Slide Lecture with Audio
This activity is based on a 45-minute slide lecture presented by Dr. Fernando J. Martinez and Mr. William F. Tulloch. This program is available as read-only and with accompanying audio.
*Scroll to read lecture and/or follow along with audio.
Slide 1: Welcome to "Guidelines and HEDIS® Measures for COPD: Strategies to Optimize Clinical and Economic Outcomes." This program is jointly sponsored by the National Committee for Quality Assurance (NCQA) and Princeton CME, and is supported by an educational grant from Boehringer Ingelheim Pharmaceuticals and Pfizer. Our presenters will be Dr. Fernando Martinez, Professor of Internal Medicine at the University of Michigan Medical Center, who will be discussing burden of COPD and guideline-based strategies to improve clinical outcomes, and Mr. William Tulloch, Director of Customer Resources at NCQA in Washington, DC, who will discuss current HEDIS® measures for the diagnosis and management of COPD. Dr. Martinez will now begin his presentation.
Slide 2: The goal of this presentation is to give a sense of current strategies in the evaluation and management of COPD, and how those strategies led to the current HEDIS® measures.
Slide 3: Why do we talk about COPD? The next slide is a slide showing the relative change in COPD mortality. It is really one of the few chronic illnesses that have actually demonstrated, over the last 30 years or so, a rising mortality. This slide came from a wonderful review estimating changes in mortality rates for various diseases over the course of about 30 years. These are US-based projections. COPD—on the left—had about a 100% rise in mortality, in the death rate, over that period of time in comparison to the wonderful events that have taken place in heart disease and stroke, where there have been steady decrements in mortality. In fact, the current estimate is that COPD is the fourth leading cause of death in the US, and it is projected to become the third over the course of the next 10 years or so. Clearly there has been a refocus on this disease, noting that, of chronic diseases, it has a dramatically rising mortality. That is one of the reasons that there has been such a push.
Slide 4: The next slide actually shows the second major issue, that COPD diagnosis has been felt to be undergenerated. The slide is titled “COPD Diagnosis Trends by Age.” This is really based on a series of data, a lot based on Dave Mannino and the work that he has done with the NHANES data set, where there have been population-based surveys in the US, including spirometric measures to document airflow obstruction. The estimate has been that about 24 million individuals have evidence of some type of impaired lung function. Interestingly enough, there is an age difference, as you can see: 30% greater than 65, 70% in a younger age range. It is interesting that the estimate has been that only about half of those individuals, or 12 million, have been diagnosed.
Slide 5: The next slide addresses some of the economic considerations. It is an estimate of the cost of managing COPD in contrast to several other chronic disorders. You can see hypertension and stroke cost a lot of money in direct and indirect expenses. But look at COPD, right up there as the third most costly of these chronic conditions. You can start to see the case for why it is so important to have an understanding of how to assess and manage COPD effectively.
Slide 6: The next slide actually addresses an important concept. It is in part related to why the last HEDIS®, or the most recent HEDIS®, measures came out. That is, a major cost driver in COPD is actually hospitalizations. There have been data that suggest patients that have COPD require 2 to 3 times more healthcare utilization than those without COPD. You can see some of the numbers in the bullets that are presented—the respiratory-related hospitalizations, 5000 versus about 2700; nonrespiratory hospitalizations, almost 6000 compared to almost 4000—these have been really consistent data. The thought is that hospitalizations, at least in the US, account for at least 60% of the direct medical costs. You can summarize a lot of the concepts behind COPD in a nice little simple line: it is a common disorder, underdiagnosed, rising mortality, extremely expensive, with a high healthcare utilization burden. You can understand, then, we have got to have a good sense of how to diagnose and manage this chronic disorder. We will now switch to some of the other concepts that will try to provide you with a sense of where the current standards are for diagnosis and management.
Slide 7: The next slide introduces the fact that we are going to talk briefly about risk factors and diagnosis.
Slide 8: The next slide is one that should not surprise you. The current thought is that there is some genetic predisposition. In fact, the NIH just launched a massive multicenter study—the COPD gene study—to try to target that particular component, identifying what genetic abnormalities predispose individuals to the development of a COPD phenotype. It is a 10,000-subject study where multiples are involved. We are going to need all your help to get patients into the study because it is going to be a very complex, but very important, study. It will test that first bullet on the left, the genes, and then there are a series of exposures that individuals have, including cigarette smoke. That is the most prevalent one that everybody realizes, but there are occupational exposures. There has been implication of indoor air pollution and even outdoor air pollution. Then there are a whole series of additional confounders on the right-hand side, including lung growth and development. Prenatal exposures have been implicated. Oxidative stress and gender differences are clearly present—it is a very hot area right now in COPD—age, respiratory infections, and so on. It is very clear that COPD is a complex disorder, and is probably a disorder manifested by some genetic predisposition, and then, boom, some exposure that leads to the phenotypes that we see manifested as COPD.
