Original Research. Jean Lachaine, PhD 1 ; Catherine Beauchemin, MSc 2 ; Anne Crochard, PhD 3 ; Sébastien Bineau, MD, MPH 4

Transcription

1 CanJPsychiatry 2013;58(4): Original Research The Impact of Memantine and Cholinesterase Inhibitor Initiation for Alzheimer Disease on the Use of Antipsychotic Agents: Analysis Using the Régie de l Assurance Maladie du Québec Database Jean Lachaine, PhD 1 ; Catherine Beauchemin, MSc 2 ; Anne Crochard, PhD 3 ; Sébastien Bineau, MD, MPH 4 1 Professor, Faculty of Pharmacy, University of Montreal, Montreal, Quebec. Correspondence: Faculty of Pharmacy, University of Montreal, PO Box 6128, Station Centre-Ville, Montreal, QC H3C 3J7; jean.lachaine@umontreal.ca. 2 Graduate Student, Faculty of Pharmacy, University of Montreal, Montreal, Quebec. 3 Market Access Manager, Lundbeck SAS, Issy-Les-Moulineaux, France. 4 Epidemiology Research Manager, Lundbeck SAS, Issy-Les-Moulineaux, France. Key Words: Alzheimer, cholinesterase inhibitor, memantine, antipsychotics Received July 2012, revised, and accepted November Objective: Patients with Alzheimer disease (AD) show a high incidence of behavioural and psychological symptoms of dementia, which often lead to the prescription of antipsychotics. Our study sought to assess the impact of the initiation of memantine or cholinesterase inhibitors (ChEIs) on the use of antipsychotics. Method: A retrospective cohort study was conducted using data from the Quebec provincial health plan database. Patients included in our study had received a diagnosis of AD and were initial users of memantine or ChEIs. The proportion of patients who used antipsychotics was estimated using prescription data dating back to 1 year before and to 1 year after the first prescription of memantine or ChEIs. The difference between the slopes corresponding to the periods pre- and postmemantine or ChEIs was analyzed using an interrupted time series design. Results: The percentage of antipsychotic users increased by 118.3% before and by 68.3% after initiation of a ChEI, and increased by 68.6% before and by 7.0% after initiation of memantine. Antipsychotic trends pre- and post-chei initiation were not statistically different (P = 0.89), while a statistical difference was observed when comparing the antipsychotic trends pre- and postmemantine initiation (P < 0.001). Conclusions: The initiation of memantine, unlike ChEIs, has a notable stabilization effect on the prescription of antipsychotics in patients with AD. W W W L effet sur l utilisation d agents antipsychotiques de l initiation de mémantine et d inhibiteurs de la cholinestérase pour la maladie d Alzheimer : une analyse menée à l aide de la base de données de la Régie de l assurance maladie du Québec Objectif : Les patients souffrant de la maladie d Alzheimer (MA) présentent une incidence élevée de symptômes de démence comportementaux et psychologiques, ce qui mène souvent à la prescription d antipsychotiques. Notre étude cherchait à évaluer l effet de l initiation de mémantine ou d inhibiteurs de la cholinestérase (IChE) sur l utilisation d antipsychotiques. Méthode : Une étude de cohorte rétrospective a été menée à l aide de la base de données du régime d assurance maladie provincial du Québec. Les patients inclus dans notre étude avaient reçu un diagnostic de MA et utilisaient pour la première fois la mémantine ou les IChE. La proportion de patients qui utilisaient des antipsychotiques a été estimée à l aide des données de prescription datant d un an avant et d un an après la première prescription de mémantine ou d IChE. La différence entre les courbes correspondant aux périodes de pré-mémantine et de post-mémantine ou d IChE a été analysée au moyen d une étude de séries temporelles interrompues. The Canadian Journal of Psychiatry, Vol 58, No 4, April 2013 W 195

