Abstract
Introduction: Persons with Alzheimer’s disease are at an increased risk of pneumonia, but the comparative risks during specific antidementia treatments are not known. We compared the risk of pneumonia in the use of donepezil, rivastigmine (oral, transdermal), galantamine and memantine.
Patients and methods: We used data from a nationwide cohort of community-dwelling individuals diagnosed with Alzheimer’s disease during 2005–2011 in Finland, who initiated monotherapy with acetylcholinesterase inhibitor or memantine (n = 65,481). The risk of hospitalization or death due to pneumonia was investigated with Cox proportional hazard models.
Results: The risk of pneumonia was higher in persons using rivastigmine patch (n = 9709) (adjusted hazard ratio (HR) 1.15, 95% confidence interval (CI) 1.04–1.27) and memantine (n = 11,024) (HR 1.59, 95% CI 1.48–1.71) compared with donepezil users (n = 26,416) whereas oral rivastigmine (n = 7384) (HR 1.08, 95% CI 0.98–1.19) and galantamine (n = 10,948) (HR 0.91, 95% CI 0.83–1.00) were not associated with an increased risk. These results did not change when adjusting for comorbid conditions, use of psychotropic drugs or with inverse probability of treatment weighting.
Discussion: The increased risk of pneumonia in this fragile group of aged persons should be taken into account. Memantine is associated with the highest risk in the comparison of antidementia drugs.
Pneumonia risk is increased in persons with Alzheimer’s disease who use memantine or rivastigmine patches.
KEY Message
Introduction
Pneumonia is common among the older population and it is responsible for a large proportion of hospitalizations among the elderly (Citation1). Alzheimer’s disease (AD) causes progressive impairment of cognitive functions and it is the most common neurodegenerative disorder among older people (Citation2). Persons with AD are more likely to have pneumonia (Citation3,Citation4), and it is one of the most common causes for hospitalization among persons with AD (Citation5). Persons with dementia are also at an increased risk of pneumonia-related mortality (Citation3). In addition, a recent paper showed an increased risk of pneumonia in persons using AD drug donepezil, than in those using rivastigmine (Citation6).
There is no curative treatment for AD, and therefore its pharmacotherapy aims to slow the rate of progression and is based on two different groups of medicines: cholinesterase inhibitors (AChEI) (donepezil (Anatomic Therapeutic Chemical (ATC) (Citation7) code N06DA02), rivastigmine (N06DA03), galantamine (N06DA04)) and memantine (N06DX01) (Citation8). AChEIs slow down the progression of AD by increasing the concentration of acetylcholine in cholinergic synapses. AChEIs produce small improvements in cognitive function in mild to moderate dementia, with marginal effects in severe disease (Citation9). While AChEIs have similar a mechanism of action, these molecules have differences in pharmacokinetics and adverse effects. In addition, rivastigmine also inhibits butylcholinesterase, and galantamine modulates nicotinic acetylcholine receptors, but the clinical importance of these properties is unknown (Citation10).
Memantine has a different mechanism of action, as it acts on the glutaminergic system blocking NMDA receptors, preventing the deleterious effects of excitotoxicity caused by pathologically elevated levels of glutamate and thus, it produces small benefits in moderate to severe dementia (Citation9) as it has beneficial effects on behavior, activities of daily living and global function (Citation11). It has also been well tolerated without significant adverse effects (Citation8,Citation10) although a lower dose should be considered in case of renal impairment (Citation12).
In Finland, the national guideline for cognitive disorders states that AChEIs are the first-line treatment for mild AD, but the treatment may be started with memantine if there is a contraindication for the use of AChEI. Both AChEIs and memantine can also be used in moderate–severe AD (Citation13).
To our knowledge, there are no reports focusing on the risk of pneumonia associated with the use of memantine or studying risks associated with oral and transdermal forms of rivastigmine. In this study, we compared the risk of pneumonia associated with the use of AChEIs (donepezil, oral and transdermal rivastigmine, and galantamine) and memantine in persons with clinically diagnosed Alzheimer’s disease.
Patients and methods
Ethics
Data were de-identified before submission to the research team and thus, no ethics committee approval was required according to Finnish legislation.
