Abstract
Objective:
We investigated whether anticholinergic drug use was related to developing mild cognitive impairment (MCI) or dementia in older adults at the population level.
Methods:
We used an Anti-cholinergic Rating (ACR) scale, Clinical Dementia Rating (CDR®); APOE genotype; and number of prescription medications. We examined time to incident MCI and incident dementia in a population-based cohort (n=1959). We assessed whether developing MCI or dementia was associated with 1) any anticholinergic drug use, 2) total ACR score, or 3) number of anticholinergic drugs taken.
Results:
Taking any anticholinergic drug was significantly associated with higher risk of developing MCI; however, higher ACR score or higher number of anticholinergic drugs, compared with lower, were not associated with greater risk of developing MCI. We found no significant relationship between anticholinergic use and developing dementia. The relationship between anticholinergic use and cognitive outcome was not affected by APOE genotype.
Conclusions:
Among cognitively normal older adults in a population-based sample, anticholinergic drug use is independently associated with subsequently developing MCI, but not dementia. Thus, anticholinergic drug use may influence risk of MCI that is nonprogressive to dementia and potentially be a modifiable risk factor for MCI.
Keywords: Anticholinergic Burden, Mild Cognitive Impairment, Dementia
INTRODUCTION
Drugs with anticholinergic effects are prescribed for a variety of conditions including Parkinson disease, incontinence, nausea, hyperhidrosis, allergies, and insomnia.1 Common non-prescription, over the counter (OTC), drugs such as some antihistamines also have anticholinergic effects. Whether prescribed or OTC, excess blockade of cholinergic receptors in the central nervous system acutely impairs memory and reaction time, and at high doses causes delirium.2 Many older adults take anticholinergic drugs chronically (e.g., diphenhydramine for insomnia),3 and longer-term, chronic use has been associated with cognitive impairment and dementia.4 These drugs have been associated with greater burden of amyloid plaques in the brains of patients with Parkinson disease,5 and in animals studies, greater anticholinergic burden has been related to increased tau pathology, synaptic loss in regions with cholinergic innervation, and hippocampal neurodegeneration.6–8 Among older adults with genetic risk factors and CSF-based AD pathophysiologic markers, anticholinergic medications significantly increased the risk of incident MCI and cognitive decline.9 However, it is still unclear whether anticholinergic medications enhances AD-related pathology and neurodegeneration.6
A recent Cochrane review addressed whether anticholinergic burden, quantified with a scale, is a prognostic factor for future cognitive decline or dementia in cognitively unimpaired older adults.10 The reviewers found “low-certainty” evidence that older adults without cognitive impairment who take medications with anticholinergic effects may be at increased risk of cognitive decline or dementia. The overall quality of evidence was low and may have exaggerated the strength of the association due to two important sources of bias: 1) protopathic bias (reverse causation) and 2) publication bias. We suggest another potential source of bias, sampling bias. In studies where participants are drawn from clinical settings, these samples are composed of individuals seeking care for the given condition and therefore skewed towards those with those conditions and their most likely outcomes. For example, in most memory clinics, the majority of individuals with mild cognitive impairment (MCI) go on to develop dementia, because they have underlying brain diseases that result in dementia; whereas, in the population setting, most individuals with MCI remain mildly impaired because their impairment is due to a variety of underlying conditions, not all neurodegenerative or cerebrovascular in etiology.11,12 Among the 25 studies that met Cochrane inclusion criteria, five were randomly sampled from the population (“population-based”).13–17
Since 2006, the Monongahela-Youghiogheny Healthy Aging Team (MYHAT) study has been following a cohort of older adults comprising a randomly drawn age-stratified sample from the population represented by the voter registration lists for a group of small-town communities of relatively low socioeconomic status in southwestern Pennsylvania.18 With up to 13 years of annual follow-up assessments, this study provides a relatively unbiased database, relative to a clinical or volunteer sample, in which to examine the effects of anticholinergic drugs on cognitive outcomes in a population-based sample. Our overall goal was to assess whether use of anticholinergic medications was related to the subsequent incidence of mild cognitive impairment (MCI) or dementia. We examined whether anticholinergic burden, as measured with a scale, was related to progression from cognitively normal to MCI or to dementia using the Clinical Dementia Rating (CDR®).19 We hypothesized that anticholinergic drug use would be associated with MCI and/or dementia 1) categorically, 2) at the burden level, or 3) in number of anticholinergic drugs taken. Post-hoc, we investigated a possible interaction effect between anticholinergic burden and APOE*4 genotype on the cognitive outcomes.