2 Original Research Résultats : Le pourcentage d utilisateurs d antipsychotiques a augmenté de 118,3 % avant l initiation d un IChE, et de 68,3 % après, et s est accru de 68,6 % avant l initiation de mémantine, et de 7,0 % après. Les tendances des antipsychotiques avant et après l initiation d IChE n étaient pas statistiquement différentes (P = 0,89), tandis qu une différence statistique était observée lorsqu on comparait les tendances des antipsychotiques avant et après l initiation de mémantine (P < 0,001). Conclusions : L initiation de la mémantine, contrairement aux IChE, a un effet notable de stabilisation sur la prescription d antipsychotiques chez les patients souffrant de MA. As a neurodegenerative disease, AD is characterized by a progressive and irreversible depletion of mental functions. It is the most common form of dementia. About half a million Canadians have AD or a related dementia, and almost 10% of Canadians over the age of 65 are currently affected by this disease. 1 This represents about 1.4% of the overall population. As the population is aging, the disease is expected to reach 1.1 million people by Twice more women are living with AD than men. In 2008, 4606 women died from AD, compared with 1967 men. 2 The financial burden of AD, from a societal perspective, is also substantial, with annual costs per patient varying from $9451 to $ depending on the AD stage. 3 These costs include informal and health care resources associated with the management of AD, and comprise drugs, community support, hospital, and home and medical care. The medications currently available to treat AD are the ChEIs (that is, donepezil, rivastigmine, and galantamine) and the N-methyl-d-aspartate receptor antagonist memantine. 4 ChEIs and memantine are prescribed in Canada for the treatment of AD and reimbursed by the RAMQ as a monotherapy for patients with mild to moderate AD or moderate to severe AD, respectively, and who are living at home. 5 During the course of the disease, patients with AD show a high incidence of BPSD, which have a great impact on the quality of life of both patients and caregivers. 6 8 The consequences of such behavioural problems are the institutionalization and difficulties related to the management of patients with AD by their family and by society. 9,10 Occurrences of BPSD often lead to the prescription of psychotropics and, specifically, antipsychotics. Antipsychotics remain the mainstay of the pharmacological treatment of neuropsychiatric symptoms, although metaanalyses indicate that they are mainly of benefit for the short-term (up to 12 weeks) treatment of aggression in people with AD, and there have been increasing concerns about serious adverse effects, including mortality. 5,11,12 The Abbreviations AD BPSD ChEIs ITS MMSE RAMQ RCT Alzheimer disease behavioural and psychological symptoms of dementia cholinesterase inhibitors interrupted time series Mini-Mental State Examination Régie de l Assurance Maladie du Québec randomized controlled trial 196 W La Revue canadienne de psychiatrie, vol 59, no 4, avril 2013 Clinical Implications The use of antipsychotics in AD patients increases overtime. Memantine stabilizes the need for psychotropics, though the initiation of ChEIs does not have a significant impact on prescriptions. The use of memantine may have a favourable impact on the burden of BPSD in AD. Limitations It was assumed that the patient took the reimbursed medications retrieved from the databases, although this may not always be the case. Medications received during hospitalization and medications that were not reimbursed by the drug plan were not taken into account. As reimbursement criteria do not allow the use of both memantine and ChEIs, the impact of this combination on the use of antipsychotics was not evaluated. use of antipsychotics in dementia is subject to controversy. Clinical practice guidelines, such as those from the Canadian Consensus Conference on the Diagnosis and Treatment of Dementia, recommend that antipsychotics can be used for severe agitation, aggression, and psychosis. Conversely, in the United States, these drugs are contraindicated for the treatment of elderly patients with dementia. Moreover, the Canadian product monographs of antipsychotic agents comprise a serious warning and precaution, indicating that there is an increased mortality among elderly patients with dementia using antipsychotics. 13,14 RCTs of antidementia drugs have demonstrated efficacy for treating and preventing BPSD, potentially leading to a decreasing need for psychotropics A French study 18 performed by Vidal et al in a real-life setting observed that on memantine treatment initiation, the use of psychotropics for BPSD was stabilized. The impact of ChEI initiation on antipsychotic use has not been evaluated yet. Our study sought to assess in the real-life practice the impact of memantine or ChEI treatment initiation on the use of antipsychotics. Methods A retrospective cohort study was conducted using data from pharmaceutical and medical services and beneficiaries demographics retrieved from the RAMQ. Similar to the other Canadian provinces, Quebec has a universal health care program that covers physician services and hospitalizations for the entire population. This universal