Study population
We used data from the MEDALZ-study (Citation14) including all community-dwelling persons diagnosed with AD during 2005–2011 in Finland. Persons with AD were identified from the Special Reimbursement register including persons entitled for higher reimbursement of drugs due to certain chronic diseases (Citation15). Current care guidelines in Finland recommend that all persons with clinically verified Alzheimer’s disease should be prescribed anti-dementia drugs if there is no contraindication for use (Citation13). For AD reimbursement, the diagnostic process is conducted according to a predefined protocol and approved by the Social Insurance Institution of Finland, which grants special reimbursement when criteria are fulfilled. The diagnostic process includes computed tomography or an MRI scan and confirmation of diagnoses by a geriatrician or neurologist, according to the NINCDS-ADRDA (Citation16) and DSM-IV criteria.
Of 70,718 people with newly diagnosed Alzheimer’s disease in 2005–2011 in Finland, 66,211 persons started using AChEI and/or memantine. In this study, we excluded those persons who started with a combination of antidementia drugs, and the final number of patients in this study was 65,481 monotherapy users. The start of antidementia drug use was considered as index date for this study.
Data sources
Data for the cohort have been collected from the Finnish nationwide registers including the Prescription register (years 1995–2012, drug use), the Special Reimbursement register (1972–2012, comorbidities), and the Hospital Discharge register (1972–2012, hospitalizations with diagnoses). Socioeconomic data since 1970 and causes of death 2005–2012 were from Statistics Finland.
Drug use in the Prescription register maintained by the Social Insurance Institution includes purchases of reimbursed drugs and is classified according to the Anatomical Therapeutic Chemical –classification system (ATC) (Citation7). Drug purchases by community-dwelling persons are recorded in the register, whereas drugs used during stays in hospitals and public nursing homes are provided by the caring unit. Drug purchases were modeled to drug use periods (when drug use started and ended) with a previously utilized method PRE2DUP (Citation14,Citation17). The modeling is based on sliding averages of daily dose for each ATC code for each person. The method takes into account regularity of purchases, hospitalizations, and stockpiling of drugs. We included AChEIs (donepezil, rivastigmine, and galantamine) in addition to memantine. Further, rivastigmine use was modelled as “any rivastigmine”, and separately as oral rivastigmine and transdermal rivastigmine based on the Nordic product numbers of drug packages. In the main analysis, only the first antidementia drug use was considered and the follow-up was censored, if or when persons discontinued usage, started concomitant use of multiple antidementia drugs or switched between these drugs.
Diagnoses of pneumonia were based on data from the Hospital Discharge and the Causes of Death registers. Pneumonia was defined according to ICD-10 as J100, J110, J12, J13, J14, J15, J16, J18 and J690. Only the first event (hospitalization or death) of each person was taken into consideration. The diagnosis and treatment of pneumonia are based on national guidelines (Citation18). From the Causes of Death register, pneumonia recorded as direct or underlying cause of death was extracted.
Comorbid conditions
Data on cardiovascular disease, diabetes (ICD-10 E10-E14, E89.1), rheumatoid arthritis and disseminated connective tissue diseases (short as rheumatoid arthritis) (ICD-10 A04.6, A39.8, A50.5, D76.0, D76.3, H20.1, H30, I33.0, J84, K50.9, K51.9, K73.2, K74.3, K83.0, L40.5, M02, M05, M06, M08, M13.9, M30–M35, M45, M46.1, M46.9, M94.1, N03, N04, Q44.2), and asthma/chronic obstructive pulmonary disease (COPD) (E84.0, J41–J45, P27.1) were collected from the Special Reimbursement register. Cardiovascular diseases were defined as having one or several of the following: chronic heart failure (I11.0, I13, I50, I97.1, P29.0), arterial hypertension (I10–I13, I15, I27.0), coronary artery disease (I20–I22, I24.0, I25) and chronic arrhythmia (I47–I49). Data on history of substance abuse, history of stroke (I60–I64), and previous pneumonia during one year before the follow-up started were collected from the Hospital Discharge register. History of substance abuse was defined as hospitalization based on the diagnoses of alcohol or narcotic use (F10–F19) or alcoholic pancreatitis (K86.0) at any time point before the index date.
