for clinicians, who frequently don’t understand why their older patient with osteoporosis is not taking a bisphosphonate.  But our patients are walking with their feet, and adherence to bisphosphonates for osteoporosis is poor and decreases with time.

When we have a medication with up front harms and downstream benefits, it’s critical that we consider the time to benefit, or how long it will take an individual to benefit from a test or treatment.  Think of the 10 years it takes to benefit from colon or breast cancer screening, which is designed to detect slow growing cancers.  For individuals with a life expectancy less than the time to benefit, the up front harms outweigh the downstream benefits.  

In this context, we talked this week with James Deardorff and Sei Lee about their study of time to benefit of bisphosphonates for osteoporosis, published in JAMA IM. What they found somewhat surprised us: it’s pretty short, about 1 year!  

Which makes me feel better about urging my mom to get treatment.  Bisphosphonates are pretty darn effective, and act quickly.

We also discuss discontinuing bispohsophantes, and if we can use the same logic we consider when starting them.  

Toward the end we talk about the launch of ePrognosis’ new Time to Benefit tool.  This tool provides a clinical recommendation for starting/stopping medications and cancer screening based on the prognosis of the patient in front of you.  See screenshot below.  You can either access it directly here.  I you use the Lee Schonberg prognostic index for community dwelling older adults you will be directed to the tool, and it will automatically place the life expectancy at the calculated prognostic estimate.  

James and Sei requested a terrific song – Bad to the Bone – much better than the Hannah Montana Bone Dance song they threatened me with (maybe next time).



Eric: Welcome to the GeriPal podcast. This is Eric Widera.

Alex: This is Alex Smith.

Eric: And Alex, who do we have with us today?

Alex: Today, we are delighted to welcome James Deardorff, who is a geriatrician and a T-32 research fellow in UCSF’s Division of Geriatrics. Welcome to the GeriPal podcast, James.

James: Thanks for having me.

Eric: And an active Twitterati. What’s your Twitter handle, James?

James: It’s @WJDeardorff.

Alex: Yes. Now James has this terrific Twitter thread. I recommend all of you follow it. He will post like articles that he’s read highlighted with key points noted, absolute treasure. We’re also delighted to welcome back Sei Lee, who is a frequent guest and host on the GeriPal podcast. He’s a Geriatrician Palliative Care Clinician Researcher, also in the UCSF Division of Geriatrics. Welcome back to GeriPal, Sei.

Sei: Great. Thanks for having me.

Eric: So Sei and James, as well as others, just published a paper in JAMA IM on The Time To Benefit Of Bisphosphonate Therapy For Prevention Of Fractures Among Postmenopausal Women With Osteoporosis, that’s a long title. We’ll shorten it to Time To Benefit Biphosphonates for this podcast. But before we get into that topic, meaty topic, does somebody have a song request for Alex?

James: I do. So I want to pick some song that was related to bones. And when you type in bone songs into YouTube, there’s a few options. What came up was Hannah Montana, Bone Dance.

Alex: Remember we’re PG. [laughter]

James: Some things by Bone Thugs and Harmony, which I didn’t think would be appropriate. And then-

Eric: Boney M., what about Boney M.? You do like Ra Ra Rasputin. [laughter]

James: There you go. But I ultimately decided to go with Bad To The Bone by George Thorogood.

Eric: Nice.

Alex: Which is an awesome choice.

Eric: Throwing out a bone there.

Alex: Like Bonin’ in the Boneyard. Anybody remember Fishbone? Oh, they were so good. Except I’m not a bass player, and clearly you need a bass player. All right. Here’s a little bit of George Thorogood and the Destroyers, Bad To The Bone.

Alex: (singing) “Now on the day I was born, the nurses all gathered round, and their gears in wild wonder had the joy they had found. The head nurse spoke up. She said, Leave this one alone. She could tell right away that I was bad to the bone. Bad to the bone. B, B, B, B, bad. B, B, B, B, bad. Bad to the bone.”

James: Well done.