3 The Impact of Memantine and Cholinesterase Inhibitor Initiation for Alzheimer Disease on the Use of Antipsychotic Agents health program is complemented, for a large proportion of the population, by a public drug plan also administered by the RAMQ. This provincial drug reimbursement program covers every person aged 65 or older, beneficiaries of the social assistance program, and people without access to a private medication insurance plan. The drug program covers more than 40% of Quebec s population. Unlike the health care plan, under which all costs are covered, the drug plan involves limited financial participation on the part of beneficiaries. The RAMQ medical services database contains information from physicians claims for services provided within and outside the hospital. The RAMQ pharmaceutical services database includes information on pharmacists claims for dispensed medications reimbursed by the program, but not for medications reimbursed in hospital. In addition, data obtained from the RAMQ include an encrypted patient identifier, which enables linkage of individual patient information while preserving anonymity. 19 The RAMQ data are anonymized and are publicly available, at a cost, for research purposes. The RAMQ direction for analysis and management of information manage the distribution of the RAMQ data to external parties. ChEIs and memantine are reimbursed by the RAMQ as a monotherapy for patients with mild to moderate AD or moderate to severe AD, respectively, and who are living at home. According to the reimbursement criteria in Quebec, no patient can receive both types of medication concomitantly. Both typical and atypical antipsychotics were considered in this analysis. All antipsychotics are reimbursed by the RAMQ without any restriction. Study Population In accordance with the RAMQ s restrictions on the number of participants available for analysis by external parties, data on medical services and pharmaceutical prescriptions were obtained for a random sample of participants who received at least 1 diagnosis of AD (International Classification of Diseases, Ninth Revision, code 331.0) during the period from January 1, 2005, to March 31, 2011, and who were participants in the drug reimbursement program. To be included in the analysis, patients needed to be initial users (no previous use for at least 3 months) of memantine or ChEI. Statistical Analysis The proportion of patients who used antipsychotics was estimated using prescription data dating up to 1 year before and up to 1 year after the first prescription of memantine or ChEI. The date of initiation of memantine or a ChEI corresponds to the index date. For each patient, the duration of the period before the index date was limited to the same duration available after the index date. For example, if a patient had 3 months of follow-up data after the index date, the same 3-month period was taken for the pre-index period. Then, for each month up to the year before and after initiation of treatment with memantine or ChEIs, the proportion of patients who used an antipsychotic was estimated. The difference between the slopes corresponding to the periods pre- and post-memantine or ChEIs was analyzed using an ITS design. ITS analysis allows estimating the trend of a given outcome before and after an intervention, as well as the change in the outcome in the post-intervention period. 20,21 The ITS analysis is a statistical method that can be used to analyze estimates that are ordered temporally and to determine if a specific event or intervention has produced a significant change in the trends. As opposed to other statistical methods, for which serial dependency violates underlying assumptions, the ITS analysis accommodates for serial dependency. 