Slide 9: The next slide highlights one of the ongoing controversies that is in this field. I wish I could tell you that I am going to give you the answer, there is no answer to this yet, and that is the separation of asthma from COPD—2 airway-centered disorders, 2 disorders that have an airway process. On the left-hand side, this actually comes from the latest GOLD guideline. The left-hand side shows asthma, which is felt to be a genetically predisposed individual exposed to some type of an allergen. Then we see a series of pathways that occur, including the important epithelial cell and mast cell interactions—lymphocytes, eosinophils. Eventually, a process characterized by airflow limitation—bronchoconstriction, airway hyperresponsiveness, and airflow obstruction—that can be reversed completely.
On the right-hand side you see already some of the concepts of COPD. There is an exposure—cigarette smoke—and a series of cells, including epithelial cells and macrophages, and again, lymphocytes, and, somewhat different, neutrophils. It is a disorder that is associated with an inflammatory process, also a disorder that has airway abnormalities. It has alveolar destruction and airflow limitation that is not completely reversible. You can see that this idea of being usually reversible versus not completely reversible is one of the major points that we use to separate these diseases. It is a fine point of terminology.
Slide 10: The next slide tries to give you a sense of what would be some of the global characteristics when you approach an individual and you are trying to ask yourself: COPD versus asthma, which one? On the left-hand side you see COPD. It is generally seen a little bit later in life with gradually progressive symptoms, and a person with an exposure—usually a smoking history in the US—symptoms that occur on a regular schedule with some activity, and the largely irreversible airflow limitation. On the right-hand side you see asthma, which is generally seen earlier, often occurring in childhood at first. There is much more symptomatic variation from day to day, from night to morning, a component of A to B, such as allergy, rhinitis, and/or eczema, family history, and largely reversible airflow limitations.
There are guidelines as to how to separate them. I am sure many of you that manage patients in a clinical setting have had some sense of how to apply these to individual patients. There clearly can be overlap. This is an area, even biologically, where there is a lot of evolution going on right now.
It, therefore, brings the concepts of diagnosis. Notice that at the bottom of both of those—the COPD on the left, the asthma on the right—there is this concept of airflow limitation. That is then a pretenet of the COPD approach.
Slide 11: That is the next slide. This is—again, this is a melding of 2 slides—a GOLD slide, and there is a bullet that I put at the bottom that came from a wonderful evidence-based review that was just published by the American College of Physicians. The concept is that the way that we diagnose COPD—because you have a C and an O, chronic obstruction—is to document that there is obstruction. That is done in a very simple, reproducible test called spirometry. I am sure most of you are familiar with it. The current recommendations that GOLD generated were: an individual with symptoms, an exposure—cigarette smoking, occupational exposure, and so on—the diagnostic test, and, in the middle, spirometry. The most recent evidence-based reviews say—I just took this quote from their spirometry question—in patients with respiratory symptoms, spirometry should be performed to diagnose airflow obstruction. That, in fact, is acknowledged as the diagnostic test for that particular diagnosis.
Slide 12: Spirometry really is a remarkably simple technique. The recommendations, as you can see, are to perform it after the administration of an adequate dose of a short-acting bronchodilator so you can determine that concept of irreversible versus largely reversible airflow obstruction. There are very well-defined methodologies for this. You can get spirometers that meet the NLHEP criteria, that meet ATS criteria, and that are remarkably cheap. There are well-publicized predictive equations that are based on age, height, gender, and race. The current guidelines have tried to simplify how to use this diagnostic modality by suggesting that if the FEV1/FVC ratio is less than 70% postbronchodilator, then you know that there is airflow limitation that is now fully reversible. That may change a bit and include some predictives. There is a little bit of under- and overdiagnosis by a function of age, but nevertheless, that less than 70% currently is what people really like to see because it is really easy to measure and it is really easy to remember.
In patients that have an FEV1 greater than 80% of predicted and FEV1/FVC less than 70% of predicted, those have been suggested as potentially having early obstruction, a kind of controversial concept. That last bullet introduces the concept that severity of disease by spirometry is really based on the FEV1 as a percent of predicted. That has been included in most of the international guidelines. You can summarize spirometry by saying it is the diagnostic test for diagnosing COPD. It is easy to do, cheap, and reproducible. Equations to predict for individuals are widely available. The diagnosis of airflow obstruction: FEV1/FVC less than 70%, severity of the chronic obstruction by FEV1 as a percent of predicted. It is really a very straightforward maneuver.
Slide 13: What does it do when you use it in actual practice settings? The next slide actually is a very interesting study that asked the question: What if you actually included spirometry as a diagnostic modality in a primary care setting in individuals that were over 40, smoked, and had symptoms? This is exactly what the recommendations are right now, that you should be considering spirometry. In this particular primary care setting—which was multiple primary care delivery sites—there were about 1500 patients, which is about 2.5% of the population that these clinicians saw during the timeframe of the study. They actually then were encouraged to do spirometry.
The bulk of that spirometry was prebronchodilator and there were some that were without bronchodilator that was administered. You saw that there were some individuals who actually completely followed the recommendation—which is spirometry before and after a bronchodilator—and that is on the left-hand side when you look at prebronchodilator airflow obstruction. That is, before bronchodilator there was airflow obstruction. On the left-hand side, those that actually got a bronchodilator and had obstruction, what proportion of those individuals actually had postbronchodilator airflow obstruction? You have seen that definition. That was almost 400 individuals. You had 1500 total, and about 800 actually had airflow obstruction. In about 470 there was a bronchodilator given. Of those, almost 400 actually had airflow obstruction after bronchodilator. Remember, these are individuals over 40 that had a smoking history and had symptoms. Then one of the more important questions is—and that bottom left-hand box, that group of individuals that after having a bronchodilator still had airflow obstruction, they met the criteria for COPD: more than 40, symptomatic, smoking history, postbronchodilator airflow obstruction—did the clinician do anything with that? Was there anything done with that piece of information? That is actually what is on the next slide.
