Risk of Cognitive Decline Associated With Paroxetine Use in Elderly Nursing Home Patients With Depression

Abstract

Objective:

This study evaluated the risk of cognitive decline associated with paroxetine use in elderly nursing home patients with depression.

Methods:

A retrospective cohort study was conducted using the 2007 to 2010 Medicare Part D claims and minimum data set (MDS) data involving new users of paroxetine and other selective serotonin reuptake inhibitors (SSRIs). The primary outcome was MDS Cognition Scale. The repeated-measures mixed model was used to examine the effect of paroxetine on cognition after controlling for other factors.

Results:

The baseline MDS Cognition Scale measures for paroxetine (n = 63) and other SSRI users (n =1018) were 2.02 (±1.85) and 2.50 (±2.39), respectively. The repeated-measures mixed model did not find statistically significant difference in cognition with the use of paroxetine (β = 0.02, 95% CI: −0.16 to 0.21]) when compared to other SSRIs.

Conclusions:

There was no differential effect of paroxetine on cognition when compared to other SSRIs.

Keywords

cognition paroxetine depression elderly

Introduction

Depression is a chronic disorder and one of the most common mental disorders among the elderly population in the long-term care setting.¹ Depression affects quality of life by adversely affecting the physical and cognitive functions.² Depression in the elderly individuals includes major depressive disorder, dysthymic disorder, and minor depressive disorder. Owing to the 5 times higher prevalence of depression in nursing homes compared with the community settings, Omnibus Budget Reconciliation Act of 1987 incorporated regulations to improve care for residents with depression.³

Patients with depression show significant cognitive deficits in executive function, attention, memory, and processing of information.⁴ A meta-analysis by Christensen revealed that patients with depression had lower performance on almost all cognitive tests.⁵ A report based on 13 studies found increased risk of cognitive decline due to depression.⁶ More recently, a study found that elderly patients with depression had 40% higher risk of mild cognitive impairment and more than 2 times higher risk of dementia than patients without depression.⁷

Past literature indicates significant association between anticholinergic drug use and risk of cognitive decline and dementia.^8
-10 A review of anticholinergic medications on cognition found that anticholinergic medications were associated with poor cognitive performance in older adults.¹⁰ A prospective study found that community-dwelling elderly patients taking anticholinergic drugs were at a higher risk for cognitive decline and dementia than the others. An increased risk of 1.4 to 2.0 times was observed in cognitive decline among continuous users but not among those who discontinued the medications.⁹ The use of anticholinergics was also associated with an increased risk of cognitive impairment in a longitudinal study involving African Americans enrolled in Indianapolis–Ibadan Dementia Project.⁸

The long-term care guidelines recommend second-generation antidepressants, selective serotonin reuptake inhibitors (SSRIs), as drugs of choice for older patients.¹¹ Second-generation antidepressants are similar in their efficacy for treating depression.^12,13 Special consideration should be given to adverse events and onset of action while selecting SSRIs. All SSRIs have the same principal mechanism of action, but they differ from each other in their pharmacodynamic and pharmacokinetic properties.¹⁴ Paroxetine has a higher affinity for muscarinic acetylcholine receptors than other SSRIs. Muscarinic receptor-binding properties of paroxetine are similar to desipramine, significantly higher than sertraline but much lower than amitriptyline.¹⁵ Consequently, the likelihood of paroxetine inducing anticholinergic side effects is expected to be higher than that of other SSRIs. A recent study indicated that paroxetine may induce fewer adverse anticholinergic effects than clomipramine but more than fluvoxamine.¹⁶ Per the 2015 American Geriatrics Society Updated Beers Criteria, paroxetine is a strong anticholinergic and is considered potentially inappropriate for elderly patients with dementia and cognitive impairment.¹⁷ The Anticholinergic Drug Scale also classifies paroxetine as an agent with clinically significant anticholinergic properties.¹⁸

Although strongly anticholinergic, paroxetine is used in elderly patients with depression.^19
-21 A review of the effects of newer antidepressants on neurocognitive function concluded that paroxetine is associated with a lower performance on neurocognitive measures such as verbal recall scores and paired-associate learning scores than other SSRIs.²² However, none of the studies have examined the cognitive impact of individual agents, specifically paroxetine. Thus, the objective of the present study was to evaluate the risk of cognitive decline associated with paroxetine use in elderly nursing home patients with depression.