22 Results From January 1, 2005, to March 31, 2011, a total of patients had received at least 1 diagnosis of AD. Data from the RAMQ database were obtained on a random sample of of these patents. Among this sample, 8.9% (n = 1929) of patients initiated memantine whereas 91.1% (n =19 787) initiated a ChEI. There were around two-thirds of women among memantine users, as well as in the ChEI group. The mean age at index date was over 80 years in both groups (Table 1). The index date was defined as the initiation date of the ChEI or memantine, while the observation period was up to 1 year before and after the index date. Proportions of antipsychotic users in the 12-month period pre- and post-chei and preand postmemantine are depicted in Figures 1 and 2. In the 12-month period before the initiation of a ChEI, the proportion of patients using antipsychotics rose from 0.06 to 0.13, for an increase of 118.3%, while in the 12-month period after the initiation of a ChEI the proportion of antipsychotic users increased by 68.3% (from 0.13 to 0.22) (Table 2). In absolute terms, the percentage of antipsychotic users increased by 7% before and by 9% after initiation of a ChEI. In the 12-month period before the initiation of memantine, the proportion of patients using antipsychotics rose from 0.24 to 0.41, for an increase of 68.6%, while in the 12-month period after the initiation of memantine the proportion of antipsychotic users increased by 7.0% (0.41 to 0.44) (Table 2). In absolute terms, the percentage of antipsychotic users increased by 17% before and by 3% after initiation of memantine. The ITS analysis indicated that the antipsychotic pre- and posttrends around ChEI initiation were not statistically different (P = 0.89), while a statistical difference was observed when comparing the antipsychotic pre- and posttrends around memantine initiation (P < 0.001). Among the 4 pre- and postintervention trends (preacetylcholinesterase inhibitors, postacetylcholinesterase inhibitors, pre-memantine, and postmemantine), only the beta post-memantine was not statistically significant (β = 0.002, 95% CI to 0.004), confirming the cessation The Canadian Journal of Psychiatry, Vol 58, No 4, April 2013 W 197

4 Original Research Table 1 Characteristics of the study population Characteristic Memantine users n = 1929 ChEI users n = Age group, years, n (%) <60 19 (1.0) 78 (0.4) (43.8) 6079 (30.7) (55.2) (68.9) Mean age (SD) 80.5 (7.8) 82.7 (7.1) Sex, n (%) Males 611 (31.7) 6922 (35.0) Females 1318 (68.3) (65.0) Year of initiation, n (%) (3.1) 1931 (9.8) (19.5) 3994 (20.2) (17.4) 4346 (22.0) (19.2) 4423 (22.4) (23.6) 4895 (24.7) (14.6) 176 (0.9) (2.6) 22 (0.1) Table 2 Percentage of change in the proportion of antipsychotic users in the 12-month periods before and after initiation of ChEIs or memantine Medications Before % After % ChEIs Memantine of the increase or the stabilization in the number of antipsychotic users (Table 3). Discussion Results of our study indicate that the use of antipsychotics in patients with AD increase overtime. Initiation of a ChEI does not have a significant impact on this increase trend. Conversely, initiation of memantine is associated with a significant change in the trend of antipsychotic use. This is in accordance with the finding of clinical trials, which showed that memantine reduces the need for psychotropics. 15,16 The beneficial effect of memantine on BPSD observed in clinical trials is also seen in a real-life setting in our study. Results of our study also confirm the results observed in a similar study performed with data of the French health insurance system, where the use of antipsychotics steadily increased before the initiation of memantine and stabilized afterwards. The change in trends before and after the index date was analyzed with an ITS analysis. This method has a clear advantage over other statistical methods, such as an ANOVA, because it accommodates the serial dependency we face in our study. Serial dependency refers to the correlation of observations of one variable at one point in time with observations of the same variable at prior time 198 W La Revue canadienne de psychiatrie, vol 58, no 4, avril 2013 