Slide 14: If you looked at those 400 or so patients, in 139 of those the diagnosis was a new diagnosis. These were new people that were picked up as a result of this spirometric screening approach in high-risk individuals. Look at the proportion of patients that actually were administered anticholinergic on the left-hand side, in the middle that were administered a long-acting β2-agonists, on the right administered an inhaled corticosteroid. The significant differences were in the bronchodilator component on the left. You could argue that if you used spirometry in a primary care setting in a group of individuals at risk, that you did, in fact, identify a lot of airflow obstruction, including postbronchodilator airflow obstruction. In a good chunk of those, about 30% to 40%, there was a new diagnosis of disease. Even more importantly, changes in therapy occurred as a result of that. You can see why there has been this groundswell of push to consider spirometry in individuals that are at risk for COPD. That really has become sort of a mantra in the COPD field.
Slide 15: There still is reluctance, particularly in primary care settings, to use spirometry. The next slide, actually, is a very interesting survey that was carried out in 57 primary care physician offices. They had a response rate of about 60%, which is okay. Two thirds of those that responded had a spirometer, but when they were asked why they were not using the spirometer, there were a whole series of issues that were raised. They were not sure if it would impact care. Remember the previous slide; they were unfamiliar with the test, they had not been trained. What this group did—in those 21 people who responded—they gave them just a 60-minute workshop on spirometry; this is how you do it, this is how you interpret it. There was an almost 60% increase in spirometry in appropriate settings 3 months later, after just a 1-hour workshop. It just goes to show that spirometry is really easy to learn, and we think it is really important to apply that. You will see when Bill talks, that is part of the reason the spirometric component has been one of the tenets of one of the HEDIS® guidelines. It is just such an important component of the COPD arena.
Slide 16:The next slide transitions into some of the concepts regarding therapy.
Slide 17: The next slide, which is titled “Clinical Goals of COPD Treatment,” is from the latest GOLD guideline summary that was published in the blue journal just recently. These are things I think all of us would agree on. We would love to have patients without symptoms, or at least with fewer symptoms. We would love to prevent disease progression. That is one of the holy grails of COPD. Improving exercise tolerance, treating complications, preventing acute flares of disease—I will show you later on why that is so important—then, obviously, reducing mortality. We would love to be able to achieve all of these goals.
Slide 18: What I am going to tell you is that we have therapeutic interventions that can favorably impact all of these. We do have the ability to modify and achieve these goals in our COPD patients. How do we do that? The next slide is a complex slide, because it is a very nice approach that the GOLD committee has set up, which is sort of a step-wise progression to COPD therapy. It is not just pharmacotherapy; it is really cotherapy because there are other therapies included here. The idea is that you can group patients of COPD into varying severities. You can see that at the top there is mild, moderate, severe, and very severe. Then you see the next row over being the spirometric way that we characterize severity—remember, I briefly mentioned that earlier. You probably make a diagnosis of FEV1/FVC less than 70%. How we judge severity is, FEV1 is a percent of predicted, and that is an example of how that is done in one of the algorithms. Then we will go through each of these to give you a sense of what the thoughts are now regarding therapeutic implications.
Slide 19: For example, the next slide removes everything in the bottom. It is sort of the same algorithm approach, but it is now showing us what is recommended across the range of disease severity, across all of the individuals that have some form of COPD, which is clearly avoidance of risk factors, and, second, vaccinations. Those are the first things we are going to talk about.
Slide 20: The next slide is the classic study in the field, the smoking cessation in COPD, The Lung Health Study. That particular study demonstrated conclusively that in mild COPD, if you intervened and individuals stopped smoking—that is the top row here, the sustained quitters—the slope of their lung function over the course of the subsequent 5 years clearly was less steep than if you were in the group that was continuous smokers or intermittent quitters. That held. The differences were even more dramatic at an 11-year time for them. There is no question smoking is bad with regard to lung function, and stopping smoking is good. There is a little bullet that is in the middle on the right-hand side which says the hazard ratio of mortality in the usual care versus special intervention was 1.18. That argument is that, if you were in the smoking cessation group, you had an 18% reduction in mortality over the timeframe that patients were followed than if you had not been in that group. In fact, if you were a sustained quitter, the survival benefit was dramatically higher. There is no question smoke is bad. It worsens lung function and worsens mortality. Stopping smoking is a good thing. Avoidance of smoking is seen and recommended across all the strata of disease severity.
Slide 21: The next slide looks at influenza vaccine. There has been one recent Cochrane review. Here is a summary of 2 of the randomized trials in COPD showing that exacerbations—flares of disease 3 to 4 weeks after influenza vaccine versus a control—clearly were associated. Look at the decrement, the total effect size: -0.39. There is a dramatic benefit that is seen in late exacerbations with influenza vaccine. Therefore, the recommendation across all disease strata has been vaccination.
