BOSTON, Massachusetts — A new study of Medicare beneficiaries shows that patients with vascular dementia and Parkinson's disease who are first misdiagnosed as having Alzheimer's disease (AD) generate substantial excess costs until they are finally accurately diagnosed.
The finding suggests a potential benefit of more accurate diagnosis of Alzheimer's so that it can be ruled out sooner, said lead author Noam Kirson, PhD, manager at Analysis Group, an economic, financial, and strategy consultancy in Boston.
"We see that a considerable share of these Medicare beneficiaries with Parkinson's disease and vascular dementia have a history of misdiagnosis of Alzheimer's disease, and we also see that those who were misdiagnosed with Alzheimer's disease incur significantly higher medical costs up to and including the period of their correct diagnosis," Dr. Kirson concluded.
"No less important is the decline in costs after the correct diagnosis; that is where we think the policy relevance comes from, because that strongly suggests that if we could arrive at a correct diagnosis earlier, we could potentially eliminate at least some portion of those excess costs," he added.
 |
Dr. Noam Kirson |
The results were presented here at the Alzheimer's Association International Conference (AAIC) 2013. The research was funded by Eli Lilly and Company. Avid Radiopharmaceuticals, a wholly owned subsidiary of Eli Lilly, is the maker of florbetapir, a positron emission tomography (PET) tracer approved for amyloid imaging used as a diagnostic tool in AD.
Complex Diagnosis
One of the biggest challenges to diagnosis of various dementias, including Alzheimer's disease, vascular dementia, and early stages of Parkinson's disease, is the considerable overlap of symptoms, Dr. Kirson noted. Recent development of tools such as PET imaging have introduced the possibility of earlier AD diagnosis, or in the context of this study, to rule out AD, he said, "and that could be very valuable."
In this study, Dr. Kirson and colleagues aimed to assemble the "economic building blocks" to make a case for the cost-effectiveness of these kinds of technologies, where, he said, there is a void of literature to inform public policy.
This was a retrospective analysis of de-identified administrative claims data for Medicare beneficiaries, using a 5% random sample. They looked at 2 mutually exclusive groups of patients, all with 2 or more claims related to vascular dementia or Parkinson's disease based on International Classification of Diseases, Ninth Revision, codes. Among these, the researchers identified groups of patients who had previously received an incorrect diagnosis of Alzheimer's disease.
They then matched each previously misdiagnosed patient with one with no history of misdiagnosis using propensity score methods, and the annual medical costs were then compared between matched pairs at 1-year intervals, stratified by time to correct diagnosis.
Dr. Kirson reported that among the 15,367 patients with vascular dementia, 16.6% had a previous misdiagnosis of Alzheimer's disease. Among the almost 31,000 patients with Parkinson's disease, 8.4% had an AD misdiagnosis. Given how common these conditions are, he noted, these amount to large numbers of patients with prior misdiagnosis.
For the costs, they analyzed 2088 pairs of patients with vascular dementia with and without a previous Alzheimer's misdiagnosis, and 2058 with Parkinson's, again with and without a previous misdiagnosis of Alzheimer's. Baseline characteristics were similar after matching.
Compared with patients with vascular dementia who received a correct diagnosis from the outset, those with a detour to an Alzheimer's diagnosis incurred significantly higher costs: Dr. Kirson reported an excess of $11,967 in year 1 over what was spent on patients with a correct diagnosis, $18,775 in excess costs for patients in whom it took 2 years to reach the correct diagnosis, and costs in excess of $21,000 for those who waited 3 years for a proper diagnosis. These numbers are per patient.
"No less important in our minds is that after a correct diagnosis is attained, those [excess] costs fall, and from a statistical standpoint, they fall to zero," he said.
The same pattern was seen for patients with Parkinson's disease previously diagnosed with Alzheimer's — excess costs per patient were $9526 for those correctly diagnosed in the first year after their previous misdiagnosis, $12,011 for those diagnosed between the first and second year, and $14,241 for those diagnosed between year 2 and year 3. However, those diagnosed at year 2 and year 3 also incurred significant excess costs in years 1 and 2: $7367 in year 1 for those diagnosed at year 2 and $20,626 in years 1 and 2 for those not diagnosed until year 3.