Data on the use of antipsychotics (N05A excluding lithium), proton pump inhibitors (PPIs, A02BC), opioids (N02A), antidepressants (N06A), benzodiazepines and related drugs (BZDRs, N05BA, N05CD, N05CF) were collected from the Prescription register. Definition of active cancer was based on use of the antineoplastic and immunomodulating drugs (ATC code L, from the Prescription register) or hospitalization due to cancer (from the Hospital Discharge register) during a one-year period before the index date.
Socioeconomic position was defined as the highest position recorded for study participants in their middle age (45–55 years old), according to classification by the Statistics Finland and categorized into four classes (high, medium, low, unknown). The highest class included entrepreneurs and higher clerical workers, medium class included lower clerical workers and employees, and the lowest class included unemployed and retired persons and students. Persons with unknown socioeconomic class and those with missing data at Statistics Finland (about 5% of the cohort) were combined to class “unknown”.
Statistical analysis
The follow-up started with the initiation of AChEI or memantine use, and ended upon death, change of or discontinuation of antidementia medication, initiation of combination of antidementia medication, >90 days hospitalization/institutionalization, end of study period (at the end of year 2012), or the outcome (hospitalization or death due to pneumonia).
The analyses were conducted using Cox proportional hazard models. Proportional hazard assumption was met for models without time-dependent covariates and for all covariates. Donepezil use was used as a reference drug in all analyses. Main analysis compared other antidementia drugs to donepezil with adjusting for antidepressant, antipsychotic, BZDR and opioid use measured at the index date. In time-dependent analysis, these drugs were modelled time-dependently during the anti-dementia drug use to better account for their possible impact on pneumonia risk. Further, rivastigmine use was divided into oral and transdermal use and in this analysis, change of drug form ended the follow-up, in addition to general criteria for the end of the follow-up. Statistical analyses were performed using SAS statistical software, version 9.3 (SAS Institute, Inc., Cary, NC).
To further take into account the differences between antidementia drug users at baseline, inverse probability of treatment weights (IPTW) (Citation19) were formed. First, we estimated a propensity score for antidementia drug treatment with a multinomial logistic regression model. Factors were selected based on being potentially associated with choice of antidementia drug and/or pneumonia. Factors included in the score are described in Appendix 1. The main analyses were weighted with IPTW based on the propensity score. The covariate balance across treatment groups before and after weighting with IPTW was assessed with standardized differences with values >10%, indicating a meaningful difference (Citation20).
Two sensitivity analyses for main analysis were conducted. Additional duration of 14 days was inserted to the duration of drug use when persons were censored from the analysis due to drug use discontinuation, switch or beginning of concomitant use of two or more antidementia drugs. In the second sensitivity analysis, antidementia drug exposure was treated as time-dependent variable and persons were allowed to change between drugs and use them concomitantly as long as there were no breaks in drug use (discontinuation as censoring event). In this analysis, exposure to rivastigmine, galantamine, memantine, multiple AChEI drugs, and memantine concomitantly with AChEI were compared with use of donepezil only.
Results
Donepezil was the most frequently used antidementia drug (used by 40.3% of the study population), followed by rivastigmine (26.1%), memantine (16.8%) and galantamine (16.7%). Donepezil was initiated on an average of 139 days (median 95) days from diagnosis of AD; the corresponding numbers were 131 (87), 127 (88) and 165 (111) days for rivastigmine, galantamine and memantine, respectively. AD medication was initiated >100 days after AD diagnosis in 47.5% (95% CI 46.9–48.1) of those on donepezil, the corresponding numbers were 46.5% (45.4–47.7), 39.9% (38.9–40.9), 43.3% (42.3–44.2) and 54.4% (53.5–55.4) of persons on rivastigmine (oral and patches), galantamine and memantine, respectively.
During the median follow-up time of 402 days (IQR 150–806), 5095 incident pneumonias were observed (8.4% of the study population). Mean (median) times from initiation of antidementia medication to pneumonia were 584 (439), 518 (378), 654 (515) and 556 (401) days for donepezil, rivastigmine, galantamine and memantine, respectively.
Persons using donepezil had less comorbidities than those using other AChEIs or memantine (). Persons using rivastigmine or memantine used antipsychotics, opioids and BZDRs as well as PPIs more frequently, and had suffered previously from pneumonia more often than those on donepezil. Persons on memantine had active cancer more often than persons on AChEIs.