METHODS
Our study cohort, called the Monongahela-Youghiogheny Healthy Aging Team (MYHAT), is an age-stratified random sample recruited from the publicly available voter registration list for a group of small towns in Southwestern Pennsylvania, USA. Participants were initially enrolled between 2006-2008. Inclusion criteria at the time of enrollment were a) being 65 years and older, b) living within the selected geographically defined areas, c) not already living in a long-term care setting, d) having sufficient vision and hearing to permit neuropsychological testing, and e) having decisional capacity.18,20 A total of 2036 people provided written informed consent and were briefly assessed using the Mini-Mental State Examination (MMSE).21 Because the study focuses on the epidemiology of MCI, those who exhibited substantial cognitive impairment at study entry by scoring lower than 21/30 on the age-education-corrected MMSE were screened out. The full assessment was then administered to 1982 participants, the majority in their homes. Cognitively normal (CDR=0) participants were considered in the analysis with MCI as the outcome. Participants both with MCI (CDR=0.5) and with those that were cognitively normal were considered in the analysis with dementia as outcome.
The University of Pittsburgh Institutional Research Board approved all study procedures for the protection of human subjects and all participants provided written informed consent.
Assessments at each visit included but were not limited to the following:
Demographic and baseline characteristics:
age, sex, education (less than high school, high school, greater than high school). Race was not included as a covariate because 97% of our original study cohort was of European ancestry.
Medication History:
At baseline and each follow-up assessment, participants were asked to show research interviewers all the prescription and non-prescription (OTC) medications they were currently taking. Interviewers recorded medication names, dosage, and schedule from the bottle or package labels, and verified the details of medication use with participants. All medications were coded by brand and generic names and classified in the study database according to therapeutic category.
APOE*4:
Participants were asked to provide a blood or saliva sample for APOE*4 genotyping.
Anti-cholinergic Rating (ACR)
A pharmacotherapeutics expert (GPS) assigned each prescription and OTC drug an ACR score on an ordinal scale using scores from previously published expert-based lists of medications with anticholinergic activity22–28 as follows; score 0 (medications with no known anticholinergic activity); score 1 (drugs with serum anticholinergic activity but without negative cognitive effects); score 2 (drugs with clinically relevant anticholinergic activity); and score 3 (drugs with high anticholinergic potency). When published scale scores differed, we selected a score where there was sufficient level of agreement. When there was no agreement, we selected the highest score assigned from one of the published reports. Table 1 shows examples of drugs in each ACR category 0-3. ACR coding of drugs was conducted blind to any participant characteristics including the cognitive outcomes. While the ACB scale28 is often taken as a “gold standard,” this scale does not include scores for numerous medications taken by older adults, particularly those with scores of 1 or 2.
TABLE 1:
Examples of prescription and non-prescription drugs and their anticholinergic cognitive burden (ACR) in the MYHAT drug database
| ACR score | Rx or OTC | Category | Generic Name | Brand Name |
|---|---|---|---|---|
| 0 | Rx | ACE Inhibitor | Quinapril | Accupril |
|
| ||||
| 0 | Rx | HMG-CoA Inhibitor Reductase | Atorvastatin | Lipitor |
|
| ||||
| 0 | OTC | Decongestant | Pseudoephedrine | Sudafed |
|
| ||||
| 0 | OTC | NSAID | Naproxen | Aleve |
|
| ||||
| 1 | Rx | Selective-Serotonin Reuptake Inhibitor | Sertraline | Zoloft |
|
| ||||
| 1 | OTC | Laxative | Bisacodyl | Dulcolax |
|
| ||||
| 1 | Rx | Calcium Channel Blocker | Diltiazem | Dilacor |
|
| ||||
| 1 | OTC | Antidiarrheal | Loperamide | Imodium |
| 1 | OTC | Antihistamine | Loratadine | Claritin |
|
| ||||
| 2 | Rx | Anticonvulsants, Misc. | Carbamazepine | Tegretol |
|
| ||||
| 2 | Rx | Opiate Agonists | Meperidine | Demerol |
|
| ||||
| 2 | Rx | Tetracyclic Antidepressant | Mirtazapine | Remeron |
|
| ||||
| 2 | Rx | Respiratory Smooth Muscle Relaxants | Theophylline | Theodur |
|
| ||||
| 3 | OTC | Antihistamine | Diphenhydramine | Benadryl |
|
| ||||
| 3 | Rx | Antimuscarinics | Oxybutynin | Ditropan |
|
| ||||
| 3 | Rx | Antipsychotic | Olanzapine | Zyprexa |
|
| ||||
| 3 | Rx | Centrally Acting Skeletal Muscle Relaxants | Cyclobenzaprine | Flexeril |
| 3 | Rx | Antispasmodic | Dicyclomine | Bentyl |
For the present analyses we included both prescription and OTC drugs and classified participants according to (a) whether they were taking any anticholinergic drug (any ACR score 1-3), (b) their total ACR score summing the scores of all the anticholinergic drugs they were taking (e.g., participants taking two medications scored 1 and 3 were assigned a total score of 4; participants taking two medications scored 2 and 3 were assigned a total score of 5, etc.), and (c) the total number of anticholinergic drugs taken. Non-prescription drugs and nutritional supplements were scored using the same scale used for prescription drugs.
Dementia Rating:
Trained interviewers rated each participant at each assessment cycle on each of the CDR functional domains based on independence in cognitively driven everyday functioning.18,29 The CDR explicitly excludes consideration of neuropsychological test performance. The global CDR ratings (stages) of 0, 0.5, and ≥1 indicate normal cognition, MCI, and at least mild dementia stages, respectively.