Methods

Data Source

The present study used the 2007 to 2010 minimum data set (MDS) linked Medicare data files to achieve the study objective. The prescription Part D claims file, MDS, and Master Beneficiary Summary File were used for this study. Medicare Part D provides the prescription drugs and prescription drug insurance premiums to Medicare beneficiaries at subsidized cost.²³ The MDS is a standardized, primary screening and assessment tool for clinical assessment of all residents in federally certified nursing home facilities.²⁴

All Medicare-certified nursing homes are required to complete a comprehensive MDS annual assessment on each resident upon admission and when the resident shows significant change in patient health status. A subset of the full MDS assessment is conducted quarterly. The admission assessment in MDS is completed within 14 calendar days of admission to the facility and the annual assessment is completed within 366 days of the admission assessment but not more than 92 days of a quarterly assessment. Quarterly assessments are brief in nature and are captured quarterly or following any adverse events. Further details regarding the MDS can be found elsewhere.^25,26

Study Design and Study Cohort

This study involved a retrospective cohort design. Figure 1 shows the schematic representation of the study cohort. Patients were defined as long-term residents if the difference between last admission and last annual assessment was ≥365 days. The index antidepressant use was defined as having a first prescription of an antidepressant after at least 6 months of no prescription fill date for any of the antidepressant medications. Study cohort consisted of patients who were (1) 65 years and older, (2) diagnosed with depression as per MDS, (3) initiated paroxetine or other SSRIs, after 6 months of washout period, (4) continuously eligible for Medicare Part D in the 6 months baseline and during 1 year of follow-up, (5) noncomatose, and (6) not diagnosed with dementia in the baseline period. This study was approved under the exempt category by the University of Houston Committee for the Protection of Human Subjects.

Figure 1.

Schematic presentation of cohort construction.

Exposures, Outcome, and Covariates Definitions

Exposure to SSRIs was measured using prescription Medicare Part D claims data; SSRIs were classified as paroxetine and other SSRIs. Other SSRIs included sertraline, citalopram, fluoxetine, fluvoxamine, or escitalopram. The National Drug Codes in the Part D file were used to measure SSRI exposure. The maximum follow-up period for the study was 1 year. Study participants were censored if they reached the end of the follow-up period, switched to a different antidepressant, had a gap of more than 15 days in the use of the index antidepressant,^27,28 used psychotherapy, or whichever occurred earlier.

The primary outcome of interest was cognition that was measured as a continuous variable using the MDS Cognition Scale. The MDS Cognition Scale measures cognition on an 11-point scale and ranges from 0 (intact cognition) to 10 (very severe impairment). It assesses short- and long-term memory, orientation, communication, and dressing.²⁹ The MDS Cognition Scale is a valid tool for measuring cognition in nursing home residents.³⁰ Also, it is continuous, intuitive, and easier to compute and better at discriminating severely cognitively impaired patients than the Cognitive Performance Scale.^29,31 Quarterly assessments after initiating antidepressants were used to assess cognition.

The baseline risk factors included demographic and clinical/behavioral characteristics of the residents. Demographic characteristics included age, gender, insurance, and race. Clinical characteristics included cognitive and behavioral measures such as index of social engagement, aggressive behavior scale, and pain scale in addition to the MDS Cognition Scale. Baseline depression and its severity was measured using Minimum Data Set Depression Rating Scale.³² Baseline medication use of other anticholinergics, antipsychotics, antianxiety agents, hypnotics, and diuretics was also used as covariates. All the baseline risk factors were selected on the basis of systematic reviews, expert opinions, and based on their availability from the data source.^6,33

-36

Statistical Analysis

Differences between the 2 treatment groups were assessed using χ² for the dichotomous variables and t test for the continuous variables. Long-term care nursing home residents have their cognition measured repeatedly at every 90-day period between admission and annual assessment. Thus, assessments made on the same resident are correlated with each other. Therefore, the repeated-measures mixed model analysis was used to examine the effect of paroxetine and other SSRIs on cognition. This model accounts for correlation among outcome measurements collected on the same resident, allows for missing data, and uses all available data for the analysis.³⁷ The baseline risk factors, demographic, and clinical/behavioral characteristics were used as covariates. Time was also included as a covariate to account for any temporal trend in cognition scores, given that participants were followed for different amounts of time. The baseline MDS Cognition Scale score was included as a covariate, and the repeated-measures model only included quarterly cognition assessments after antidepressant initiation. Results were presented as β estimates along with 95% confidence intervals. Statistical significance was set at an a priori α level of 0.05. All analyses were conducted using SAS Software Release 9.3. The PROC MIXED in the SAS with REPEATED statement and AR(1) error-covariance structure were used to perform the repeated-measures mixed model analysis to examine the association between antidepressant use and cognition. Distribution of MDS Cognition Scale score was found to be positively skewed. Thus, MDS Cognition Scale score was log transformed to achieve normal distribution and consequently repeated-measures mixed model was used to examine the effect of paroxetine on cognition after controlling for other factors.

