The cost-utility of the rivastigmine transdermal patch in the management of patients with moderate Alzheimer’s disease in the US Nagy B1, Brennan A1, Brandtmuller A1, Thomas SK2, Sullivan SD3, Akehurst R1 1School of Health and Related Research, University of Sheffield, Sheffield, England; 2Novartis Pharmaceuticals Corporation, East Hanover, NJ, USA; 3Pharmaceutical Outcomes Research and Policy Program, University of Washington, Seattle, WA, USA Figure 2. Predicted MMSE scores using data from a 24-week, randomized clinical trial Table 1. Input parameters for the base case cost-utility and sensitivity analyses in the US. Figure 3. Results for one-way deterministic sensitivity analyses on the base case cost-
extrapolated for up to 5 years (9.5 mg/24 h rivastigmine patch treatment group; 9.5 mg/24 h
effectiveness of the rivastigmine patch from a US perspective. Introduction Parameter Base case value Sensitivity analysis
rivastigmine patch treatment group adjusted for dropout bias; placebo group).
The management of Alzheimer’s disease (AD) in the US is a major healthcare challenge, for
which the costs are substantial. In the US there are more than 5 million people currently
living with AD, and the total economic burden of AD is estimated to be around $100 billion
annually, the majority of which is due to indirect costs of care and institutionalization1.
Indirect costs of care include non-pharmacologic social support and assistance with activities
Probability of NHP (0.490 – 0.014 • MMSE)8
of daily living (ADL), such as home caregiver assistance, support groups, community care,
disease education and nursing home care. The cost of AD is expected to rise considerably
Currently, cholinesterase inhibitors, such as rivastigmine (Exelon®, Novartis), donepezil
Costs of institutional setting ($63,580 pa)6 –$3371
(Aricept®, Eisai) and galantamine (Razadyne®, Johnson & Johnson) and NMDA receptor
antagonist memantine (Namenda®, Forest), form the mainstay of pharmacologic therapies
Costs of community care setting ($13,318 pa)6
to attenuate the symptomatic progression of AD patients.
In 2007, a novel, once-daily rivastigmine transdermal patch was approved in the US for thetreatment of mild-to-moderate AD and Parkinson’s disease dementia (PDD), and in the EU
NHP: Nursing Home Placement; pa: per annum.
for the treatment of mild-to-moderate AD. The rivastigmine patch provides comparable
exposure to the highest doses of capsules (12 mg/day) with improved tolerability, allowing
easier access to optimal therapeutic doses. This in turn has the potential to improve
The number of institutional days avoided was derived from the probability of NHP, and was
This analysis aimed to evaluate the cost-utility of the rivastigmine transdermal patch from
calculated as the difference between the average number of days/year a patient spent in an
the perspective of a US payer, and to provide evidence to support reimbursement of the
institutional setting for the rivastigmine and placebo arms.
rivastigmine patch in the treatment of AD.
Incremental costs in the model included: drug costs, costs of institutionalization and
community care costs. Costs of informal care were not included in the base case cost-utilitycalculations, but were included in the sensitivity analysis. Costs of informal care were
assumed to be higher in the community versus the institution6.
institutionalization; the relationship between MMSE states and quality of life; and the analytic
Costs of institutionalization for the base case were derived from the average direct
perspective adopted. Rivastigmine patch proved to be a dominant strategy, resulting in cost
The aim of this analysis was to model the incremental costs and benefits associated with the
savings, in the majority of sensitivity analyses (Figure 3).
Mortality rates reported in the IDEAL trial were substantially lower than those published
healthcare costs of all patients living in nursing homes or in assisted living facilities7.
rivastigmine transdermal patch versus best supportive care (BSC) in the management of AD,
in epidemiological studies. The cost-utility model incorporated the mortality rates from
Community care costs for the base case were derived from the average direct healthcare
Since patients on the rivastigmine patch arm stayed in the community longer, resulting in
from the perspective of a US healthcare payer.
epidemiological studies, and not those from the IDEAL trial, to reflect the application of the
costs of patients in managed care settings or in academic medical centres7.
delayed institutionalization, inclusion of informal care costs raised the cost-effectiveness
model to real-world clinical practice.
Input parameters for the base case cost-utility analysis are summarized in Table 1. Costs
ratio to $30,899 per QALY gained. Inclusion of informal care, however, did not consider
treatment-related benefits and potential cost-offsets from reduced caregiver burden, etc.
Disease progression was modeled using Mini-Mental State Examination (MMSE) scores
were converted to year 2006 values, and a discount rate of 3.5% in the base case was
extrapolated over a 5-year period by fitting Mendiondo curves3 to the IDEAL data to obtain
applied. Base case analysis did not assume any MMSE-based stopping rule for treatment.