Outcomes:
The primary outcomes were defined as time to incident MCI (CDR=0.5) and time to incident dementia (CDR>=1), respectively, as described further below.
STATISTICAL ANALYSES
Descriptive statistics.
Participants’ demographic characteristics were calculated as mean (SD) for age and frequency (proportions) for sex and educational level (less than high school, high school, more than high school). Total number of prescription medications was presented by median [lower quartile Q1; upper quartile Q3]. Proportion of participants taking any anticholinergic drug (ACR score 1-3) was calculated. The distribution of baseline total ACR score and total number of anticholinergic drugs by age/sex/education groups are displayed by bar plots. To compare characteristics of participants who took and did not take anticholinergic drugs at baseline, chi-square test was conducted for categorical variables; and t-test and Wilcoxon-Mann-Whitney test for continuous variables.
Adjusted Cox models were used to test whether anticholinergic drug use was associated with time to developing MCI (CDR=0.5) or dementia (CDR>=1). For those who developed MCI or dementia during follow-up, time to MCI was calculated as time (in years) from baseline (cycle 1) to the cycle that the participant was first rated CDR=0.5, and time to dementia was calculated as time (in years) from baseline (cycle 1) to the cycle that the participant was first rated CDR>=1. For those who did not develop MCI or dementia during follow-up, time from baseline to the last available assessment was calculated.
Covariates used in the adjusted Cox models were age, sex, education, and total number of prescription medication (to control for overall morbidity). We examined 3 versions of anticholinergic drug use: any anticholinergic drug use (binary), total ACR score, and total number of anticholinergic drugs. Categorical versions of total ACR score (categorized into 3 levels based on quartiles: 0, 1-3, >3) and total number of anticholinergic drugs (categorized into 3 levels based on quartiles: 0, 1-2, >2) were used to fit the Cox models to relax violating linearity assumption between log-hazard and the two covariates. Total number of prescription medications was categorized into 3 levels based on quartiles: <3, 3-6, >6. Age, anticholinergic drug use, and total number of prescription medications were treated as time-dependent covariates. Adjusted hazard ratios with 95% confidence intervals (CIs) were calculated, as well as the corresponding p-values. The proportional hazards assumption was assessed using Schoenfeld residuals. Multicollinearity was checked with variance inflation factor (VIF). All statistical analyses were conducted using R statistical package, version 3.6.3.30
RESULTS
Baseline characteristics of anticholinergic drug users and non-users
Over the course of the study the cohort (n=1959) had 81% attrition with 373 participants remaining in the study at the 13th year point. The median length of follow-up was 5.1 years (lower quartile 2, upper quartile 10.1). At study entry/baseline, 1959 participants were free of dementia (CDR<1). Four participants had no anticholinergic drug information at baseline but did have it at subsequent assessments. Among the remaining 1955 participants (Table 2), mean (SD) age was 77.6 (7.4) years, 61.3% were women; 13.5%, 45.2%, and 41.3% had education less than high school, equal to high school, and higher than high school, respectively. Regarding anticholinergic drug exposure, 69.5% participants took at least one anticholinergic drug (with any ACR score 1, 2, or 3). At baseline, age, sex, and education were not significantly associated with any anticholinergic drug use. Participants who took any anticholinergic drug were more likely taking a higher number of prescription medications. Total ACR score ranged from 0 to 13, with median (Q1; Q3) being 1 (0; 3). Total number of the anticholinergic drugs ranged from 0 to 10, with median (Q1; Q3) being 1 (0; 2). Their distributions are presented in FIGURE 1 in bar plots, stratified by age (panel A and D), sex (panel B and E) and education (panel C and F) groups, respectively. We grouped age into 3 groups: 65-74, 75-84, and >=85.
TABLE 2.
Baseline demographic characteristics by any anticholinergic drug use
| All (n=1955) | any anticholinergic drug: no (n=596) | any anticholinergic drug: yes (n=1359) | P value | |
|---|---|---|---|---|
| Age | 77.6 (7.4) | 77.1 (7.3) | 77.8 (7.4) | 0.059* |
|
| ||||
| Female | 1198 (61.3%) | 360 (60.4%) | 838 (61.7%) | 0.634** |
|
| ||||
| Education | 0.090** | |||
| < high school | 264 (13.5%) | 78 (13.1%) | 186 (13.7%) | |
| = high school | 883 (45.2%) | 250 (41.9%) | 633 (46.6%) | |
| > high school | 808 (41.3%) | 268 (45.0%) | 540 (39.7%) | |
|
| ||||
| Total number of prescription medications | 4 [2; 6] | 1 [0; 3] | 5 [3; 7] | <0.001*** |
Table shows mean (SD), or frequency (%), or median [Q1; Q3]. P values reflect tests for significant differences between any anticholinergic drug use groups using the following tests:
t test;
Chi-squared test;
Wilcoxon-Mann-Whitney test.