Eric: Alex liked that too much. You should have picked the Hannah Montana song. [laughter]

James: Way too much. We need to torture him somehow. [laughter]

Eric: Yeah. I like when he struggles a little bit. [laughter]

Eric: All right, so we’re going to be talking about Time To Benefit Of Bisphosphonate Therapy For Treatment Of Osteoporosis. Before we get into the subject, James, how did you get interested in this?

James: Yeah, so I completed my geriatrics fellowship last year, and I think as we go through fellowship, you get a lot of teaching about osteoporosis since it’s so common in older adults. And I was really excited just to learn how to manage these patients, because in the patients in geriatrics clinic that we see, fractures just can be so devastating and result in loss of independence. And so many older adults with diagnosis of osteoporosis have never received any therapy for it. And I think what I realized, especially during COVID, is that there’s just so many barriers and hesitations to starting medications as clinicians were maybe hesitant to start treatments because we’re dealing with patients with multiple comorbidities, polypharmacy, frailty.

James: I know a lot of my patients, there were concerns of that ability to take the medications without supervision, since they weren’t having as much caregiver support during COVID, or they had difficulty coming into the clinic for an infusion. They hadn’t seen their dentist in a while and were maybe about to get some dental extractions. And with all those barriers, it just made me wonder. Like one question, how much would some of these patients benefit? And really, the more important question is how quickly they would benefit from starting these medications. And Sei had done a lot of these time to benefit type papers with colorectal cancer screening, mammography, and statin use. And so this felt like a natural extension to those papers.

Eric: Well, before we get to talk about time to benefit and what that is, what do we know about… Like where does bisphosphonates fit in for osteoporosis treatment?

James: Yeah. So if you look at clinical guidelines on management of osteoporosis based on specific criteria with your high risk for fractures, and those vary a little bit by guideline to guideline, but really bisphosphonates are our first line therapy just because we have so much experience with them. And they really have been shown in several clinical trials to have reductions in vertebral fractures, non-vertebral fractures, and hip fractures. So they really are considered first line therapy for postmenopausal women with osteoporosis.

Eric: Great. So bisphosphonates work, they’re often the first line therapy, aside from maybe some other reasons you may want to consider others, including people with CKD, things like that. Then the question is, how quickly do they work?

Eric: Before we talk about that, so you’ve done a lot of research around time to benefit. It’s in the title of this article. What does that mean?

Sei: Yeah, it’s something that I’ve been thinking about and working on for a while. I think what I talk about is that almost everything that we do for patients it has good things, but also can have some bad things. Whether it’s polypharmacy, whether it’s adverse effects. And for a lot of the things that we do, the bad things happen pretty quickly as soon as the medicine is started, for example, but the good things take a longer time. Maybe it’s six months, maybe it’s two years, maybe it’s 10 years. And while for older adults, especially, when something is going to help is just as important as if it’s going to help and how much it’s going to help.

Sei: So, for example, if we find that a time to benefit for some medication is actually 10 years but the harms happen immediately, then that may not be appropriate for the frailest of our patients. Whereas, if the time to benefit is only two months, we would expect that to be beneficial for almost all of our patients. So I just wanted to kind of introduce this idea that it’s not just if it’s going to be helpful, or how much it’s going to be helpful. It’s also important for us to think about when is it going to be helpful.

Eric: So for like a bisphosphonate, if somebody has maybe… Let’s say many of us care for seriously ill, older adults, maybe in the range of many months to a couple years, if it takes five years for bisphosphonate to see a benefit, sounds like maybe starting that person on a bisphosphonate may not be ideal, versus if it takes six months or two months to see a benefit. Is that right?

Sei: Yeah, absolutely. So one of the things that we’re… I think the main thing I think about in terms of time to benefit is, what is somebody’s life expectancy? What is the time to benefit for the intervention? If the life expectancy is much shorter than the time to benefit, that tells me that it is unlikely that this medication is going to benefit the person, which means they’re going to be exposed to all of the harms and burdens, which usually almost always happen up front, but they’re unlikely to get any of the benefits.