Slide 22: Now we will look and see what the current recommendation is in individuals that have more moderate disease. Their symptomatic pharmacotherapy should include treatment with a long-acting bronchodilator. What does that involve?
Slide 23: You have got a whole series of choices. The left-hand side is the short-acting, which are the ones that we use to force acute symptom relief. I want you to concentrate on the middle, which are the long-acting. We have a series of long-acting agents that are available, including β2-agonists and long-acting anticholinergics—currently it is tiotropium. We have methylxanthine that is available, but not used very often. We are going to concentrate most of our thinking in the next series of slides on the long-acting β2-agonists and the long-acting anticholinergics.
Slide 24: This is a complex slide. There is the NICE guide, which is a UK, evidence-based guideline which does really a wonderful job. It is a little bit outdated; it is being revised now. It was published in an evidence-based review of a whole series of the literature. I have taken the liberty of adding some of the more recent studies—particularly the Stockley study at the bottom, for example—with my interpretation of benefit. This is just looking at a whole series of really relevant outcomes. For example, there is improvement in dyspnea, improvement in exercise tolerance, improvement in quality of life, and decreasing exacerbations. Anytime you see an arrow pointing up, it suggests that the drug that was tested was better than placebo. This is just the placebo comparisons. You can see that over the course of the last 10 to 12 years there have been a whole series of studies with either salmeterol—which is the S under the medication column—or formoterol, which is the F. There are general improvements that are seen in symptoms, improvements in health status. This is part of the reason that there has been a push toward the use of a long-acting bronchodilator. Clearly, long-acting β2-agonists have that benefit and clearly count as one of the potential options in that setting.
Slide 25: The next slide is the same thing for tiotropium versus placebo—that is the long-acting anticholinergic we currently have available. I have actually added to these NICE guidelines because there has been a whole series of additional studies with tiotropium that have been published more recently. Some of those arrows are my interpretation of what I would imagine. You can see that there has been a very consistent benefit in terms of the improvement in dyspnea, improvement in exercise tolerance more recently documented, improvements in exacerbation rates, and improvements in health status. You can see why the guidelines have been so vocal in saying that people who are symptomatic with COPD, long-acting bronchodilators—you have 2 choices, long-acting β2-agonists or a long-acting anticholinergic—really meet many of the goals of therapy that we set out. These include improving symptoms, improving exercise, improving exacerbation rates, and improving health status. That is why they have taken such a prominent role in guidelines.
Slide 26: The next slide is highlighting a slightly different concept, and that is rehabilitation. What is rehabilitation?
Slide 27: The next slide is a very recent ATS/ERS consensus conference report. It is a multidisciplinary program of care for patients with chronic respiratory impairment. It is individually tailored, designed to achieve some of those great goals; improving physical and social performance and autonomy. It is comprehensive; it includes evaluation and treatment, exercise training, education, and psychological counseling. It is a wonderful approach, and not terribly expensive, mind you. And what does it achieve?
Slide 28: This is a really great systematic review published in the Journal of General Internal Medicine a few years ago. This is just now looking at breathlessness improvements. It segregates the studies by the severity of patients, so that when you look at the left-hand side you see a whole series of studies dating back 10, maybe 15 years or so. Those are mild-to-moderate patients. On the bottom are more severe patients. When you look at the overall benefit, you can see that there is clearly an improved effect size when it comes to breathlessness with rehab that is really seen both in moderate and severe patients. Clearly, rehab is a way of achieving goals, of improving symptoms, and improving exercise capacity.
Slide 29: You see again we are building on this GOLD data. In the more severe patients they suggest inhaled corticosteroids. This is where there has been a lot of movement and a lot of controversy. Where do inhaled steroids come in? Clearly, if you think a patient has asthma, inhaled steroids are the preferred maintenance therapy in those individuals. And so that is, even in milder disease, steroids are clearly and prominently placed in the asthma world. In the COPD world, there has been a lot of movement and there has been a lot of controversy in this area. It is really an evolution of the data.
Slide 30: The next slide is a pulmonary function study of what is currently approved with regard to inhaled steroids in the US by the FDA. It is the combination of fluticasone 250 and salmeterol 50. This is one of the pivotal studies that was used to get an indication for this disease. This was with COPD with chronic bronchitis phenotype, and this was a randomized controlled trial where you had 4 arms—placebo at the bottom, salmeterol and fluticasone kind of in the middle, and then the combination of fluticasone and salmeterol on top. There was clearly a greater improvement with the combination product than with the individual components. This is really in a pulmonary function parameter. That is what is approved from the inhaled steroid point of view in the US. What you have available is a fluticasone/salmeterol 250/50.
Slide 31: The next slide is where you are starting to see some of the issues regarding exacerbations, because this has been such a prominent component. This I actually pulled out from the ACP evidence-based reviews published in the past couple of months. This was the effect in several studies of a combination of a long-acting β2-agonists and an inhaled corticosteroid on exacerbation rate, in comparison to placebo. The subtotal down there shows an almost statistically significant benefit. There are additional data that are coming out that are going to solidify this concept that this combination does, in fact, improve exacerbation rates.