FIGURE 1:
Distribution of baseline total Anticholinergic Burden (ACR) score and total number of anticholinergic drugs by age, sex, and education groups
Longitudinal analyses: association between anticholinergic drug use and incident MCI
Among the 1959 participants with CDR<1 at baseline, 546 participants who already had CDR=0.5 at cycle 1 were excluded from the longitudinal analysis, and 154 participants were further excluded because they were lost to follow-up after baseline. During the follow-up, 405 participants developed MCI (CDR=0.5). We examined the associations between 3 different versions of anticholinergic drug use (any anticholinergic drug use, total ACR score, and total number of anticholinergic drugs) and developing MCI (CDR=0.5) (TABLE 3). Analysis samples included those with CDR=0 at baseline (cycle 1).
TABLE 3.
Adjusted hazards ratios for anticholinergic drug use on developing MCI#
| Model 1 | Model 2 | Model 3 | ||||
|---|---|---|---|---|---|---|
| Variable | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value |
| Age | 1.1 (1.1, 1.1) | <0.001 | 1.1 (1.1, 1.1) | <0.001 | 1.1 (1.1, 1.1) | <0.001 |
|
| ||||||
| Female | 1.2 (1.0, 1.5) | 0.120 | 1.2 (1.0, 1.5) | 0.115 | 1.2 (1.0, 1.5) | 0.121 |
|
| ||||||
| Education | ||||||
| < high school | Reference | Reference | Reference | |||
| = high school | 0.8 (0.6, 1.0) | 0.070 | 0.8 (0.6, 1.0) | 0.067 | 0.8 (0.6, 1.0) | 0.073 |
| > high school | 0.6 (0.4, 0.8) | 0.001 | 0.6 (0.4, 0.8) | 0.001 | 0.6 (0.4, 0.8) | 0.001 |
|
| ||||||
| Total number of prescription medications | ||||||
| <3 | Reference | Reference | Reference | |||
| 3-6 | 1.1 (0.8, 1.4) | 0.608 | 1.1 (0.8, 1.4) | 0.591 | 1.1 (0.8, 1.4) | 0.659 |
| >6 | 1.2 (0.9, 1.6) | 0.274 | 1.2 (0.9, 1.7) | 0.245 | 1.1 (0.8, 1.6) | 0.469 |
|
| ||||||
| Take any anticholinergic drug | 1.4 (1.0, 1.8) | 0.029 | ||||
|
| ||||||
| Total ACR score | ||||||
| 0 | Reference | |||||
| 1-3 | 1.4 (1.0, 1.8) | 0.026 | ||||
| >3 | 1.3 (0.9, 1.8) | 0.15 | ||||
| >3 versus 1-3 # | 0.9 (0.7, 1.2) | 0.687 | ||||
|
| ||||||
| Total number of anticholinergic drugs | ||||||
| 0 | Reference | |||||
| 1-2 | 1.3 (1.0, 1.8) | 0.037 | ||||
| >2 | 1.5 (1.0, 2.1) | 0.037 | ||||
| >2 versus 1-2 * | 1.1 (0.8, 1.4) | 0.508 | ||||
Abbreviations: HR, hazard ratio; CI, confidence interval.
MCI was defined as clinical dementia rating (CDR) score =0.5.
is comparing total ACR score of >3 to 1-3 (the reference)
is comparing total number of anticholinergic drugs of >2 to 1-2 (the reference).
Models 1, 2, 3 used any anticholinergic drug use, total ACR score, total number of anticholinergic drugs as the main predictor, respectively.
Taking any anticholinergic drug was significantly associated with higher risk of developing MCI (CDR=0.5). The expected hazard was 40% higher in those who took any anticholinergic drug as compared to those who took none, adjusting for age, sex, education, total number of prescription medications (TABLE 3, Model 1).
We further assessed the association between total ACR score and risk of developing MCI (CDR=0.5). Compared to those with a total ACR score of 0 (taking no ACR drug, the reference group), taking anticholinergic drugs with total ACR scores of 1-3 was significantly associated with a 40% higher risk of developing MCI, adjusting for age, sex, education, total number of prescription medications. Compared to total ACR scores of 0, taking anticholinergic drugs with total ACR scores of >3 was associated with a 30% higher risk of developing MCI, but the difference was not statistically significant (p=0.15). Taking anticholinergic drugs with a total ACR score >3 was not significantly associated with higher risk of developing MCI (CDR=0.5), compared to having a total ACR score of 1-3 (TABLE 3, Model 2).
When using the total number of anticholinergic drugs as the predictor, compared to taking no anticholinergic drugs, taking 1-2 or >2 anticholinergic drugs was significantly associated with higher risk of developing MCI, respectively, adjusting for age, sex, education, total number of prescription medications. However, taking >2 anticholinergic drugs was not significantly associated with higher risk of developing CDR=0.5, compared to taking 1-2 anticholinergic drugs (TABLE 3, Model 2).