Table 1. Demographics of the study population (percentage (95% CI), n).
When persons using rivastigmine were compared based on the drug form (oral vs. transdermal), those using transdermal patches were more often female, older, and/or used more often opioids and PPIs, but the use of antipsychotics and BZDRs was less common. In addition, history of substance abuse was less frequent in this group ().
In unadjusted analysis, the risk of pneumonia was higher among users of rivastigmine and memantine compared to those on donepezil, and these risks persisted after adjustments (HRs (95% CI) 1.14 (1.06–1.23) and 1.59 (1.48–1.71) for rivastigmine and memantine, respectively), but not for galantamine ().
Table 2. Risk of pneumonia associated with cholinesterase inhibitors and memantine use.
When the analysis was adjusted for time-dependent use of psychotropic drugs and opioids, these associations did not change. When the main analysis was weighted with IPTW, the results remained similar. However, when oral and transdermal rivastigmine use was separated, only the transdermal drug form was associated with higher risk of pneumonia (HR 1.15, 95% CI 1.04–1.27) although the point estimate for oral rivastigmine use was suggestive of an increased risk (HR 1.08, 95% CI 0.98–1.19).
Sensitivity analyses with 14 days added to the duration of drug use resulted in similar results as the main analyses (). When antidementia drug use was modeled time-dependently allowing switches and concomitant use, the associations between rivastigmine (HR 1.20, 95% CI 1.12–1.29) and memantine (HR 1.71, 95% CI 1.60–1.83) with pneumonia remained. Concomitant use of AChEI and memantine also was associated with an increased risk of pneumonia (HR 1.25, 95% CI 1.17–1.35).
Discussion
In this study we show, that risk of pneumonia in AD patients is higher in those treated with rivastigmine or memantine than with donepezil. This result differs from the study of Lai et al. (Citation6), who found that the risk of pneumonia among patients using AChEIs was the smallest in patients with rivastigmine. They hypothesized, that the beneficial effect of rivastigmine results from greater selectivity for brain tissue than peripheral tissues, resulting in decreased likelihood of unintended respiratory and digestive outcomes (dyspnea, gastroesophageal reflux or esophageal immotility). Our results do not support this hypothesis. One possible explanation to our results might be that fewer patients have been reported to suffer adverse events when using donepezil than rivastigmine (Citation21), but it is likely that this explains only a small part of the result.
Donepezil was the most commonly used AChEI in our study population, which is in agreement with the Swedish results (Citation22) and results from previous, smaller Finnish AD cohort (Citation23). We have previously reported that the use of antidementia medication in this population is in line with the Finnish Care Guideline regarding the treatment of dementia disorders (Citation23).
In the present study, the increased risk of pneumonia in rivastigmine users was more evident among those who used a transdermal drug form than among oral rivastigmine users. Transdermal rivastigmine users were more often women and older than those using oral rivastigmine. In addition, a bigger proportion of them used opioids and PPIs, whereas a smaller proportion of them used antipsychotics as well as benzodiazepines and related drugs, and had a history of substance abuse. Opioids modify the immune system affecting its ability to respond to pathogenic attack (Citation24), but there is no solid evidence whether opioids contribute to the infection risk (Citation25). In addition, there may be an increased risk of pneumonia among persons using PPIs, but the clinical significance is questionable (Citation26,Citation27). However, we controlled for use of these drugs in our analyses. As transdermal rivastigmine has less adverse effects such as nausea and vomiting (Citation28), the risk of pneumonia would have been expected to be smaller in persons using patches rather than capsules. However, in addition to the lower rate of adverse effects, they may be more often used among the most vulnerable group of AD patients and e.g. at older age as in our study. Also, it is possible that a transdermal form is favored in those with poor adherence, possibly due to behavioural and psychological symptoms of dementia (BPSD), but adjustments for psychotropic drug use should at least partly take this into account. Unfortunately, we do not have data on severity of AD or BPSD. To our knowledge, there are no previous studies on prevalence or frequency of BPSD symptoms in those diagnosed at different ages, i.e. are BPSD symptoms more frequent if the diagnoses is made at older age.