One-way, deterministic sensitivity analysis was used to look at the effect on the model when
The cost-utility of the rivastigmine patch was assessed using an Excel-based economic
using different values for drug costs, utilization parameters, event rates and utilities, varying
evaluation model developed to compare incremental costs and Quality Adjusted Life Years
Utility gains were calculated using prior evidence correlating health-related quality of life
one parameter at a time. The following regression – based on the relationship between the
(HRQoL) directly to MMSE score. Health Utilities Index Version 3 (HUI3)4 point estimates
Health Utilities Index Version 2 (HUI2)4 and MMSE – was used:
were regressed to derive the following equation, which was then used for the base case:
Using this model, rivastigmine patch, from the perspective of a US payer, was calculated to
The findings of the IDEAL study2 (Investigation of transDermal Exelon in ALzheimer’s
achieve cost savings for AD patients. Rivastigmine patch is considered to be a cost-effective
disease), a 24-week randomized double-blind clinical trial with a 28-week open label
extension, provided the efficacy data for the modeling analysis (focusing on 9.5 mg/24 h
QALYs were then calculated by applying this regression function to the mean MMSE score
The main limitations of this analysis relate to the evidence available to quantify long-term
patch treatment groups). The clinical pathway for the current model was populated with
for each cycle of observation, and then multiplying by 0.5 to adjust for the 6-month cycle
efficacy and the costs and consequences of treatment. While an MMSE score is a standard
data from the IDEAL study. Rivastigmine patch data came from patients who received patch
measure of cognitive function, physical functioning scales (ADL) could offer another
for 12 months during the double-blind and open-label extension phases of the IDEAL trial
The probability of nursing home placement (NHP) was estimated using the figures reported
The mean MMSE at baseline for patients modeled was similar for both the rivastigmine
potential means for analysis where data are available. Secondly, estimates of the probability
(n = 383). Placebo data came from patients who received placebo during the double-blind
by Hauber et al5, who built a piecewise Cox proportional hazard model to predict the
patch (16.7) and placebo (16.4) groups, and indicated moderate AD.
of institutionalization are variable within the literature, although sensitivity analyses
likelihood that an individual will be institutionalized at a given MMSE for given values of
provided consistent results favoring cost-effectiveness of rivastigmine patch.
From the perspective of a US payer, the base case cost-utility model predicted that the
To model the patients’ clinical pathway (Figure 1), two assumptions were made: (a) that a
covariates such as age, race and marital status. The Hauber et al model specifies the hazard
rivastigmine transdermal patch (9.5 mg/24 h) would provide an incremental 0.09 QALYs,
In conclusion, using this model the rivastigmine patch demonstrated a favorable
certain number of patients would discontinue treatment, and (b) that some patients would
rate (the likelihood that an individual will be institutionalized) at time t for a given covariate
help avoid 64.3 institutional days over 5 years, and save $1,986 per patient (Table 2).
cost-effectiveness profile, well within the range of currently accepted thresholds for
die. In the IDEAL trial, a discontinuation rate of 21.8% was reported for the 9.5 mg/24 h
vector X as: h(t) = h0(t)eβX, where h(t) is the hazard rate at time t when the covariate is
cost-effectiveness by US healthcare providers.
patch2. Similar rates were applied for the entire 5-year economic analysis, with adjustments
X, h0(t) is the hazard rate at time t when the covariate is 0, X is the vector of covariates,
One-way sensitivity analysis suggested that the results were consistent with varying
to correct for any drop-out bias resulting from discontinuations of rivastigmine treatment.
and β is a vector of unknown parameters to be estimated.
parameters. The main determinants of the cost-effectiveness ratio were: the likelihood of
References Figure 1. Clinical pathways for (a) patients receiving rivastigmine treatment, and (b) patients receiving best supportive care (BSC). Table 2. Base case cost-utility of the 9.5 mg/24 h rivastigmine patch versus best supportive care
Bloom BS, et al. Gerontologist. 2003;43:158–164. et al. Int J Geriatr Psychiatry. 2007;22:456–467. 9.5 mg/24 h Best supportive Incremental et al. Stat Med. 2000;19:1607–1616.
4. Furlong W, Feeny D. McMaster University Centre for Health Economics and Policy Analysis:rivastigmine patch Working Paper no. 98-11. 1998. Benefits et al. Pharmacoeconomics. 2000;17:351–360.
Bell C, et al. Alzheimer Dis Assoc Disord. 2001;15:129–136.
Leon J, et al. Health Affairs. 1998;17:206–216.
Stewart A. Cost of care for people with dementia aged 75 and over. PSSRU discussion paper1303/2. 1997.
Ganguli M, et al. Arch Neurol. 2005;62:779–784.
10. Neumann PJ, et al. Med Decis Making. 2000;20:413–422. Cost-utility of 9.5 mg/24 h rivastigmine patch versus best supportive care
Number of institutional days avoided over 5 years
Poster presented at the American Association for Geriatric Psychiatry Annual Meeting, Orlando, Florida, USA, 14th–17th March 2008
Funding for the study was provided by Novartis.
Werner Bartens: Heillose Zustände. Warum die Medizin die Menschen krank und das Land arm macht, Droemer/Knaur Verlag, München, 2012, ISBN-13: 978-3-426-27581-8, ISBN-10: 3426275813; 223 S. Format 208mm/ 130mm/22mm; Amazon-Preis für gebundene Ausgabe 18 Euro; neu- ab 14,00 Euro; gebraucht- ab 10,99 Euro. Was ist die Medizin? Ist sie eine marktferne, dem Eid des Hippokrates treue und freie Hei
The new england journal of medicine This Journal feature begins with a case vignette highlighting a common clinical problem. Evidence supporting various strategies is then presented, followed by a review of formal guidelines, when they exist. The article ends with the author’s clinical recommendations. A 50-year-old black American has a blood pressure of 160/110 mm Hg on repeated