Alex: I wonder if we could talk about those harms briefly, and I’ll introduce this issue by way of talking about my mom, who I called for permission to discuss this prior to recording this podcast. My mom is an older Asian woman and she has osteoporosis, which is quite prevalent among older Asian women. And she tried an oral bisphosphonate, she had terrible acid reflux, and it was awful. She said never again.

Alex: So she went without any treatment for a period of time, and just tracked her Z scores, T-scores, and they were getting worse. So she decided to try IV bisphosphonate, and she had horrible flu-like symptoms afterward, right? Just terrible. And she said never again. And then we talked to an endocrinologist friend of ours, and she said, “You know, with the IV bisphosphonates it’s usually worse. You get that reaction the first time, but not so much the second. And sure enough, she had it again this past year and was totally fine. No symptoms afterward. So we finally settled on one, but those harms were not inconsequential to her. She’s somebody who has a knock-on-wood long life expectancy, but nonetheless was hesitant to use them just because of those harms.

Alex: So, of course you’re weighing this against the potential benefits in terms of fracture prevention, et cetera, but just some thoughts from you about the harms of bisphosphonate therapy.

Eric: Also, a big shout out to Alex’s mom, who’s also a generous supporter of the GeriPal podcast. [laughter]

Alex: Thank you, mom. [laughter]

Sei: Salute.

James: Yeah. I think Alex, you hit on the head the two major things that we worry about we’re right away. And it’s very interesting looking at the literature on GI irritation, because I think that’s one of the common symptoms that we see when we’re starting these medications. And when these medications were first started, there was a good case series in the New England Journal in 1996 that really documented some cases of severe esophagitis. A lot of it were people that were taking these medications inappropriately that you typically, when on these oral medications, you want to take them in the morning on an empty stomach, at least 30 minutes before a meal, and then stay upright or standing for at least 30 minutes to reduce that risk of esophagitis or irritation.

James: A lot of the times these case reports of esophagitis were happening in people that were taking them in the middle of the night, or right before they went to bed without water, things like that. And so in the clinical trials it does appear that the rates of GI irritation are pretty similar between placebo and bisphosphonate treated groups.

James: There is this disconnect, like in the biggest trial with alendronate, the fracture intervention trial, about 30% in both groups had some GI irritation. So it might just be the case that there’s a lot of background GI complaints, that people could be taking them inappropriately, so there are definitely ways to minimize it. But we do see a lot of people reporting this, and the GI irritations are the most common reason why people discontinue these medications.

James: And then the other one, like you were saying with the IV infusions, zoledronic acid can cause this acute phase reaction, which is fever, headaches, muscle aches, fatigue, and it can really happen in 20% to maybe 40% of people on their first dose. And so, as you said before, with repeated doses typically less, and you can pretreat with Tylenol or ibuprofen to reduce these types of symptoms, but it can be a concern. It can last for a few days.

James: And so, as Sei was saying before, when you think about some of these immediate harms and burdens, we want to make sure that we’re treating people appropriately that can benefit from these medications.

Sei: I wanted to kind of jump in here and talk about… Like, it’s fascinating. This is one of those really interesting situations where what patients feel and what they experience versus what doctors are focusing on, there’s this such huge disconnect. And from a physician perspective, there’s a feeling of like, “You have no idea how terrible fractures are, and this is such a great drug for decreasing fracture risk.” But from the patient perspective, and as when I’m sick, I’m like, “What do you mean I’m going to be coughing for another five days? I can’t be…”

Sei: Medically classified as minor inconveniences are huge to patients, and I think we have to be able to hold both of those things in our heads as health professionals. And maybe it’s terrible, but it feels like the lesson here is that every older adult needs a physician’s son who can walk them through, who can listen and validate their concerns. Or daughter, yeah. And actually have physician friends to actually kind of walk them through, “Well, maybe this is a path forward,” but it’s certainly not minimizing the patient concerns, but it’s actually working our way through the real symptoms and real side effects that she had to figure out, “How can we get the benefit while trying to make…?” But ultimately it requires a level of trust because what I heard, she said never again to the IV bisphosphonate, and then you managed to convince her to do it, which is kudos to you and, yeah, great. That has a good ending.