Longitudinal analysis: association between anticholinergic drug use and incident dementia
Similarly, we assessed the association between the 3 versions of anticholinergic drug use and developing dementia (CDR>=1). Analysis samples included those with CDR<1 at baseline (cycle 1). Among the 1959 participants who had CDR<1 at baseline, 135 participants developed CDR >= 1 during follow-up; 258 participants were further excluded for loss to follow-up after cycle 1. We did not find significant associations between anticholinergic drug use and developing dementia (see Models 4, 5, 6 in TABLE 4).
TABLE 4.
Adjusted hazards ratios for anticholinergic drug use on developing dementia#
| Model 4 | Model 5 | Model 6 | ||||
|---|---|---|---|---|---|---|
| Variable | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value |
| Age | 1.1 (1.1, 1.2) | <0.001 | 1.1 (1.1, 1.2) | <0.001 | 1.1 (1.1, 1.2) | <0.001 |
|
| ||||||
| Female | 1.0 (0.7, 1.5) | 0.799 | 1.0 (0.7, 1.5) | 0.846 | 1.0 (0.7, 1.5) | 0.812 |
|
| ||||||
| Education | ||||||
| < high school | Reference | Reference | Reference | |||
| = high school | 0.7 (0.4, 1.1) | 0.126 | 0.7 (0.5, 1.1) | 0.137 | 0.7 (0.5, 1.1) | 0.130 |
| > high school | 0.6 (0.4, 1.0) | 0.040 | 0.6 (0.4, 1.0) | 0.042 | 0.6 (0.4, 1.0) | 0.041 |
|
| ||||||
| Total number of prescription medications | ||||||
| <3 | Reference | Reference | Reference | |||
| 3-6 | 0.9 (0.6, 1.4) | 0.563 | 0.9 (0.6, 1.4) | 0.524 | 0.9 (0.5, 1.4) | 0.515 |
| >6 | 0.9 (0.6, 1.6) | 0.824 | 0.9 (0.5, 1.5) | 0.647 | 0.9 (0.5, 1.6) | 0.641 |
|
| ||||||
| Take any anticholinergic drug | 0.9 (0.6, 1.4) | 0.641 | ||||
|
| ||||||
| Total ACR score | ||||||
| 0 | Reference | |||||
| 1-3 | 0.9 (0.5, 1.4) | 0.516 | ||||
| >3 | 1.1 (0.6, 1.9) | 0.825 | ||||
| >3 versus 1-3 # | 1.2 (0.8, 1.9) | 0.340 | ||||
|
| ||||||
| Total number of anticholinergic drugs | ||||||
| 0 | Reference | |||||
| 1-2 | 0.9 (0.6, 1.4) | 0.573 | ||||
| >2 | 1.0 (0.6, 1.9) | 0.915 | ||||
| >2 versus 1-2 * | 1.2 (0.7, 1.9) | 0.498 | ||||
Abbreviations: HR, hazard ratio; CI, confidence interval.
Dementia was defined as clinical dementia rating (CDR) score ≥1.
is comparing total ACR score of >3 to 1-3 (the reference)
is comparing total number of anticholinergic drugs of >2 to 1-2 (the reference).
Models 4, 5, 6 used any anticholinergic drug use, total ACR score, total number of anticholinergic drugs as the main predictor, respectively.
Post-Hoc Analysis of APOE*4
For all the models, we included an interaction term between ACR drug use and APOE*4. The interaction term effect was not significant in any of the models, suggesting the effect of ACR on the outcome is not statistically different in people with and without APOE*4 (See Supplementary TABLEs 1 and 2).
DISCUSSION
Our analysis of data from a population-based sample of cognitively normal older adults from small-town communities in Southwestern Pennsylvania revealed that any anticholinergic drug use was associated with elevated risk of subsequent development of MCI, but not of dementia. Unexpectedly, however, higher ACR score and greater number of anticholinergic medications were not associated with greater risk of developing MCI compared with lower ACR or fewer anticholinergic medications. First, our findings support the understanding that anticholinergic drug use is related to cognitive impairment. We also found that anticholinergic drug use predicts MCI but not dementia. This apparent paradox is consistent with the now well-established finding that, at the population level, most individuals with MCI do not progress to dementia.11,12,31,32 This phenomenon is at odds with the typical outcome in specialized memory clinics where most MCI is prodromal dementia, likely reflecting the underlying etiology of the MCI and subsequent dementia in different settings. Put simply, individuals seeking assessment at Alzheimer disease clinics have an elevated probability of having Alzheimer’s disease or another progressive brain disorder. In the general population, there are additional, including possibly reversible, causes of cognitive impairment that would not be seen in patients referred to specialized memory clinics.
Previously, we identified characteristics of MCI that progresses to dementia, remains stable, or reverts to normal cognition in our study cohort.11 We found that most individuals who had MCI that remained stable without progressing to dementia were characterized by taking >3 prescription drugs, having diabetes and low diastolic blood pressure, attributed to medication effects or heart failure and the APOE*4 genotype which is an established risk factor for MCI and dementia. Potentially, anticholinergic drug use might contribute to this stable category of MCI due to their acute effects on cognition.
