Stakeholders in the Alzheimer’s disease field often cite the tremendous burden the disease puts on the healthcare system—estimated at $600 billion annually worldwide (see World Alzheimer Report 2010 and ARF related news story)—as a compelling reason to treat the disorder. The argument assumes that an effective, disease-modifying treatment for AD would lower healthcare costs. This may not be the case, according to an analysis presented at the 5th Clinical Trials in Alzheimer’s Disease conference, held 29-31 October in Monaco. Researchers led by Anders Wimo and Bengt Winblad at the Karolinska Institutet, Stockholm, Sweden, modeled a hypothetical disease-modifying treatment and found that the expense of the drug could more than offset the money saved in nursing home care, leading to a higher price tag overall. The scientists stress, however, that the gains in quality of life for AD patients would be big enough to make such a treatment cost effective for society. The research appeared in the October 2 Current Alzheimer Research.

To create the model, first author Anders Sköldunger and colleagues used Swedish data for factors such as the incidence and prevalence of AD and mild cognitive impairment, conversion rates from MCI to AD, and the costs of caring for people with these conditions. They faced a challenge in modeling the disease-modifying drug, however, since no such drug has yet hit the market and therefore no actual data on cost or efficacy exist. The researchers assumed that a hypothetical disease-modifying treatment might lengthen life by about a year. (AD shortens lifespan by two to four years on average, Wimo noted.) They set the cost of the drug at $7,400 per year—about five times the cost of brand-name cholinesterase inhibitors in Sweden. Then they used a 20-year Markov mathematical model to look at the effects of long-term treatment.

Under this scenario, treated AD patients would incur about $35,000 more in lifetime medical costs than untreated patients. The price tag for untreated patients included the cost of nursing home care, home health services, and informal care. Wimo pointed out that most people in the early stages of the disease receive care at home, only incurring expensive nursing home fees in the end stages of the disease. Treated patients were assumed to need less nursing home care, but the cost of the drug itself and patients’ longer life expectancy more than wiped out these savings. The researchers varied cost and efficacy parameters to see how this would affect the final result. For a drug that did not lengthen life, costs came out about equal for treatment or no treatment, Wimo reported.

“Before we did this model, my colleagues and I thought there would be enormous cost savings. It was a bit surprising when we tested all the assumptions and there were no savings,” Wimo told Alzforum.

Should this discourage society from developing disease-modifying drugs? Not at all, Wimo said. Health economists use a measure called “quality adjusted life years” (QALYs) to take quality of life into account when comparing treatments. To arrive at this number, economists multiply life expectancy by the quality of the remaining life-years. On the quality scale, perfect health rates a value of 1, while death equals 0. This quality measure takes into account factors such as mobility, mood, pain, and ability to care for oneself. Some health outcomes are considered worse than death and are given a negative value—for example, a treatment that prolongs life but leaves the patient confined to bed in extreme pain would score below 0. Health economists then divide the QALYs gained from a treatment by its cost to get a rough idea of its value. A drug that costs less per QALY delivers better value compared to one with a high cost per QALY, the thinking goes.

By this measure, the hypothetical disease-modifying treatment scored well. Wimo and colleagues noted that Swedish society is willing to pay about $90,000 per QALY, based on actual decisions made by Swedish health reimbursement authorities. This number may be even higher in the U.S., as some chronic treatments such as kidney dialysis cost as much as $200,000 per QALY (see commentary by economist Alex Tabarrok on the blog Marginal Revolution). The hypothetical Alzheimer’s drug came in at about $43,000 per QALY, less than half the assumed Swedish limit. “For me it’s a signal to go on [developing treatments],” Wimo said.

Howard Fillit, founding director and chief science officer of the Alzheimer’s Drug Discovery Foundation, New York City, noted that models like this one are important, but remain very hypothetical at this point. The cost of the drug is a critical variable, he noted. “We have no idea what any potential disease-modifying drug might cost,” Fillit said, pointing out that some biological agents could be much more expensive than this model assumes.

The difficulty of determining quality of life in AD also complicates the analysis of drug benefit. Most likely, a drug that prolonged the mild stages of the disease would have much more value to people than one that prolonged the later stages, Fillit said. A final challenge for any model is to factor in the comorbidities that typically afflict older people. These also drive healthcare costs and can lead to death from causes other than AD. All of this points to the need for real-world data. “Ultimately, the true economic value of a disease-modifying agent will have to be shown in some sort of cost-effectiveness trial, such as a Phase 4 study comparing the agent to standard of care,” Fillit said.—Madolyn Bowman Rogers

This is Part 7 of a seven-part series. See also Part 1, Part 2, Part 3, Part 4, Part 5, and Part 6. Read a PDF of the entire series.


  1. The results of a model depend, of course, on the assumptions used for the variables in the model. For example, Dr. Wimo and colleagues concluded, in another model, that treatment of dementia with an ACEI
    was cost neutral. A truly successful disease-modifying treatment would allow patients to remain at home, avoiding dementia-related long-term care costs. It is true, though, that any treatment that prolongs
    life among the non-productive aged does have a net cost to society. The issue is, what is the quality of life for those who are treated? Here the discussion devolves into even more speculation. We don’t have good measures of life quality among older people with memory loss.


    . The economic impact of galantamine vs placebo: an analysis based on functional capacity in a Swedish cohort study. J Med Econ. 2012;15(5):1019-24. Epub 2012 Apr 23 PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. World Alzheimer Report: Care Costs 1.0 Percent of World GDP
  2. CTAD: Turning the Ship Around Toward Early Trials
  3. CTAD: New Data on Sola, Bapi, Spark Theragnostics Debate
  4. CTAD: Regulatory Science Gains Prominence in AD Research
  5. CTAD: Adaptive Antibody Trial to Try Bayesian Statistics
  6. CTAD: ApoE Carriers Sought for Trial of NSAID Derivative
  7. CTAD: EEG Gains Luster as More Trials Incorporate Biomarkers

Other Citations

  1. Read a PDF of the entire series.

External Citations

  1. World Alzheimer Report 2010
  2. Marginal Revolution

Further Reading

Primary Papers

  1. . Mortality and treatment costs have a great impact on the cost-effectiveness of disease modifying drugs in Alzheimer's disease. Curr Alzheimer Res. 2012 Oct 2; PubMed.