Again, however, after the correct diagnosis was made, the excess costs dissipated, Dr. Kirson said.
Retrospective analysis of claims data has inherent limitations, Dr. Kirson acknowledged, including the fact that they can only control for what they can observe. "If there is any sort of underlying heterogeneity which is not captured in the claims data, that's a limitation of an observational study — which is a long-winded way of saying this is not a randomized trial."
The excess costs were made up from various medical services, Dr. Kirson noted: clinician visits, outpatient visits, inpatient stays, emergency department visits, diagnostic imaging, and laboratory testing — "any interaction with medical services that gets reimbursed by Medicare is in here." They don't have information on this dataset about drugs or costs paid out by other providers, such as Medicaid, for the dual-eligible patients.
Findings Confounded?
After his presentation, session co-moderator Constantine Lyketsos, MD, from Johns Hopkins University in Baltimore, Maryland, said he was "struggling" with the implication that the excess costs were linked to the misdiagnosis. "I think that's very closely confounded with what would call case complexity and I appreciate that the variables you had in your propensity matching were trying to get around that, but…what would be other explanations that would explain these additional costs, and how different would be, for instance, from getting a diagnosis of cancer right? Is this a phenomenon of more complicated presentation as opposed to missing the right diagnosis?"
Dr. Kirson agreed this was an important question, and acknowledged that their analyses can't rule out the possibility that some other unobserved factors, such as clinical severity, are not being picked up, "which is why I would be careful to say that it's not the entirety of these costs that we think are avoidable, and a lot more work needs to go into sort of exactly unpacking how much of this can be saved."
But that said, he added, "I think that the full convergence in post period is strongly suggestive of the fact that there don't seem to be lingering clinical residuals, and I think that gives more credibility to the interpretation that at least a significant chunk here is coming from misdiagnosis itself."
On July 3, the Centers for Medicare & Medicaid Services (CMS) released a proposed decision memo[1] with regard to PET amyloid-β imaging, saying the evidence is insufficient to conclude that use of this imaging "improves health outcomes for Medicare beneficiaries with dementia or neurodegenerative disease," and so is not "reasonable or necessary" under the Social Security Act. However, they did agree there is sufficient evidence to exclude Alzheimer's disease in "narrowly defined and clinically difficult differential diagnoses, such as AD versus frontotemporal dementia" and to help enrich clinical trials for treatment or prevention strategies.
Their decision is posted for comments for 90 days, after which a final decision will be made. The Alzheimer's Association issued a statement[2] at that time expressing disappointment with the decision, pointing out it could hinder both the pace of scientific discovery in the field and the needs of a growing population of patients.
In an interview with Medscape Medical News, William Thies, PhD, the Alzheimer's Association senior scientist in residence, pointed out that, "Dr. Kirson's information was right to the point about the value of a diagnosis, because there is this sense that the only value of a diagnosis is if you get a definitive therapy. But it's perfectly clear that's not the case, and while we tried to make that case for CMS, we didn't have his information, which I thought was very powerful.
 |
Dr. William Thies |
"So some of that may actually help us in the 90-day period and we'll see how it finally comes out," he said. "But in any case, there will still be an opportunity to at least gather the information that will get us to the point where we can say, well this is worthwhile or it's not."
Ralph Nixon, MD, chair of the Alzheimer's Association's Scientific Advisory Committee, who moderated a press conference that included these findings here, pointed out that the association has "long been a proponent" of early diagnosis of Alzheimer's so that patients can participate in early-stage clinical trials and do the planning that is required after such a diagnosis.
These new findings, he said, look at the consequences of making the wrong diagnosis. The medical costs of evaluating that wrong disorder, he said, "were much, much higher than would have been the cost if the correct diagnosis had been made," Dr. Nixon said. Those with a wrong diagnosis were "being deprived of their ability to get the proper treatment, which may be different than what we do with Alzheimer's disease. So there are many adverse implications of making a wrong diagnosis."
Alzheimer's Association International Conference (AAIC) 2013. Abstract DT-02-04. Presented July 17, 2013.
0 comments:
Post a Comment