Eric: All right. Let’s jump into the JAMA IM publication, James. Can you just briefly describe for me, what did you do? What were you trying to figure out?

James: Yeah, so what we were really trying to figure out is to help with decision-making around the initiation of bisphosphonates for osteoporosis. We wanted to identify what the time to benefit for the prevention of different fracture types among postmenopausal women with osteoporosis, and to do that involves a lot of looking at previous randomized clinical trials and looking at the survival curves, and trying to see when those curves really separate, when you start to see a benefit.

Eric: All right, so you took a lot of randomized control trials out there, you looked at how long it took to see a benefit in each one, and kind of pooled them all together?

James: Yeah, exactly. So what we did is we looked at trials from five previously published systematic reviews, just reviews that gathered evidence from a whole bunch of different trials, and so we picked the ones that met our eligibility criteria. These were clinical trials involving postmenopausal women with a diagnosis of osteoporosis, based on either an existing vertebral fracture or a bone mineral at the density T-score of less than negative 2.5, which is kind of the typical definition of osteoporosis at that time.

James: And so this is a relatively high-risk population, and most of the clinical trials that were initially used to get these medications approved by the FDA use this eligibility criteria.

Sei: Yeah. So what I would say is just to kind of try to simplify this. What we looked at were these trials. And so they get a whole bunch of relatively high-risk women, and then they get randomized to either a dummy pill or bisphosphonate. And then we follow these groups of women, this thousands of women, in each arm over time. Usually what you see is there’s a slow rate of increase of some women have fractures, whether they’re on the dummy pill or the medicine. But at some point after the start of the study, after half of the women get these bisphosphonates, what you see is that the people who got bisphosphonates have lower fracture risk, and the people who didn’t get bisphosphonates have higher fracture risk.

Sei: And what we were trying to do was to see when exactly those separation became big enough for it to be clinically meaningful. So, that’s the graphical way of explaining what we did. Of course, we use kind of fancier masks to try to come up with quantitative estimates of time to benefit.

Eric: And two questions about that. One, you looked at… Your primary outcome was non-vertebral fractures. Why that instead of just all fractures?

James: Yeah. So I think most commonly the clinical trials assess non-vertebral fractures, and so it is definitely the most common outcome. So non-vertebral fractures include fractures like hip, clavicle, humerus, wrists, pelvis, and leg fractures. And it varied a little bit study to study, but that was generally the case. And so these fractures are relevant to patients. I mean, hip fractures can certainly cause the most morbidity, mortality. I think cited mortality rate after one year, mortality rate after a hip fracture is somewhere in the range of like 20 to 30%, and even fractures of the humerus or the leg can cause some decreases in function that can have long lasting impacts. So these are really relevant outcome events. But we also looked at other outcomes, like specifically hip fractures, which are a little bit more rare, but certainly very important. And then we looked at clinical vertebral fractures as well.

Eric: I guess I got a question for Sei. You said when the graph starts separating to a clinically important difference. How do you define that? How do you define when what’s clinically meaningful as far as when we’re going to say, “Yeah, this is five years down the line. They’re different now, clinically meaningful. But two months down the line, oh, they started to separate.”

Sei: Yeah. I mean, I think that is a very thorny question. And I think I can say, well, one in a billion isn’t clinically meaningful, and one in one is clearly meaningful. But I think it’s quite subjective to see what is a clinically meaningful difference. And so one of the things that we look at is how frequent the harms are. If the harms are happening in one in 10 women, then I think we can say that the benefits have to be at least on par of that. Maybe it’s a little bit less if the preventing fracture is much more important than avoiding the harm. So that’s one way of answering the question of what is clinically meaningful. The other thing that I would say is because there is no single right answer for all patients.