Slide 32: One of those, for example, comes from the TORCH study, which is the next slide. There is a caveat to this, this is with a 500 fluticasone/50 salmeterol, which is not approved in the US for treating COPD. It was the drug dose that was used in this really important study that you cannot avoid when you are talking about COPD because it is such a sentinel study. This study included 4 arms—it was a huge study, almost 6000 individuals—placebo, salmeterol at 50, fluticasone at 500, remember. There is the combination of fluticasone and salmeterol, with, again, the unapproved dose, the 500/50. There was clearly a benefit in moderate-to-severe exacerbations that were seen with the combination, and the combination seemed to be a little bit better than the components. I think when you look at the totality of the data, I would have to agree that the combination of a long-acting β2-agonists and an inhaled corticosteroid does have an exacerbation reduction. You can see why that is being increasingly put within guidelines. Exactly where to put it is where there is still some controversy. At what level should you use it still leads to some level of angst and discussion. By the way, in this TORCH study, there was no exacerbation requirement for patients coming in. In fact, the FEV1 to inclusion went up to as high as 60%. It is sort of expanding a little bit what some of the guidelines said.
Slide 33: The next slide is one of the interesting things that has come out of that TORCH study, and subsequent studies. That is this slide which is titled “Reduction of Exacerbations.” Again, this is the higher dose, the unapproved in the US dose. In fact, in both steroid-containing arms, which are the 2 groups to the right—look at the number of patients in the study, 6000 people—fluticasone and the fluticasone/salmeterol combination, there was an increased incidence during the course of the trial of investigator-reported pneumonias. That has raised a little bit of concern. The difficulty that we have had is—remember the previous slide—exacerbations and acute flares actually were decreased by the combination, although there was this pneumonia signal. We are not exactly sure what to make of that, and that is still something that is in evolution and is being investigated.
Slide 34: The next slide is really the crux of that TORCH study. Remember, that study, as I briefly mentioned, was really meant to examine the hypothesis that the combination of an inhaled steroid—again, in this case the higher dose, the unapproved dose in the US—and salmeterol would improve survival over 3 years in comparison to placebo. These are the final results of that sentinel study. The probability of death is the top set of data. Over 3 was 15.2% in the placebo, 12.6% in the combination therapy arm. The absolute difference between those 2—15.2 minus 12.6—is 2.6. The relative reduction is 17.5%. That is actually pretty good, and it is actually quite good compared to a lot of the other chronic therapies that we use in chronic diseases. The rub is in the bottom part of this slide, what the statistical significance was. There was an unadjusted analysis—that is the top row—with a hazard ratio of .82, an 18% relative reduction, and a P value of .041…so far, so good. But here is the problem, the bottom row is the key. They were 2 interim analyses that were done because the primary safety endpoint—mortality—was the same as the primary efficacy endpoint—mortality reduction. They had to adjust the analysis for these 2 interim analyses and now they got a .825—that is the 17.5% reduction—with a P value of .052. Not .05, but .052.
What does that mean? I am a biostatistician. My masters is biostatistics. To me, this is a study that was weakly positive. There are plenty of explanations as to why they had a relative reduction that was a bit less than they anticipated. There was a differential drop out in the placebo arm, crossover to the active therapy, and so on. This has led to a lot of controversy. I will leave you with that as a whetting of the appetite, because there is going to be, over the course of the next year, a lot of discussion regarding where we position steroids in COPD management.
Slide 35: The next slide—and this will finish the pharmacotherapy component of what I am presenting—is a symptom approach to therapy. I wanted to give equal weight to the ATS/ERS. You have seen the GOLD recommendations, which are based on the algorithm by severity. The ATS/ERS took a slightly different approach. I was on this committee, so I can tell what the arguments were. The arguments were that, when people on the left-hand side—these are people who have COPD that are diagnosed—have symptoms, the first thing you do, to the right, is give them a short-acting bronchodilator. If they were persistently symptomatic—now, mind you, we did not put any criteria for spirometry for guiding therapy—it was based on symptom relief. The recommendation was the same in many ways as the GOLD, a regular long-acting bronchodilator, and an as-needed, short-acting bronchodilator was felt to be the most important. If the patient was still symptomatic, then you added. What did you add? That is where inhaled steroids came in, in combination with long-acting bronchodialaters. At the bottom, you would add or substitute theophylline. The approach is similar, in that the recommendations are the same. Long-acting bronchodilater is a central tenet. Inhaled steroids can be added, particularly if combined with a long-acting β2-agonists.
But when to add them was a little bit different, in that we felt a bit more strongly in arguing that this was based predominantly on symptoms. I am going to show you 2 slides regarding combination, because there is a lot of interest now in the combination of varying interventions.
Slide 36: The first slide I am going to show you is really a wonderful study by Rich Casaburi which combined a pharmacotherapeutic intervention with a nonpharmacotherapeutic intervention. This is a study where individuals were put on either tiotropium or placebo, so a long-acting anticholinergic or a placebo. Then all patients got a pulmonary rehab component, and the endpoint was endurance time. That is the Y-axis. What you can see is, after being put on tiotropium, there was a mild increase in exercise endurance with the drug than placebo. But as soon as you combined it with a nonpharmacological intervention—pulmonary rehab—the difference was dramatically separated. Again, this argues that combining interventions is not a bad idea. This is combining a bronchodilator with pulmonary rehab.