With MCI that progresses to dementia, prior reports have suggested that increased brain atrophy may be linked to the central effects of taking anticholinergic drugs.33 In AD mouse models, cholinergic receptor antagonists induce cell death, while increased cholinergic neurotransmission reduces neurodegeneration.34,35 These findings suggest that anticholinergic drug use might accelerate expression of pathological processes in the subgroup of individuals with MCI due to AD. However, in individuals without neurodegeneration, anticholinergic drugs might cause cognitive deficits simply by blocking cholinergic transmission, which would be potentially reversible if the drugs were stopped.
Our findings add nuance to the recent meta-analysis from Cochrane finding evidence, albeit of low certainty, that cognitively normal older adults who take medications with anticholinergic effects may be at increased risk of both cognitive decline and dementia. The Cochrane review included several population-based studies but also some studies based in primary and secondary care settings, and both cohort studies and case-control studies, which among them used 10 different methods of measuring anticholinergic risk or burden, and multiple approaches to classifying study participants’ cognitive decline, MCI and dementia, with one to eleven years of follow-up. Our disparate findings may therefore be related to differences in study settings and design, ours being set in a large, well-characterized, and unbiased cohort of 1959 older adults, prospectively assessed for up to 13 years, using self-reported prescription and OTC drug use, and the CDR for MCI and dementia rating.
Four of the five population-based reports cited by the Cochrane review also failed to link a relationship between anticholinergic medications and dementia. Whalley and colleagues17 recruited 281 volunteers at age 77-78 without overt dementia who had participated in the Scottish Mental Survey as children in 1932. During the 10-year follow-up, while anticholinergic drug use was associated with worse cognition, it was not related to developing dementia within their narrow age range. This Scottish study did not identify individuals with MCI, but rather diagnosed dementia using ICD-10 criteria and neuropsychological test scores corrected for age and childhood IQ. Among 7027 participants in The Irish Longitudinal Study on Ageing15 aged 50 years and older and followed for two years, new use of anticholinergics, and to some extent recurrent anticholinergic drug use, was associated with statistically significant reductions in recall scores, but not evidence of decline on MMSE or verbal fluency. In the community-based Australian PATH Through Life cohort,14 among 2058 randomly selected persons 60-64, interviewed twice over four years, exposure to anticholinergic medications was associated with lower level of complex attention in the young-old, but not with greater cognitive decline over time. Papenberg and colleagues investigated the effects of anticholinergics in a population-based sample of older adults without dementia collected from the Swedish National Study on Aging and Care in Kungsholmen.16 They found older adults using anticholinergics both at baseline and follow-up declined more on episodic memory than nonusers. However, the detrimental effect of anticholinergic drugs was confined to episodic memory; there were no negative effects for speed, verbal fluency, semantic, and short-term memory.
In contrast, Grossi and colleagues examined data from the Medical Research Council Cognitive Function and Ageing Study, a population-based, prospective, multi-center cohort study in England and Wales, specifically designed to estimate the prevalence, risk factors and course of dementia.13 They found no evidence of an increased risk of dementia associated with benzodiazepines or anticholinergic medications with ACB score of 1 or 2; however, they did find a statistically significant increase in dementia risk among recurrent users of anticholinergic medications with an ACB score of 3, and particularly those with good baseline cognitive function.
A major strength of our study is its large population-based cohort that we followed for 13 years, minimally affected by selection bias seen in clinical samples. However, population-based studies lack the in-depth clinical assessments and expert diagnoses that are standard in clinical studies. Further, being population-based, we are restricted to the demographic composition of older adults residing in the study area, and they were mostly of European descent. Our results will therefore need to be replicated in more ethnically diverse populations. Next, we relied on participant self-report of medications taken and inspection of bottle labels that varies from what would be found in pharmacy databases, which do not include OTC drugs nor reflect what patients are actually taking. Given the large variety of conditions for which anticholinergic drugs are taken (Table 1), we did not include every possible health condition as a covariate in the models to control for bias (confounding by indication); however, we included total number of prescription medications to adjust for overall morbidity. Further, we used an ACR grading system based on the literature, acknowledging that no method of measuring ACR is entirely satisfactory given that medications have different pharmacodynamics and kinetics, all possible interactions cannot be known, and individual rates of drug metabolism depend on their liver and kidney functions. Finally, we did not take drug dosage or total daily ACR dose into account as these cannot be reliably calculated for drugs taken on an as-needed basis.
With the aging population and increasing burden of Alzheimer’s disease and related dementias, efforts at mitigating cognitive impairment on a population require multi-pronged approaches.36 Promoting healthy life-style behaviors, including minimizing the use of medications with adverse effects on cognition, when possible, will be a part of this approach.37 At the population level, anticholinergic drug is associated with a subgroup of MCI that does not progress to dementia, and could be a potentially modifiable risk factor for MCI.
Supplementary Material
ACKNOWLEDGMENTS.