Sei: One of the things that we do is actually identify the time to benefit for several different clinically meaningful benefit thresholds. So we talk about, if you believe avoiding one fracture for 100 women treated is a clinically meaningful benefit, this is the time to benefit for one in 100. If you believe avoiding one fracture in one in 500 women is a clinically meaningful benefit, this is the time to benefit for that absolute risk reduction threshold. So we provide a range of these clinically meaningful benefits. And depending on, I think, the individual patients values and preferences, a differing threshold may be appropriate.

Eric: All right, James. Let’s drop the suspense. What did you find? What’s the time to benefit for bisphosphonate treatment of osteoporosis?

James: Yeah, so overall we identified kind of 10 clinical trials involving three different bisphosphonates, alendronate, risedronate, and zoledronic acid. And so we did all the fancy statistical analysis with the survival curves, and really what we found is that the time to benefit to prevent one non-vertebral fracture among a hundred postmenopausal women with osteoporosis was 12.4 months. So in other words, this meant that at an absolute risk reduction threshold of 0.01, a hundred plus menopausal women with osteoporosis would need to be treated with a bisphosphonate for 12.4 months to prevent one non-vertebral fracture. And so that’s kind of the headline one.

James: We also looked at other outcomes, like hip fracture. And so hip fracture is a little bit more of a rarer outcome. And so what we saw is that the time to benefit for hip fracture, it would take about 20.3 months treating 200 postmenopausal women to prevent one hip fracture. So since it was a little bit rarer, we would need more time and need to treat more women.

Sei: I want to jump in here and just say, overall, I feel like when I saw the results, I was like, “Dang, bisphosphonates are really good. They work pretty fast.” And my kind of takeaway from the results was like, “Yeah, almost every older adult should be on bisphosphonate. We should try to figure out a way to get them on,” because not only do they work, which we knew from previous studies, from the RCTs themselves and previous systematic reviews and meta-analyses, our time to benefit analysis shows that they work pretty fast. 12.4 months is not very long. And so I would say my takeaway from this was like, “Good drugs, works so good and works fast.”

Alex: Yeah. And osteoporosis is common, but you said almost every older adult. Do you want to amend that to every older adult with osteoporosis?

Sei: Yes. Yes. Thank you. Thank you for that clarification, Alex.

Alex: My question is, like when you say this is powerful, this works really well, is there a way that you can contextualize that in terms of time to benefit for other treatments?

James: Yeah, I mean, first of all, I think anything in medicine is not going to be like a slam dunk type thing, except when we’re dealing with like insulin and type one diabetes it’s pretty obvious. And so when you first start to look at these numbers, like one in a hundred for 12 months, sometimes it doesn’t seem like that much of a benefit, but when you’re especially being on a population level, the results can be pretty dramatic when we’re dealing with preventing fractures that can have long-term morbidity and mortality.

James: And so when you think about other things that we do, like colorectal cancer screening or mammography, or even like statin therapy for preventing heart attacks and strokes, they all have pretty dramatic lengthy times. So things like colonoscopy for cancer screening at somewhere in the range of 10 years, same thing with mammography as well for preventing breast cancer. For statin therapy in preventing heart attacks and strokes is probably in the range of two and a half, three years as well. So, in terms of other interventions that we commonly do, the bisphosphonates seem like a pretty reasonable option.

Eric: My question is what do I do with this clinically?

James: Yeah. So I think the implications are pretty interesting, and this is where I wish that there was more space in the article to really dive into the details on this, because it is challenging. I think the numbers that we show in our paper suggest that, really, the short term benefits of bisphosphonate therapy are going to outweigh the short term harms for the vast majority of post-menopausal older women, and really that these numbers can be used to guide discussions among patients in terms of their values and preferences.

James: One of the limitations of analysis like this is that a lot of the women in these clinical trials were relatively healthy in the range of 60 to 70. And in a geriatrics clinic, we’re seeing some of the oldest old, like 80 to 90 year olds with frailty and comorbidities. And so the question inevitably comes like, if we’re talking about people with limited life expectancy, how do I know that these results can apply to them? And I think that’s one of the most challenging parts about this. And what I would say to that is that one, we do have limited evidence from other subgroup analyses in clinical trial, suggesting that these drugs are still effective in older adults with like the oldest old and with advanced frailty. And so I think that these numbers likely still apply to them.