The risk of pneumonia among those who used memantine monotherapy was 1.6 times higher than the risk of those on donepezil. Memantine is well tolerated and has less adverse effects than AChEIs, so it is likely that the drug itself is not associated with increased pneumonia risk. Use of memantine is associated with disease severity (Citation29), and therefore the increased risk for pneumonia may be due to more advanced stage of dementia. This may also explain higher pneumonia risk associated with concomitant use of AChEI and memantine compared with donepezil use found in sensitivity analyses. Pneumonia is a common cause of death in AD patients (Citation30,Citation31), and the association with (aspiration) pneumonia and severe dementia has been reported (Citation32–34). Furthermore, multimorbidity is associated with the risk of pneumonia (Citation35), and in our study, e.g. the prevalence of active cancer and history of pneumonia were more frequent in those persons on memantine, although these were adjusted for in the analyses. In addition, the use of memantine was more common among the eldest population than among younger persons.
Memantine is primarily excreted unchanged in the urine (Citation36). As renal dysfunction is common among the aged, this should be taken into account when prescribing memantine. It seems that no dosage adjustments are needed in persons with mild to moderate renal impairment, but the lower dose is recommended for subjects with severe renal impairment (Citation12). A too high dose increases the risk for experiencing potentially problematic adverse effects such as dizziness, headache and somnolence or fatigue (Citation37). On the other hand, although somnolence and fatigue may increase person’s susceptibility to pneumonia, it is likely that these adverse effects have a limited role when considering pneumonia risk.
Antipsychotics have often been used to treat BPSD in AD. Their use is associated with an increased risk of aspiration pneumonia due to the increased latency of the swallowing reflex (Citation33). However, adjusting for the use of psychotropic drugs time dependently during the follow-up did not affect the results in our study, so it seems that the increased risk of pneumonia among memantine users is not due to the use of psychotropics.
The strength of our study was the nationwide cohort that included all persons with new clinically diagnosed AD in 2005–2011. AD medication data was obtained from the Prescription Register and thus, it represents actual drug purchases, which were modeled to drug use periods by taking into account personal drug use behavior and hospitalizations. Further, we were able to differentiate between oral and transdermal use. We also made further analyses to minimize the impact of psychotropic drugs on risk of pneumonia by adjusting their time-dependent use during the follow-up. Because our study is limited to community-dwelling persons, the risk of acquiring pneumonia is closer to the “common people” compared with those living in nursing homes where infections such as pneumonia and flu are more common.
Pneumonia cases in our study were hospitalized or died for pneumonia which excludes pneumonia cases treated in home care and represents the most severe infections. However, the limitation of our study is the lack of data about severity of AD or BPSD. In addition, we did not have data on lifestyle factors such as smoking. Furthermore, we do not have data on frailty status. However, our community-dwelling setting excludes persons who are too fragile to survive on their own. Further, it is unlikely that the risk of pneumonia associated with memantine use would be completely explained by frailty as it would mean that memantine is selectively prescribed to the frailest persons despite the restrictions for use in severe renal dysfunction. Therefore, missing frailty status is not expected to create bias in our results.
The increased risk of pneumonia in AD patients should be remembered when treating those persons. Pneumococcal vaccination has been beneficial in the older population (Citation38) and its use should be considered in these persons with an increased risk of pneumonia.
In conclusion, the risk of pneumonia was higher among patients using rivastigmine patch or memantine when compared to donepezil users. This may be due to the severity of AD and other medical conditions of the patient rather than characteristics of the antidementia medication in use. However, further studies should be conducted regarding the severity of AD impacting the association between antidementia drug use and pneumonia.
Disclosure statement
HT and AT have participated in research projects funded by Janssen with grants paid to the institution where they were employed. JT has served as a consultant at Lundbeck, Organon, Janssen-Cilag, Eli Lilly, AstraZeneca, F. Hoffman-La Roche, and Bristol-Myers Squibb. He has received fees for giving expert opinions at Bristol-Myers Squibb and GlaxoSmithKline, lecture fees from Janssen-Cilag, Bristol-Myers Squibb, Eli Lilly, Pfizer, Lundbeck, GlaxoSmithKline, AstraZeneca and Novartis; and grant from Stanley Foundation. JT is a member of the advisory board in AstraZeneca, Janssen-Cilag, and Otsuka. SH has received a lecture grant from MSD. Other authors declare no conflicts of interest.