The authors thank all MYHAT study personnel for their efforts, particularly Ms. Amy Carper for creating the ACR dataset, and all MYHAT participants for their cooperation in this multi-year study.
The work reported here was supported in part by grants R37 AG023651 and R01 AG055389 from the National Institute on Aging, National Institutes of Health, US DHHS
SPONSOR’S ROLE:
The National Institutes of Health had no role in the design, methods, subject recruitment, data collections, or preparation of the paper.
Footnotes
CONFLICT OF INTEREST: Each of the authors reports no conflicts of interest in connection with this report.
DATA AVAILABILITY:
The datasets generated during and/or analyzed during the current study are not publicly available but are available on reasonable request, https://www.dementia-epidemiology.pitt.edu/contact-us/.
REFERENCES
- 1.Ghossein N, Kang M, Lakhkar AD. Anticholinergic medications. In: StatPearls [Internet]. Treasure Island, FL: StatPearls Publishing; 2022:32310353. [PubMed] [Google Scholar]
- 2.Egberts A, Moreno-Gonzalez R, Alan H, et al. Anticholinergic Drug Burden and Delirium: A Systematic Review. J Am Med Dir Assoc. Jan 2021;22(1):65–73 e4. [DOI] [PubMed] [Google Scholar]
- 3.Lopez-Alvarez J, Sevilla-Llewellyn-Jones J, Aguera-Ortiz L. Anticholinergic Drugs in Geriatric Psychopharmacology. Front Neurosci. 2019;13:1309. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Chatterjee S, Talwar A, Aparasu RR. Anticholinergic medications and risk of dementia in older adults: Where are we now? Expert Opin Drug Saf. Oct 2020;19(10):1251–1267. [DOI] [PubMed] [Google Scholar]
- 5.Perry EK, Kilford L, Lees AJ, et al. Increased Alzheimer pathology in Parkinson’s disease related to antimuscarinic drugs. Ann Neurol. Aug 2003;54(2):235–8. [DOI] [PubMed] [Google Scholar]
- 6.Yoshiyama Y, Kojima A, Itoh K, et al. Does Anticholinergic Activity Affect Neuropathology? Implication of Neuroinflammation in Alzheimer’s Disease. Neurodegener Dis. 2015;15(3):140–8. [DOI] [PubMed] [Google Scholar]
- 7.Yoshiyama Y, Kojima A, Itoh K, et al. Anticholinergics boost the pathological process of neurodegeneration with increased inflammation in a tauopathy mouse model. Neurobiol Dis. Jan 2012;45(1):329–36. [DOI] [PubMed] [Google Scholar]
- 8.Chuang YF, Elango P, Gonzalez CE, et al. Midlife anticholinergic drug use, risk of Alzheimer’s disease, and brain atrophy in community-dwelling older adults. Alzheimers Dement (N Y). Sep 2017;3(3):471–479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Weigand AJ, Bondi MW, Thomas KR, et al. Association of anticholinergic medications and AD biomarkers with incidence of MCI among cognitively normal older adults. Neurology. Oct 20 2020;95(16):e2295–e2304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Taylor-Rowan M, Edwards S, Noel-Storr AH, et al. Anticholinergic burden (prognostic factor) for prediction of dementia or cognitive decline in older adults with no known cognitive syndrome. Cochrane Database Syst Rev. May 5 2021;5:CD013540. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ganguli M, Jia Y, Hughes TF, et al. Mild Cognitive Impairment that Does Not Progress to Dementia: A Population-Based Study. J Am Geriatr Soc. Feb 2019;67(2):232–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Mitchell AJ, Shiri-Feshki M. Rate of progression of mild cognitive impairment to dementia--meta-analysis of 41 robust inception cohort studies. Acta Psychiatr Scand. Apr 2009;119(4):252–65. [DOI] [PubMed] [Google Scholar]
- 13.Grossi CM, Richardson K, Fox C, et al. Anticholinergic and benzodiazepine medication use and risk of incident dementia: a UK cohort study. BMC Geriatr. Oct 21 2019;19(1):276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Low LF, Anstey KJ, Sachdev P. Use of medications with anticholinergic properties and cognitive function in a young-old community sample. Int J Geriatr Psychiatry. Jun 2009;24(6):578–84. [DOI] [PubMed] [Google Scholar]
- 15.Moriarty F, Savva GM, Grossi CM, et al. Cognitive decline associated with anticholinergics, benzodiazepines and Z-drugs: Findings from The Irish Longitudinal Study on Ageing (TILDA). Br J Clin Pharmacol. Jul 2021;87(7):2818–2829. [DOI] [PubMed] [Google Scholar]
- 16.Papenberg G, Backman L, Fratiglioni L, et al. Anticholinergic drug use is associated with episodic memory decline in older adults without dementia. Neurobiol Aging. Jul 2017;55:27–32. [DOI] [PubMed] [Google Scholar]
- 17.Whalley LJ, Sharma S, Fox HC, et al. Anticholinergic drugs in late life: adverse effects on cognition but not on progress to dementia. J Alzheimers Dis. 2012;30(2):253–61. [DOI] [PubMed] [Google Scholar]