points. Serial dependency violates assumptions underlying ANOVA and other traditional statistical models. As our study used data from an administrative database, it enabled the inclusion of a very large number of subjects who are assumed to accurately represent a real-world clinical setting, contrary to a clinical trial with a small sample size and controlled clinical environment. However, as in other studies based on administrative databases, there were some inherent limitations. It was assumed that the reimbursed medications retrieved from the databases were taken by the patient, although this may not always be the case. This could result in an overestimation of treatment adherence. Moreover, medications received while hospitalized and medications that were not reimbursed by the drug plan were not taken into account. While donepezil, the first ChEI available in Canada, was reimbursed by the RAMQ starting in 2000, memantine reimbursement started in October It is possible that physicians were less familiar with memantine, than with the ChEIs, and may have been more reluctant to prescribe this medication concomitantly with an antipsychotic. Also, the memantine sample size was much smaller that the ChEI sample size and therefore could be more sensitive to time trends. In our study, the impacts of ChEIs and memantine on antipsychotic use were analyzed separately. Direct comparison of the impact of memantine and ChEIs on the use of antipsychotics would not have been appropriate as the use of these medications are not in the same population. In fact, according to the reimbursement criteria by the RAMQ, ChEIs are reimbursed for patients with an MMSE score of 10 to 26, while the reimbursement criteria for memantine is an MMSE score of 3 to 14. According to the results of a RCT by Tariot et al, 23 memantine combined with a ChEI is associated with significant clinical benefits in terms of cognitive function, ability to perform daily activities, and mood and behaviour compared with a ChEI alone. 23 More recently, an observational study by Lopez et al 24 showed that combining memantine with a ChEI treatment significantly delayed admission to a nursing home in patients with AD. 24 According to these findings, it would have been of interest to estimate the impact of this combination on the need for an antipsychotic. But, given the reimbursement criteria do not allow for the use of both memantine and a ChEI in combination, it was not possible to evaluate the impact of the combination of memantine and a ChEI on the use of antipsychotics. Conclusions The results of this retrospective analysis with the RAMQ database confirmed that the initiation of memantine has a notable stabilization effect on the prescription of antipsychotics in patients with AD as it has already been observed in a previous study. 18 They also showed that, unlike with memantine, the initiation of ChEIs was not associated with such a stabilization effect on the use of antipsychotics.

5 The Impact of Memantine and Cholinesterase Inhibitor Initiation for Alzheimer Disease on the Use of Antipsychotic Agents Figure 1 Proportion of patients using an antipsychotic during the 12-month periods before and after the initiation of a ChEI Proportion of patients Months before and after index date Figure 2 Proportion of patients using an antipsychotic during the 12-month periods before and after the initiation of memantine Proportion of patients Months before and after index date Table 3 ITS estimates according to antipsychotic users and Alzheimer medications β coefficients Alzheimer medication Pre-intervention trend (95% CI) Postintervention trend (95% CI) Change in trends (95% CI) P ChEIs (0.005 to 0.008) (0.005 to 0.008) 0.00 ( to 0.002) 0.89 Memantine (0.014 to 0.016) ( to 0.004) ( to 0.012) < The Canadian Journal of Psychiatry, Vol 58, No 4, April 2013 W 199