Eric: But maybe if their prognosis is less than a year, reconsider it, and certainly if we’re thinking about hospice patients. I guess my question, too, is I’m caring for hospice patients right now. It takes a year to see that that benefit in one in a hundred. Does this also tell us about… Like, I see a lot of patients who may be on Alendronate. Does this tell us anything about, can we stop it? Sei, what do you think?

Sei: Oh, I get all the easy questions, right?

Sei: So the problem is that I would say it is… I think the fairest answer is it’s unclear. I don’t know. What we were able to do was to look at trials that were starting these medications. And so I can tell you with a high degree of confidence that this is how long after you start this medicine that we can see this magnitude of benefit. The question you’re asking is if somebody’s been on it for a long time and we stop it, is the time for the harms for the fracture risk to go back up, is that the same as the time that it takes when we started for the fracture risk to go down? And I would say that it’s unclear.

Sei: I think going out on a limb, I would say compared to other drugs, this drug, especially from what we know about the mechanism of action that the bisphosphonates actually get deposited into bone, the stuff that I remember learning about these drugs is that they actually seem to have a long tail of remaining in the body. So I think the correct answer is we don’t know. But if we need to make a decision, I would say that my sense is that if we take it off, my sense is that it will probably remain having an effect for substantially longer than the 12.4 months that we found for the time to benefit.

Eric: And that’s my understanding of the whole role of drug holidays. So after being on bisphosphonates, and depending on which one, five years, 10 years, actually doing a drug holiday and monitoring this stuff, because the mechanism action, it’s still working.

James: Yeah. I think that’s exactly right, is that even like a single dose of the IV zoledronic acid can decrease bone resorption for several years. And so you can really still derive some benefit kind of far out from that initial dose. And especially it becomes even more interesting when you’re dealing with things like hospice patients or nursing home patients, whereas they start to approach near the end of life, they may have to be more bedbound and just not have as many opportunities to really have a fracture. And so it becomes challenging in that sense that a lot of the models that we’re using might overestimate their fracture risk if they have functional impairments where they’re bedbound for most of the day. I mean, everything about medicine is a risk benefit discussion, and in the context of the patient factors, that’s what ultimately is the most important part.

Eric: I guess my last question about this trial before we can talk a little bit more general about time to benefit is, did this include trials like the zoledronic acid, like post-hip fracture studies? Did it include not just primary prevention, but like secondary prevention after hip fracture?

James: We did not. So, that one’s the Horizon Recurrent Fracture Trial we published in New England Journal. I don’t remember the exact year, but it was like…

Eric: It was a ways ago. It’s like 2007 or something.

James: Yeah.

Eric: I remember it shortly after fellowship.

James: And so in our inclusion criteria, we actually did not include that one. One, because it had some men involved, and two, we just felt like after a hip fracture, really all individuals should be on a medication. That was a different population than what we were targeting, and so we ultimately excluded that. And we also excluded some with… there was a big trial in women with osteopenia, not osteoporosis, with zoledronic acid, so that was excluded as well. So really we wanted to create kind of a homogenous patient population for our study of just postmenopausal women with osteoporosis based on a baseline fracture or a T-score less than 2.5.

Alex: But not a baseline hip fracture.

James: Yeah.

Alex: Yeah.

James: Exactly.

Alex: Yeah. I think from what I remember that study, the graph started separating around like six months for any clinical fractures, non-vertebral fractures. And then like at a year, I think their graph started separating for death too. It was like a year and a half.

Eric: All right. So maybe we can take a big step back again. Alex brought up this question about time to benefit. We talked about some other examples. I think now is an interesting time, right? The ePrognosis has something new in it. Is that right, Sei?

