many people may not know that Dr Gavin was one of my introductions into academic medicine, as he was the first faculty member at Washington University in ST louis to interview me for a research Fellowship position. And that was back in the mid 1980s. So Jim we're not changing too much. I don't think as we get more gray hair, I want to thank the organizers of this meeting for giving me the opportunity to talk about health system outcomes and bottom line budget and far miccoli's economic impacts associated with sensor based C. G. M. Across the diabetes spectrum. These are my dualities. A lot of people just show it too quick to read. I just want to make very transparent. This is where I do my research and my consulting, so do our home monitoring tools, improve outcomes and costs. Not many people in the audience remember this. This goes way back decades and decades. A home urine chemistry test to test for glucose urea. Then in the 40s and 50s and 60s, even into the 70s, this is what was used. We eventually, in the 1980s got into looking at glucose levels on the strips, which we would cut to save money. Eventually, things were commoditized into the B G. M blood glucose monitoring, which is still around today. And I want to think about this for a moment during this 100th anniversary of the discovery of insulin. Do we have data to suggest these older tools had an impact on acute complications and costs? Because as opposed to the days when we were just starting all of this, I have to say. Um Even back in the days that I remember with these urine glucose testing um in the 60s, 70s and even into the 80s, nobody really thought about the cost of monitoring. And now, of course, the economic costs is very high in everybody's mind. So the question is what about the pharma co economics or the technology economics of our continuous glucose monitors today Because there is a cost to everything. And I would argue the cost of diabetes um took a huge jump literally the day the first insulin injection was given to Leonard Thompson because this all comes at a cost. So I want to specifically focus on trips to the emergency department and looking at these glycemic costs for the glycemic emergencies which are extremely expensive. This is a very recent article from the Jama network looking at the Association of cumulative multi morbidity and got eczema control and medications with hypoglycemia related E. D. Visits. So We're going to look first at hypoglycemia over 200,000 adults with diabetes. Now this was an older population And most of them of course would have type two diabetes. But when we're talking about hypoglycemia, not surprisingly, we could document about 69% on insulin with or without cell phone to your area. These data come from 2014 to the end of 2016. And the data are very interesting. Were looking at crude rates of hypoglycemia related emergency visits and hospitalizations. And if we just look at the age group, we see this inflection. When we get to the Medicare age, we are seeing more emergency department and hospitalizations in the seniors than we did for the younger group. And if we now look at stratification bi annual household income, I don't think this really surprises people that as the income goes up on the X axis, the amount of events actually go down. So it is those people who make the lower annual household incomes that have the higher rates of hypoglycemia. What about looking at total comorbidities. If we go up to as of eight comorbidities, the more comorbidities you have sicker you are, the greater your risk of having one of these hypoglycemic emergencies. And this one is very interesting because this is what we also showed in the T. One of the exchange. If we stratify this by hemoglobin A. One C. There's this U. Shaped curve. And perhaps paradoxically when you first think about it. But I don't think it is As we showed this before. These high a. one c levels have the highest amounts of severe hypoglycemia requiring emergency department or hospitalization. And it actually comes down to a low level in the 5.7-6.4% range. But once it gets below that you see this trend upward. And the way I have always thought about this, but I don't know for sure is these people up here don't check glucose levels as much. And because of that, they see a high number and they give too much insulin and they end up in the hospital because of that can't prove that. But that's what I think. And then finally, not surprisingly, the insulin sensitive type one individuals Have a 34% increased risk in hypoglycemia-related emergency department visits or hospitalizations. And I think that's what we would expect. Now. What about the annual cost of health care resource from hypoglycemia? This is a complicated slide and I'm going to make this try to make this simple. The blue is no hypoglycemia. The orange is hypoglycemia that's not severe. And then this green or gray is severe hypoglycemia, both from the healthcare professional in the emergency department in the hospital. And then the total cost. And I think the bottom line when we look at the total cost is that the severe hypoglycemia cost? About $15,000 over $15,000 for health care resources. When you look at all of the severe hypoglycemia and this is in $2,013. So even though the study was published in 2018, this was actually quite a while ago and I would argue this number is hired now. So what about deka and hyperglycemia emergencies? Because we've been up until now looking at hypoglycemia. Well The rates also are not encouraging. This is from 2009 to 2015. Even though this was published just last year in diabetes care. And you can see that emergency department and inpatient admissions for DK has been increased. Of course this does not include what we just saw over the past 15 months or so with COVID. I have to thank the number is much higher. And what we also know with this is that part of the reason for the increase in numbers here. Certainly not. The only reason Is difficulty in access to insulin for individuals with type one diabetes. This study from Jammeh in 2019. Also I think is um important looking at the incidence of hospitalizations for major diabetes-related complications. So we're looking at everything I want to focus here on hyperglycemia. And you can see this bump up between 2009 and 2015 in this younger age group. And as it turns out the black line, which is hyperglycemia, there is also a bump up in the 45-64 year old age group. and actually between 2009 and 2015, Hyper glycemic crisis in the United States increased by 81%. So this is actually a huge number now. What about length of stay when in the hospital for? DK. Well the first thing you see is that if we just look at length of stay in these blue bars, the length of stay actually went down. So perhaps that's a good thing yet the cost per episode actually went up To over 26. Almost $27,000 per episode for the cost. So the length of stay went down. But the cost went up between 2003 and 2014. And we are talking about a lot of people here. 100 and almost 189,000 people. So I want to do the math and I can only convert this to 2020 numbers. But in 2014 if we look at the admission number and we multiply that by the cost, we are talking about over $5 billion dollars in 2014. And if we calculate that out with this inflation calculator that I really like to use. We are now actually just a tad under six billion dollars. And that's just for DK in the United States. So the number of cases of decay in the US. costs well, in excess of $6 billion 2020 because we are seeing more Dkk now. So this number is actually an under estimate. So deka is expensive. And I think the real question as we think about our evolution into C. G. M can C G. M reduce the cost of vodka. Well, this is a study from the UK published not long ago looking at freestyle, liberate in the UK. Looking at 2400 patients with type one diabetes from 30 hospitals. The mean age was 34, more than half were women Diabetes duration 14 years. Bmi was 24 and the hospital admission for hyperglycemia For DK was reduced from 5% to 1% after six months on the freestyle, the break. Now, that's an 80% reduction and zika that is, in my view, huge. Well, let's look at another population. This also happens to be with freestyle negre. But it's another UK analysis. It was a prospective observation study, 900 people using the lead break. It was an adult population with 20 years duration. And in this study, DK was reduced from 10 to 2 episodes in the six months following the initiation of the freestyle debris. And once again, how often does this happen? It was an 80% reduction. So, these two studies from the UK showed the exact same results, looking at reduction of decay with the freestyle a break. So What if we reduce the DK costs in the us by 80%? Well, we don't have US data, But if we look at the number that we had almost six billion And we multiply that by .2, that means the cost would be $1.2 billion. Now I can't You know, these aren't the numbers in my check checking account. So I can't tell you what $1.2 billion importantly, We would save almost $5 billion dollars in costs if these data were also true in the United States. And there's no reason to think that they wouldn't be so similar reductions in acute diabetes, complications from this real world experience in France. This is called the relief study. It was just published earlier this year with 74,000 individuals with either Type one or Type two diabetes. Initiating lee brae and the french national claims database And looking at type one diabetes where not surprisingly, you have more dramatic results. DK was reduced by 56%. There was a reduction in Hypoglycemic coma is by almost 40 and the overall reduction In these guys Emmick emergencies was 49%. The same calculations were performed in type two diabetes. The numbers were not as high, but you still saw over a 39% reduction in these acute ischemic emergencies. If we look at this For 12 months after the Lee Brae and we look at this based on how frequently they are testing their blood sugar. What you see is is that there are not huge differences. Like you would think, for example, between one and three tests per day, that was pretty similar 4-5 tests per day. But overall, even with more than five tests per day, There was still a 51% reduction. 51% reduction in um hospitalizations. This is hospitalizations, a big ticket item in terms of costs. So if we translate this study to us, deka cost of 54% reduction. We are still saving over $3 billion. No matter how you look at c. g. m. We are saving money on these big ticket items. In every study that has been published, more data can see GM improve outcomes and costs. This is more real world evidence in the US. This is a study that Richburg install was the first author on looking at acute diabetes events combined combining inpatient and emergency outpatient events including hyperglycemia, hypoglycemia, DK, A coma and hyper osmolarity secondary all cause inpatient hospitalizations. Over 2400 individuals with type two diabetes receiving multiple injections before and six months after starting the freestyle library. It was an adult population in their mid 50s on average. And let's look at the data. I mean when I first saw this data, I literally got chills because looking at the pre acquisition compared to the post acquisition of the library, you saw acute diabetes events either in or outpatient emergency. The curves changed right away like in the first week or two and the hazard ratio was .39. So there was a 61% reduction of all of these acute diabetes events all cause hospitalizations. You see the same sort of relationship With again, a strong hazard ratio of .68. Very statistically significant. And the conclusion was these findings provide support for the use of C. G. M. And type two diabetes patients treated with short or rapid acting insulin therapy to improve clinical outcomes and potentially reduce costs. Now there's more real evidence because there are two types of C. G. M. And this was data that will be presented as a late breaking abstract this year. Looking at an IBM markets can research database comparing the decks calm and the freestyle leverage. There were Over 3500 individuals with type one diabetes. Over 3900 people with type two diabetes. And all of them were using multiple injections of both the type ones and the type twos for both Type one and type two diabetes. The liberal and the decks. Com users were propensity score match. Now this is important because propensity score matching means any differences in baseline data were normalized so there were no differences in demographics, provider visits comorbidities, insulin pump use and baseline events. Looking at all cause hospitalizations for type one diabetes. The bottom line, the freestyle library and blew the decks come in pink. There was no difference. There was a reduction, but there was no difference between the two systems. Similarly, when looking at the type ones for acute diabetes events, The two did not show any difference and the p levels for both were non significant. If we look at acute hyperglycemia events and acute hypoglycemic events. Again, the curves are almost interchangeable In type one diabetes, and if we look at type two patients for all cause hospitalizations and acute diabetes events, once again, the two systems worked great. They reduced the events, They reduced the hospitalizations, but there were no difference between the two. The point is, you got to get on a sensor and which one, at least from this analysis doesn't matter if we look at hyper glycemic events and hypoglycemic events again. Um we don't see any difference between the two. I think the data are pretty convincing. So this real world evidence conclusion from this late breaking abstract is that patients with both types of diabetes experience similar reductions of acute diabetes events and all cause hospitalizations when using either brand of C. G. M. So I would like to conclude this discussion by saying that the acute diabetes emergencies are increasing in the US both on the hypo and hyperglycemia. Side data continues to accumulate from around the world that C. G. M. Reduces the incidence of these life threatening events. And more formal cost effectiveness studies assessing the impact of this technology in different countries are now needed. We've seen three different studies from the UK and from France and we need more of this data. Um I showed you some of the data that we saw here um in the United States with this late breaking abstract, but what we now need our formal cost effectiveness analyses of that. And with that, I'd like to thank you from Seattle Washington and the University of Washington where you can see the hospital, the whole medical school here, our beautiful university of Washington campus and of course husky Stadium, which I am told. And I think this is true, the best place to watch college football in the United States. So with that, I thank you all very much.
Related Presenters