References
- Simonetti AF, Viasus D, Garcia-Vidal C, Carratalà J. Management of community-acquired pneumonia in older adults. Ther Adv Infect Dis. 2014;2:3–16.
- Pasqualetti G, Tognini S, Calsolaro V, Polini A, Monzani F. Potential drug-drug interactions in Alzheimer patients with behavioral symptoms. Clin Interv Aging. 2015;10:1457–66.
- Foley NC, Affoo RH, Martin RE. A systematic review and meta-analysis examining pneumonia-associated mortality in dementia. Dement Geriatr Cogn Disord. 2015;39:52–67.
- Manabe T, Teramoto S, Tamiya N, Okochi J, Hizawa N. Risk factors for aspiration pneumonia in older adults. PLoS One. 2015;10:e0140060.
- Rudolph JL, Zanin NM, Jones RN, Marcantonio ER, Fong TG, Yang FM, et al. Hospitalization in community-dwelling persons with Alzheimer’s disease: frequency and causes. J Am Geriatr Soc. 2010;58:1542–8.
- Lai EC, Wong MB, Iwata I, Zhang Y, Hsieh CY, Kao Yang YH, et al. Risk of pneumonia in new users of cholinesterase inhibitors for dementia. J Am Geriatr Soc. 2015;63:869–76.
- WHO Collaborating Centre for Drug Statistics Methodology. The Anatomical Therapeutic Chemical Classification System. Structure and principles; [cited 2016 Mar 10]. Available at http://www.whocc.no/atc/structure_and_principles/.
- Hugo J, Ganguli M. Dementia and cognitive impairment: epidemiology, diagnosis, and treatment. Clin Geriatr Med. 2014;30:421–42.
- Buckley JS, Salpeter SR. A risk-benefit assessment of dementia medications: systematic review of the evidence. Drugs Aging. 2015;32:453–67.
- Rodda J, Carter J. Cholinesterase inhibitors and memantine for symptomatic treatment of dementia. BMJ. 2012;344:e2986.
- Matsunaga S, Kishi T, Iwata N. Memantine monotherapy for Alzheimer’s disease: a systematic review and meta-analysis. PLoS One. 2015;10:e0123289.
- Periclou A, Ventura D, Rao N, Abramowitz W. Pharmacokinetic study of memantine in healthy and renally impaired subjects. Clin Pharmacol Ther. 2006;79:134–43.
- Finnish Medical Society Duodecim, Helsinki (2010). Current care guideline: memory disorders [in Finnish with English summary]. www.kaypahoito.fi. Date last updated: August 13 2010. Date last accessed: March 10 2015.
- Taipale H, Koponen M, Tanskanen A, Tolppanen AM, Tiihonen J, Hartikainen S. High prevalence of psychotropic drug use among persons with and without Alzheimer’s disease in Finnish nationwide cohort. Eur Neuropsychopharmacol. 2014;24:1729–37.
- Tolppanen AM, Taipale H, Koponen M, Lavikainen P, Tanskanen A, Tiihonen J, et al. Use of existing data sources in clinical epidemiology: Finnish health care registers in Alzheimer’s disease research: the Medication use among persons with Alzheimer’s disease (MEDALZ-2005) study. Clin Epidemiol. 2013;5:277–85.
- McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimer’s disease: report of the NINCDS-ADRDA work group under the auspices of department of health and human services task force on Alzheimer’s disease. Neurology. 1984;34:939–44.
- Tanskanen A, Taipale H, Koponen M, Tolppanen AM, Hartikainen S, Ahonen R, et al. From prescription drug purchases to drug use periods: a second generation method (PRE2DUP). BMC Med Inform Decis Mak. 2015;15:21-015-0140-z.
- Finnish Medical Society Duodecim, Helsinki (2015). Current care guideline: lower respiratory infections [in Finnish with English summary]. www.kaypahoito.fi. Date last updated: January 2 2015. Date last accessed: April 11 2015.
- Curtis L, Hammill BG, Eisenstein EL, Kramer JM, Anstrom KJ. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Med Care. 2007;45:S103–S7.