Sei: Yeah. So I think one of the things that I’ve always been thinking about is how we could use time to benefit as a way of prioritizing different preventive interventions in older adults, because obviously if something is going to help you in one year, whereas something else is going to help you in 10 years, it doesn’t seem like too much of a stretch to say, I think we should focus on the thing that’s going to help you in one year. And so, one of the things that we looked at was trying to combine these concepts of life expectancy with the time to benefit for a variety of different interventions. And so we’ve been working on this new page for ePrognosis where we try to put all those things together. And Dr. Smith magically made it appear. It’s amazing.

Alex: So for those of you who are watching on YouTube, and this might be a moment where if you’re listening to this podcast, consider checking out our YouTube channel in order to see it.

Eric: Unless you’re driving, you don’t.

Alex: You don’t. We’ll describe it for you. Sei, do you want to describe what you’re seeing here?

Sei: Yeah. So basically what you have rows of different type of interventions, like we have intensive blood pressure control, statins for primary prevention, and that’s the different rows. Then across the X axis is the number of years. And what it shows is that if your life expectancy is, for example, five, when it goes down to intensive blood pressure control, it’s in the green because intensive blood pressure control has a life expectancy estimate of around 1.7 years. And that’s where at 1.7 years, it’s red to the left and then it’s green to the right. And for different types of interventions, like statins for primary prevention, the time to benefit is at 2.5 years. And so the nice thing is for a given life expectancy, for example, at five years, what you see is that intensive blood pressure control, green. Yes, it makes sense to do that. Statins for primary prevention, yes, it makes sense to do that.

Sei: If you go further down and look at colorectal cancer screening, since that time to benefit is closer to 10 years, and mammography is a little bit over 10 years, both of those interventions are in the red. It doesn’t make sense to do that because the time to benefit substantially exceeds the life expectancy of five years in this example. And the life expectancy can move left or right, depending on what somebody’s life expectancy is. And what we’re working on now is making the life expectancy tie into some of the mortality calculators that are available on ePrognosis, so that once you calculate the life expectancy, it can immediately link to this page and anchor the life expectancy as given the patient’s life expectancy as estimated as this. It’s easy to see these interventions that are in green are the ones that would be recommended.

Alex: And we have that linkage already for the Lee Schulberg Index. That’s Sei Lee and Marc Schulberg’s Index combined indices, so that when you enter in the risk factors and you get a life expectancy from those indices, at the bottom of that page you automatically see this life expectancy is set to the life expectancy from those calculators. And I’d also add that bisphosphonates for osteoporosis are on here. Intensive glycemic control is on here, and that we are hopefully going to have some additions to this as new data comes out and as we build out this feature.

Sei: Yep.

Eric: If there was a magic wand, are there one or two additional interventions that you really are eager to see on this page? Or can you not tell us because of some grant or something?

Sei: Well, I think the next thing that I’ll be working on is I think quantifying the intensive glycemic control. It’s already on there, but I think that’s actually from just a single study. So I think that’s the next thing to try to quantify what the time to benefit for intensive glycemic control. But I’m certainly open to ideas. I mean, I think bisphosphonates was really an idea that’s been percolating, and James really kind of took the ball and ran with it, so…

Eric: James, what’s on your radar?

James: Yeah, I don’t know.

Eric: What would you like to see?

Alex: I have one.

Eric: Yeah, go ahead, Alex.

Alex: I’d like to see time to benefit for… Da, da, da, drum roll, drum roll, drum roll. Anti-dementia medications, cholinesterase inhibitors, Memantine, right? What is the time to benefit? And also, what happens when we stop them, right? How long does it take, right? You often taper cholinesterase inhibitors, but how long does it… And there’s a concern that there may be a rise in agitation after discontinuing acetylcholinesterase inhibitors, for example. But how long does it take until you see those events? And is that within the lifetime of many of the patients that we care for, for example?

Eric: I like that one.

Alex: That’s my request.

Sei: We’re on it.