6 Original Research Acknowledgements Dr Lachaine and Dr Beauchemin received consulting fees from Lundbeck SAS. Dr Crochard and Dr Bineau are employees of Lundbeck SAS. References 1. Alzheimer Society of Canada. Rising tide: the impact of dementia on Canadian society executive summary [Internet]. Toronto (ON): Alzheimer Society of Canada; 2010 [cited 2011 Feb 8]. Available from: Rich Text Editor/~/media/Files/national/pdfs/English/Advocacy/ ASC_Rising Tide-Executive Summary_Eng.ashx. 2. Statistic Canada. Les principales causes de décès [Internet]. Ottawa (ON): Statistic Canada; 2011 [cited 2011 Feb 8]. Available from: 3. Labelle C. Alzheimer s disease [Internet]. Ottawa (ON): Library of Parliament; 2007 [cited 2011 Feb 8]. Available from: prb0239 1e.pdf. 4. Massoud F, Leger GC. Pharmacological treatment of Alzheimer disease. Can J Psychiatry. 2011;56(10): Ballard C, Day S, Sharp S, et al. Neuropsychiatric symptoms in dementia: importance and treatment considerations. Int Rev Psychiatry. 2008;20(4): Covinsky KE, Newcomer R, Fox P, et al. Patient and caregiver characteristics associated with depression in caregivers of patients with dementia. J Gen Intern Med. 2003;18(12): Ropacki SA, Jeste DV. Epidemiology of and risk factors for psychosis of Alzheimer s disease: a review of 55 studies published from 1990 to Am J Psychiatry. 2005;162(11): Boller F, Verny M, Hugonot-Diener L, et al. Clinical features and assessment of severe dementia. A review. Eur J Neurol. 2002;9(2): Hashimoto M. Pharmacologic therapies of BPSD: messages from psychiatrists. Rinsho Shinkeigaku. 2011;51(11): Senanarong V, Jamjumras P, Harmphadungkit K, et al. A counseling intervention for caregivers: effect on neuropsychiatric symptoms. Int J Geriatr Psychiatry. 2004;19(8): Schneider LS, Dagerman KS, Insel P. Risk of death with atypical antipsychotic drug treatment for dementia: metaanalysis of randomized placebo-controlled trials. JAMA. 2005;294(15): Gill SS, Bronskill SE, Normand SL, et al. Antipsychotic drug use and mortality in older adults with dementia. Ann Intern Med. 2007;146(11): US Food and Drug Administration. FDA Public health advisory: deaths with antipsychotics in elderly patients with behavioral disturbances [Internet]. Silver Spring (MD): US Food and Drug Administration; 2005 [cited 2011 Feb 8]. Available from: PostmarketDrugSafetyInformationforPatientsandProviders/ DrugSafetyInformationforHeathcareProfessionals/ PublicHealthAdvisories/ucm htm. 14. Singh S, Wooltorton E. Increased mortality among elderly patients with dementia using atypical antipsychotics. CMAJ. 2005;173(3): Gauthier S, Loft H, Cummings J. Improvement in behavioural symptoms in patients with moderate to severe Alzheimer s disease by memantine: a pooled data analysis. Int J Geriatr Psychiatry. 2008;23(5): Wilcock GK, Ballard CG, Cooper JA, et al. Memantine for agitation/aggression and psychosis in moderately severe to severe Alzheimer s disease: a pooled analysis of 3 studies. J Clin Psychiatry. 2008;69(3): Rodda J, Morgan S, Walker Z. Are cholinesterase inhibitors effective in the management of the behavioral and psychological symptoms of dementia in Alzheimer s disease? A systematic review of randomized, placebo-controlled trials of donepezil, rivastigmine and galantamine. Int Psychogeriatr. 2009;21(5): Vidal JS, Lacombe JM, Dartigues JF, et al. Evaluation of the impact of memantine treatment initiation on psychotropics use: a study from the French national health care database. Neuroepidemiology. 2008;31(3): Régie de l Assurance Maladie du Québec (RAMQ). Régie: mission et organisation clientèles [Internet]. Montreal (QC): RAMQ; 2011 [cited 2011 Feb 15]. Available from: regie/missorg/clienteles.shtml. 20. Wagner AK, Soumerai SB, Zhang F, et al. Segmented regression analysis of interrupted time series studies in medication use research. J Clin Pharm Ther. 2002;27(4): Ramsay CR, Matowe L, Grilli R, et al. Interrupted time series designs in health technology assessment: lessons from two systematic reviews of behavior change strategies. Int J Technol Assess Health Care. 2003;19(4): Hartmann DP, Gottman JM, Jones RR, et al. Interrupted time-series analysis and its application to behavioral data. J Appl Behav Anal. 1980;13(4): Tariot PN, Farlow MR, Grossberg GT, et al. Memantine treatment in patients with moderate to severe Alzheimer disease already receiving donepezil: a randomized controlled trial. JAMA. 2004;291(3): Lopez OL, Becker JT, Wahed AS, et al. Long-term effects of the concomitant use of memantine with cholinesterase inhibition in Alzheimer disease. J Neurol Neurosurg Psychiatry. 2009;80(6): W La Revue canadienne de psychiatrie, vol 58, no 4, avril