- 18.Ganguli M, Chang CC, Snitz BE, et al. Prevalence of mild cognitive impairment by multiple classifications: The Monongahela-Youghiogheny Healthy Aging Team (MYHAT) project. Am J Geriatr Psychiatry. Aug 2010;18(8):674–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Morris JC. The Clinical Dementia Rating (CDR): current version and scoring rules. Neurology. Nov 1993;43(11):2412–4. [DOI] [PubMed] [Google Scholar]
- 20.Ganguli M, Snitz B, Vander Bilt J, et al. How much do depressive symptoms affect cognition at the population level? The Monongahela-Youghiogheny Healthy Aging Team (MYHAT) study. International journal of geriatric psychiatry. Nov 2009;24(11):1277–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Folstein MF, Folstein SE, McHugh PR. “Mini-mental state”. A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research. 1975;12(3):189–98. [DOI] [PubMed] [Google Scholar]
- 22.Carnahan RM, Lund BC, Perry PJ, et al. The Anticholinergic Drug Scale as a measure of drug-related anticholinergic burden: associations with serum anticholinergic activity. J Clin Pharmacol. Dec 2006;46(12):1481–6. [DOI] [PubMed] [Google Scholar]
- 23.Rudolph JL, Salow MJ, Angelini MC, et al. The anticholinergic risk scale and anticholinergic adverse effects in older persons. Arch Intern Med. Mar 10 2008;168(5):508–13. [DOI] [PubMed] [Google Scholar]
- 24.Chew ML, Mulsant BH, Pollock BG, et al. Anticholinergic activity of 107 medications commonly used by older adults. J Am Geriatr Soc. 2008;56:1333–1341. [DOI] [PubMed] [Google Scholar]
- 25.Han L, Agostini JV, Allore HG. Cumulative anticholinergic exposure is associated with poor memory and executive function in older men. J Am Geriatr Soc. Dec 2008;56(12):2203–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Sittironnarit G, Ames D, Bush AI, et al. Effects of anticholinergic drugs on cognitive function in older Australians: results from the AIBL study. Dement Geriatr Cogn Disord. 2011;31(3):173–8. [DOI] [PubMed] [Google Scholar]
- 27.Bishara D, Harwood D, Sauer J, et al. Anticholinergic effect on cognition (AEC) of drugs commonly used in older people. Int J Geriatr Psychiatry. Jun 2017;32(6):650–656. [DOI] [PubMed] [Google Scholar]
- 28.Boustani M, Campbell N, Munger S, et al. Impact of anticholinergics on the aging brain: a review and practical application. Aging Health. 2008;4(3):311–320. [Google Scholar]
- 29.Ganguli M, Bilt JV, Lee CW, et al. Cognitive test performance predicts change in functional status at the population level: the MYHAT Project. J Int Neuropsychol Soc. Sep 2010;16(5):761–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.R-Core-Team. A language and environment for statistical computing. Vienna, Austria: 2020. [Google Scholar]
- 31.Larrieu S, Letenneur L, Orgogozo JM, et al. Incidence and outcome of mild cognitive impairment in a population-based prospective cohort. Neurology. Nov 26 2002;59(10):1594–9. [DOI] [PubMed] [Google Scholar]
- 32.Ganguli M, Dodge HH, Shen C, et al. Mild cognitive impairment, amnestic type: an epidemiologic study. Neurology. Jul 13 2004;63(1):115–21. [DOI] [PubMed] [Google Scholar]
- 33.Risacher SL, McDonald BC, Tallman EF, et al. Association Between Anticholinergic Medication Use and Cognition, Brain Metabolism, and Brain Atrophy in Cognitively Normal Older Adults. JAMA Neurol. Jun 1 2016;73(6):721–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Del Pino J, Zeballos G, Anadon MJ, et al. Cadmium-induced cell death of basal forebrain cholinergic neurons mediated by muscarinic M1 receptor blockade, increase in GSK-3beta enzyme, beta-amyloid and tau protein levels. Arch Toxicol. May 2016;90(5):1081–92. [DOI] [PubMed] [Google Scholar]
- 35.Capsoni S, Giannotta S, Stebel M, et al. Ganstigmine and donepezil improve neurodegeneration in AD11 antinerve growth factor transgenic mice. Am J Alzheimers Dis Other Demen. May-Jun 2004;19(3):153–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention, intervention, and care. Lancet. Dec 16 2017;390(10113):2673–2734. [DOI] [PubMed] [Google Scholar]
- 37.By the American Geriatrics Society Beers Criteria Update Expert P. American Geriatrics Society 2019 Updated AGS Beers Criteria(R) for Potentially Inappropriate Medication Use in Older Adults. J Am Geriatr Soc. Apr 2019;67(4):674–694. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The datasets generated during and/or analyzed during the current study are not publicly available but are available on reasonable request, https://www.dementia-epidemiology.pitt.edu/contact-us/.