Slide 37: The next slide is a great study from the Canadians, the OPTIMAL study, combining a series of pharmacological interventions. This study had tiotropium plus a placebo, tiotropium, an anticholinergic, plus salmeterol—the long-acting β2-agonists—or tiotropium plus fluticasone/salmeterol. Again, the disclosure is that this is the 500/50 fluticasone/salmeterol, the dose that is not approved for COPD in the US. The primary endpoint is at the top, the proportion that had at least 1 acute exacerbation. There was no difference between the groups. But one of the first predefined secondary endpoints is at the bottom, which was the likelihood of being hospitalized. Look at that—the combination of tiotropium and salmeterol started to look pretty good. The combination of tiotropium plus the fluticasone/salmeterol combination—it is kind of a triple combination—looked even better. There is a lot of movement now in this to be able to test this kind of an idea. What if we combine interventions? If you think about what the GOLD guidelines have recommended and what the ATS/ERS has recommended, this idea of a combination really has been a very important concept.
Slide 38: On the next slide, you also have to have some fair balance to all this. There are great points to all those drugs that we just talked about, but there are some potential side effects for long-acting β2-agonists. You can have some cardiovascular death. This asthma-related death argument has been an issue that has put a black box on the long-acting β2-agonists when it comes to asthma. By the way, that is another reason to consider asthma versus COPD. For the long-acting anticholinergic, a dry mouth is really the major thing that you see, although there are other rarer components that you have to keep in mind. Then the inhaled corticosteroids and LABA really are driven by their oral side effects, those are the side effects of corticosteroids. Then there are a whole series of potential cardiovascular/CNS effects, and that is common. The decreased bone mineral density with inhaled steroids is probably dose related and it is controversial how much of that is to effect. Glaucoma and interocular pressure are something that you probably do not see very often, but you do have to keep it in the back of your mind. The respiratory infections—that pneumonia argument that we made earlier—are something that are currently in evolution and are being argued.
There are a lot of good pharmacological agents, and that is why they are recommended in the guidelines.
Slide 39: The next slide gives you what the guidelines look like in that respect. We will spend the last couple of minutes talking about long-term oxygen therapy and surgical treatments.
Slide 40: The next slide is a classic slide that many of us have used over the years. It shows that there are 2 studies that were done more than 20 years ago that looked at oxygen therapy in individuals with resting hypoxemia that had low PO2 or SAT at rest. There was the MRC, which is a UK study that compared no oxygen versus nighttime oxygen. The bottom survival curve is no oxygen. The thick line that says MRCO2 is the nighttime oxygen. You can see there was clearly an improvement in oxygen therapy. Around the same time, the NIH was doing the nocturnal oxygen therapy trial—the NOTT study—which compared nocturnal oxygen. That is the NIH COT curve with continuous oxygen, which is the top curve. Clearly there was an improvement in survival seen with continuous oxygen. These studies were sentinel studies because they defined oxygen therapy in COPD. Oxygen therapy in COPD is useful in people with resting hypoxemia, and continuous oxygen is better in those people than nighttime oxygen; both are better than no oxygen at all.
Mind you, there is another huge NIH study that is just about to start called the Long-Term Oxygen Therapy Trial, the LOTT. That study is going to test the advantage of oxygen versus no oxygen in patients with mild hypoxemia who do not meet the criteria that these studies defined. I am involved in that study, so I am biased. But that study is going to define oxygen therapy utilization worldwide for the next several decades, if we can get this thing completed. I encourage you to keep that in the back of your mind, because that is going to be another study that is going to require a lot of aid from individuals in the practice setting to refer patients to the centers that are participating in that LOTT study. There is my little plug for that NIH study.
Slide 41: The next slide is “Surgical Options to Treat COPD.” There are several of them, one being lung-volume reduction surgery. I have spent a lot of time with the NIH and CMS studying an aggressive intervention for emphysema. It clearly improves excess capacity and symptoms. In fact, it offers a survival advantage over no surgical therapy. You have to realize that it is for a very, very narrow group of people, particularly for patients that have upper lobe predominant emphysema that still have impaired excess capacity after completing rehab, and maximum medical therapy. You can imagine this was a wonderful study that we completed with the NIH and CMS and Medicare that clearly demonstrated a benefit, but it was an exceedingly narrow group of people. This is now being extended into bronchoscopic techniques, and those are in evolution. The bottom is lung transplantation. Again, COPD and emphysema are the most frequently listed diagnoses for transplant, but that is really a miniscule component of the COPD population.
Slide 42: The last 2 or 3 slides that I am going to highlight are slides that I put in here because it will set up what Bill is going to talk about in terms of the most recent HEDIS® measures. This relates to the management of an acute exacerbation of COPD, the definition of which is: Patients with COPD have variations of the disease, COPD exacerbation is really a sustained worsening of that fluctuation, more than you see from day to day, that is acute onset and has to require some type of intervention. You generally try to exclude something else—pneumonia or heart failure or something of that sort—so it is really an increase in symptoms out of proportion to what the patient normally has, without anything else obvious to explain it.