Alex: I would like to see it because we’ve had this before, talking about atrial fibrillation, anticoagulation, what are we looking at as a time… Because this often comes up in like our hospice patients. I guess this is the hard part, it’s often, “Should we continue this?” And Sei, I’m hearing from you that is a slightly different discussion than should we start this. Is that right?

Sei: Yeah. I think it’s probably more parallel than in bisphosphonates just because the pharmacokinetics of the anticoagulant medications. But what I can say is-

Alex: They’re so short acting.

Sei: Yeah. When I took a look at that data, it does seem like the… That’s a really hard one because I think the benefits occur pretty quickly. However, the flip side is the harms are real and the burdens are real. I remember my mother, she complained bitterly about how she hated her gums always bleeding and how easily she bruised. So I think it’s a substantial burden on the quality of life, but it’s also clearly beneficial in terms of avoiding this terrible outcome of stroke.

Alex: I like how we both worked our mothers into this.

Sei: They’d be proud.

Alex: Yeah. And, how much these things that are viewed as nuisances are very real and concerning to many older adults. Here’s another one. This is something Eric talks about a lot. Finasteride, right? Several of us have practiced at the VA. We take care of a lot of older men with limited life expectancies. When can we stop these medicines for BPH?

Eric: That one, we see a lot of people who come in the hospital, some they have retention. Somebody starts them on alpha blocker and Finasteride. And again, Finasteride works by slowly decreasing the size of the prostate. That’s not going to happen tomorrow if I start it today, it takes many months. So the question is, how long does it take if we stop it? It’s a really interesting discussion.

Eric: Before we end up with the song too, James, what’s next for you?

James: So during my research year, I’ve been working on whole bunch of projects with Alex and Sei, so I think we’re doing some mortality models and community dwelling older adults with dementia and some other prognostic calculators. So yeah, stay tuned.

Eric: Great. Looking forward to having you on our future podcast.

James: Yes.

Eric: Before we end though, hey, I want to thank both of you for being on, but maybe a little bit more, was it Bad To The Bone?

James: Bad To The Bone?

Eric: I was thinking Hannah Montana. [laughter]

Sei: Boney M. More Boney M.

Eric: Boney. Yeah. Rah, rah, Rasmutin.

Sei: I was going to say, the reason why we don’t have good data on discontinuation is because RCTs are expensive. And therefore the only people doing RCTs are most commonly done by drug companies because there’s a profit benefit, a profit motive of, if we show benefit, then we can sell lots of this medication. What we really need are discontinuation trials, where if we stop this, is it going to cause harm? But the problem is there is no financial incentive, except for CMS and from a societal perspective, on doing any of these discontinuation trials.

Sei: It takes millions of dollars, tens of millions, sometimes hundreds, at least tens of millions of dollars to run these trials, and there is nobody who is willing to put up that sort of money to see if we should use less of any of these drugs. So I think that is a fundamental problem and where, unfortunately, I think the NIH, even though it has a big budget, just doesn’t have enough to be able to do these discontinuation trials for all the different medications for whom that information would be really helpful.

Alex: And that’s somewhere where this idea of like drug holidays may come in useful, right? Because we could potentially study it and then it’s not stopping it completely. But on a large scale, we’ll get some sense of what happens to people as we stop it, not just right at the end of life.

Alex: All right. Terrific points. So, here’s a little bit of Bad To The Bone.

Alex: (singing) I broke a thousand hearts before I met you. I break a thousand more baby before I’m through. I want to be yours, pretty baby, yours and yours alone. I’m here to tell you, honey, I’m bad to the bone. I’m bad to the bone. Buh, buh, buh, bad. Buh, buh, buh, bad. Buh, buh, buh, buh, bad. Bad to the bone.

Eric: I feel like it should be good for the bone given the topic, but maybe not so good for the-

Alex: That’s what I was going to say. I think I did that for AGS. I’m good for the bone. I think that was a change.

Eric: Well, James, Sei, big thank you for joining us for the GeriPal podcast.

James: Thanks so much for having us.

Eric: Thank you, Archstone Foundation for your continued support, and to all of our listeners.

Alex: Thanks.

Back To Top