- Austin PC. Goodness-of-fit diagnostics for the propensity score model when estimating treatment effects using covariate adjustment with the propensity score. Pharmacoepidemiol Drug Saf. 2008;17:1202–17.
- Birks JS. Cholinesterase inhibitors for Alzheimer’s disease. Cochrane Database Syst Rev. 2006;CD005593.
- Johnell K, Fastbom J. Concurrent use of anticholinergic drugs and cholinesterase inhibitors. Register-based study of over 700,000 elderly patients. Drugs Aging. 2008;25:871.
- Taipale H, Tanskanen A, Koponen M, Tolppanen A-M, Tiihonen J, Hartikainen S. Antidementia drug use among community-dwelling individuals with Alzheimer’s disease in Finland: a nationwide register-based study. Int Clin Psychopharmacol. 2014;29:216–23.
- Roy S, Ninkovic J, Banerjee S, Charboneau RG, Das S, Dutta R, et al. Opioid drug abuse and modulation of immune function: consequences in the susceptibility to opportunistic infections. J Neuroimmune Pharmacol. 2011;6:442–65.
- Rittner HL, Roewer N, Brack A. The clinical (ir)relevance of opioid-induced immune suppression. Curr Opin Anaesthesiol. 2010;23:588–92.
- Reimer C. Safety of long-term PPI therapy. Best Pract Res Clin Gastroenterol. 2013;27:443–54.
- Masclee GM, Sturkenboom MC, Kuipers EJ. A benefit-risk assessment of the use of proton pump inhibitors in the elderly. Drugs Aging. 2014;31:263–82.
- Nieto RA, Deardorff WJ, Grossberg GT. Efficacy of rivastigmine tartrate, transdermal system, in Alzheimer’s disease. Expert Opin Pharmacother. 2016;17:861–70.
- Törmälehto SM, Martikainen JA, Väätäinen ST, Hallikainen IT, Hallikainen M, Bell JS, et al. Use of anti-dementia drugs in relation to change in cognition, behavior, and functioning in Alzheimer’s disease over a three-year period: Kuopio ALSOVA study. J Alzheimers Dis. 2015;48:1033–41.
- Chouinard J. Dysphagia in Alzheimer disease: a review. J Nutr Health Aging. 2000;4:214–17.
- Attems J, König C, Huber M, Lintner F, Jellinger KA. Cause of death in demented and non-demented elderly inpatients; an autopsy study of 308 cases. J Alzheimers Dis. 2005;8:57–62.
- Kukull WA, Brenner DE, Speck CE, Nochlin D, Bowen J, McCormick W, et al. Causes of death associated with Alzheimer disease: variation by level of cognitive impairment before death. J Am Geriatr Soc. 1994;42:723–6.
- Wada H, Nakajoh K, Satoh-Nakagawa T, Suzuki T, Ohrui T, Arai H, et al. Risk factors of aspiration pneumonia in Alzheimer’s disease patients. Gerontology. 2001;47:271–6.
- Mitchell SL, Teno JM, Kiely DK, Shaffer ML, Jones RN, Prigerson HG, et al. The clinical course of advanced dementia. N Engl J Med. 2009;361:1529–38.
- Ticinesi A, Nouvenne A, Folesani G, Prati B, Morelli I, Guida L, et al. An investigation of multimorbidity measures as risk factors for pneumonia in elderly frail patients admitted to hospital. Eur J Intern Med. 2016;28:102–6.
- Jarvis B, Figgitt DP. Memantine. Drugs Aging. 2003;20:465–76.
- Dodler C, Nelson M, McKinsey J. Memantine dosing in patients with dementia. Am J Geriatr Psychiatry. 2009;17:170–3.
- Tsai YH, Hsieh MJ, Chang CJ, Wen YW, Hu HC, Chao YN, et al. The 23-valent pneumococcal polysaccharide vaccine is effective in elderly adults over 75 years old–Taiwan’s PPV vaccination program. Vaccine. 2015;33:2897–902.
Appendix 1.
Characteristics of antidementia drug users and unweighted and inverse probability of treatment weighted (IPTW) standardized differences comparing covariate distributions between antidementia drug users. IPTW weights were based on propensity score derived with multinomial logistic regression and the score includes all covariates in this table.