Slide 43: Why is that important? Because the next slide—which is taken from the GOLD algorithm—says that there is already an acceptance of what we do for therapy in that setting. The first thing is, we increase bronchodilators to improve symptoms. That is the very top. Initiate optimized bronchodilators—that is the current recommendation. Bronchodilators must be used in an exacerbation. Consider antibiotics. When? That is a controversy. At the very least, my own personal bias is that there is a change in sputum characteristics. And then as you move down to the last box on the left side, if a patient is improving, great. If a patient is not improving—on the right—oral steroids are considered.
Slide 44: The last data slide that I will show you is a slide which is a meta-analysis of systematic steroid use in exacerbations of COPD. This is a Cochrane review of steroids in acute exacerbations—9 studies, almost 1000 people. There is no question that steroids reduce treatment failure in this setting. They improve FEV1 greater than placebo. They do not seem to have much impact on mortality, but they do have some side effects.
The summary of all this, particularly with the way to randomize trials, is that if you are dealing with a personal, acute exacerbation—particularly in an emergency room or hospital setting—the role of steroids is very well defined. That is considered now to be one of the standard treatments. Remember that as you think about the most recent HEDIS® measures.
Slide 45: I will conclude that there is clearly a big burden to COPD. You have seen it. It is a common disorder, underdiagnosed, increasing mortality, with big costs to the healthcare system. COPD can be distinguished from asthma. It can be a little bit difficult, but clearly spirometry is the diagnostic modality that we use. It is how we stratify disease severity currently. The treatments are clearly available that meet all of the goals that we have established for COPD—improving symptoms, improving exercise tolerance, and decreasing exacerbations; some even improve mortality. There is clearly an aggressive approach that can and should be used in these patients because you will improve their overall outcome invoking those.
That is my component. Bill will now give you a different view.
Slide 46: Thank you, Dr. Martinez. In this section, I am going to discuss the efforts the NCQA are undertaking to measure and improve the quality of care that COPD patients receive.
NCQA is an independent, non-profit organization whose mission it is to improve the quality of healthcare through measurement transparency and accountability.
Slide 47: One of the key NCQA programs for quality measurement is HEDIS®, the Healthcare Effectiveness Data and Information Set. HEDIS® measures a variety of parameters relating to the quality of care, and those measures are grouped into the domains you see here.
However, most HEDIS® measures are concentrated in the first 3 domains. The effectiveness of care domain includes the HEDIS® measures on preventive care and proper management of chronic conditions, while the access and satisfaction domains largely include information gathered from healthplan members using the consumer assessment of health plan satisfaction, or CAPS, survey.
NCQA has also recently begun measurement of healthcare efficiency in the cost of care domain, using measures of relative resource use, or RRUs. The RRU measures look back at the total annual cost of care for members with chronic conditions, and the COPD RRU measure will be reported for the first time in 2008.
Slide 48: The HEDIS® measure set allows for true plan-to-plan comparisons across a wide spectrum of quality and performance measures. HEDIS® results are used by commercial employers and other health insurance purchasers, as well as state and federal governments, to assess health plan quality.
HEDIS® measure specifications allow organizations that report the measure to choose either electronic data—typically claims and encounter administrative information—or electronic data supplemented by medical record reviews when assessing their own performance.
The HEDIS® measure set is measured by the Committee on Performance Measurement, a broad-based committee of 24 individuals representing employers, consumers, plans, federal and state agencies, etc.
The CPM works to reach consensus on both the HEDIS® measures and the processes to include in the HEDIS® measure set. Assisting the CPM in this work are measurement advisory panels, or MAPs, which comprise experts in specific clinical areas. As with the CPM, the MAP members are not NCQA employees, but rather represent the diverse stakeholders we serve.
For example, Dr. Martinez is a member of NCQA’s respiratory MAP. The MAPs propose and develop new measures in their clinical areas, as well as review existing measure performance over time. At least every 3 years, NCQA does a thorough review of each measure, both to assess progress in the clinical state of the art, and to assess any technical challenges that we may be experiencing with the measure.
Slide 49: In evaluating either a proposed or an existing HEDIS® measure, NCQA concentrates on 3 desirable measure attributes. Those are relevance, scientific soundness, and feasibility.
Relevance refers to either the clinical and/or economic importance of a clinical condition, as well as known quality gaps in treating or providing care for members with that condition.
Scientific soundness is an evaluation of the support for a specific measure in clinical literature and clinical consensus, which includes available practice guidelines and peer review evidence.
Finally, feasibility refers to whether measures could be gathered using available data, and in a cost-effective manner.
Slide 50: Now, as I mentioned, NCQA has several MAPs, and we have had a respiratory MAP for many years. That MAP was instrumental in developing our asthma measures. It was this respiratory MAP itself that determined that COPD was the proper area for new respiratory measure development because of the concerns raised by Dr. Martinez in his presentation. Certainly, the significant health and cost burden of COPD and concerns about quality of care provided to those with the disease—including misdiagnosis of the disease as asthma and underutilization of available guidelines—were all factors in the selection of this disease for HEDIS® measurement.
However, measurement would not have been possible at all without the existence of those consensus guidelines, as they provide a key foundation for measure development.
The first measure launched as part of the HEDIS® measure set for COPD focuses on appropriate diagnosis of COPD using spirometry. Although GOLD guidelines recommend spirometry as a key management tool for COPD to assess both disease severity and the efficacy of treatments, the only time spirometry is recommended for all patients is at diagnosis. That serves as the focus of the NCQA measure.
As you have already heard, spirometry is a vital test, both to rule out asthma and to properly assess the patient. It is underused.
Slide 51: In the HEDIS® measure, eligible patients are those who are 40 years of age or older, and who have a new diagnosis of COPD. Because there is no specific diagnosis code indicating an initial diagnosis of the disease, NCQA defines a new diagnosis using a clean claims look-back period. This means that as long as the patient has not had any claim or encounter with the COPD diagnosis code within the previous 2 years, they are considered a new diagnosis.
We realize that some patients may simply have newly active disease and have been diagnosed several years before, but it is impossible to remove all of these patients using the available data sets.
However, spirometry would be indicated in those patients who have newly active disease, so it is clinically appropriate to include them in this measure.
To show compliance with the measure, the doctor must document evidence of the spirometry test sometime between the 2 years prior to and 6 months after the COPD diagnosis. This extended time period reflects the common clinical practice of using spirometry to assist patients who are at risk for the disease, which may mean that appropriate testing is occurring even before an official diagnosis is made. Of course, the patient must also be a member of the health plan gathering the data for that same 2-year prior or 6-month after time period in order to be eligible for the measure.
Slide 52: The NCQA reports only aggregate results in the first year of a HEDIS® measure, which was 2006 for the spirometry measure. Primary results are only released in the second year, which was 2007 for the spirometry measure, and those commercial rates are shown here on this slide.
The commercial rates show results that are very close, both to the first-year aggregate results and our field testing, which we did in 2004. The results show that spirometry is underused and the problem appears to be nationwide. The commercial product averages shown here were about 35% or 36%, which is the same again as we had in field testing and in first-year results.
The Medicare and Medicaid results were actually lower and are not shown here, but they were around 26% in the first year.
The best reported plan rate, shown here, was just below 60%, which represents one of the lowest performance levels for a relatively new HEDIS® measure. This shows there was a clear need for improvement in this rate across the board.
Slide 53: NCQA did not want to limit our COPD quality measurement to a diagnostic measure, and we have been developing a pharmacotherapy measure for several years. This measure we reported for the first time in 2008.
The measure focuses on exacerbations of COPD because they represent both a significant economic burden to the healthcare system as well as a high risk to the patient’s health. Exacerbations and appropriate treatment of COPD are also an area where proper patient management could substantially reduce the disease impact.
Slide 54: The pharmacotherapy management of exacerbations measure was initially developed in conjunction with the spirometry measure, and was designed to examine the use of pharmacotherapy for COPD patients hospitalized or treated in the emergency department for exacerbation.
Field tests of this measure, however, revealed a significant problem. Many of the hospitalizations that were included in the measure were for nonrespiratory reasons. The COPD diagnosis code was still on the claim form because it was a comorbid condition, but the hospitalization was for an unrelated reason.
Using electronic data as the HEDIS® measure does almost always mean we have some of these ineligible patients in the measure, but NCQA wants to minimize that effect and its impact on the rates. Thus, the CPM recommended the pharmacotherapy measure be respecified and retested to try and reduce this false positive rate.
Slide 55: The newly respecified pharmacotherapy measure still looks at appropriate treatment for an exacerbation treated either on an inpatient or emergency department basis. The COPD diagnosis codes on the claim form must be the primary diagnosis code for a patient to be included in this measure.
Testing of this small change showed a much smaller number of ineligible patients included in the measure, and that is the specification that NCQA chose to go forward with.
To demonstrate compliance with the measure, members hospitalized or treated in the emergency department for a COPD exacerbation must have an appropriate pharmaceutical prescribed. The prescription can be active at the time of treatment or provided within 14 to 30 days from discharge, depending on the drug classification.
Slide 56: NCQA will report 2 rates for this measure, one for clinical steroids and one for bronchodilators, and the bronchodilators include anticholinergics, β2-agonists, and methylxanthines. Although there are other medications—for example, antibiotics—which may be appropriate for some patients having exacerbations, health plans would have to go to their medical records to determine which patients would be appropriate for those other drugs. Because the pharmacotherapy measure uses only electronic data, NCQA made the decision to limit rates to those medications that had both strong clinical evidence of effectiveness and sufficient electronic data to assess prescribing patterns.
Of course, the ultimate goal of both of these measures is to improve the quality and appropriateness of care received by COPD patients. Because each health plan is unique with its own network, geographic service area, population mix, benefit design, and other factors, each health plan will have to analyze their own data on both COPD diagnosis and treatment to determine what barriers and challenges they may be facing in attempting to improve care in this area.
Slide 57: Some common challenges that many health plans may face are listed here, including the problems of mis- and underdiagnosis, which can lead to missed opportunities and late treatment for patients, practitioners’ lack of familiarity with the available guidelines, or the need for lung function testing, the difficulties in removing triggers, particularly smoking, and the general communication difficulties within our healthcare system.
After plans identify the unique set of challenges they face, they will have to implement interventions and other actions to address those challenges and begin the process of improving COPD care.
Next >>