Sensitivity Analyses

Additional sensitivity analyses were conducted to evaluate the robustness of the study findings. In the first sensitivity analysis, repeated-measures mixed model analysis along with propensity score adjustment was used to examine the association between cognition and use of paroxetine and other SSRIs. Propensity score was estimated using a logistic regression model with treatment allocation as the dependent variable and baseline confounders and risk factors as the independent variables. The estimated propensity score was used as a covariate in the adjusted analysis. In the second sensitivity analysis, a propensity score-adjusted Poisson regression model was used; the first quarterly assessment of MDS Cognition Scale after the index antidepressant use was the outcome measure and baseline covariates included demographic, clinical/behavioral characteristics and baseline MDS Cognition Scale score.

Results

Figure 2 outlines the process of sample selection and cohort development. There were 1518 nursing home residents who met the inclusion and exclusion criteria and had at least 1 quarterly assessment during the follow-up. Of these, 1081 were new users of SSRIs. Among the SSRI users, 63 (5.83%) received paroxetine and 1018 (94.17%) received other SSRIs. Table 1 presents baseline characteristics of paroxetine and other SSRI users. The baseline MDS Cognition Scale measure (standard deviation, SD) for paroxetine and other SSRI users was 2.02 (±1.85) and 2.50 (±2.39), respectively. The MDS Cognition Scale measure (SD) at the end of the study for paroxetine and other SSRI users was 2.30 (±1.95) and 2.61 (±2.10), respectively. There was no difference in the mean persistence between paroxetine 201.04 (±98.17) and other SSRI group 194.67 (±90.14) at P = .588.

Figure 2.

Flowchart of study sample selection.

Table 1.

Baseline Characteristics of Elderly Nursing Home Patients With Depression Using Paroxetine and Other SSRIs.

	Paroxetine Users (n = 63)	Other SSRI Users (n = 1018)	P Value
Demographic characteristics
Gender, %			.920
Female	30	30
Male	70	70
Age in years, mean (SD)	81.33 (9.41)	80.50 (9.85)	.519
Race, %			.181
White	86	78
Nonwhite	5	14
Missing	9	8
Behavioral characteristics, mean (SD)
Baseline MDS Cognition Scale	2.02 (1.85)	2.50 (2.39)	.115
Index of social engagement	2.51 (1.47)	2.47 (1.50)	.859
Depression rating scale	0.78 (1.26)	0.86 (1.55)	.696
Aggressive behavior scale	0.33 (1.05)	0.29 (0.94)	.729
Pain scale	0.78 (1.85)	0.73 (0.75)	.622
Medical characteristics, %
Arthritis	36	29	.180
Diabetes	33	39	.396
Hypertension	65	75	.087
Cancer	2	6	.161
Stroke	24	23	.823
Congestive heart failure	27	22	.309
Chronic obstructive pulmonary disease	17	20	.658
Parkinson	8	3	.066
Other cardiac disorders	16	23	.199
Schizophrenia	2	4	.362
Anxiety disorder	16	15	.805
Asthma	5	4	.806
Manic depression	2	2	.695
Depression	14	13	.834
Medication use characteristics, %
Antipsychotics	13	11	.734
Antianxiety	13	18	.252
Hypnotics	9	11	.732
Diuretics	40	36	.583
Other anticholinergics	3	4	.600

Abbreviations: MDS, minimum data set; SD, standard deviation; SSRIs, selective serotonin reuptake inhibitors.

Table 2 reports results of repeated-measures mixed model analysis after adjusting for baseline confounders and risk factors. Results from this model suggest that there was no statistically significant difference between paroxetine users (β = 0.02, 95% CI: −0.16 to 0.21]) in terms of MDS Cognition Scale compared to other SSRIs users. Results from the sensitivity analyses supported the main findings. Results from the first sensitivity analysis indicate that there was no significant association between cognition and paroxetine use (β = 0.05, 95% CI: −0.16 to 0.26) when compared with other SSRIs after adjusting for propensity score and time. Appendix A presents the mean MDS-cog score and the sample size at different assessments. Appendix B presents the distribution of propensity scores for paroxetine and other SSRIs. The graph indicates a major region of overlap across the 2 treatment groups, thereby justifying the use of propensity score adjustment in the sensitivity analysis. Results of the second sensitivity analysis indicate that there was no significant association between paroxetine use and first quarterly assessment of MDS Cognition Scale after the index antidepressant use (β = −0.06, 95% CI: −0.23 to 0.12) when compared with other SSRIs after controlling for all the baseline covariates in the study.

Table 2.

Model for the Association Between Cognition and Use of Paroxetine and Other SSRIs in Elderly Nursing Home Residents With Depression.

Variables	Parameter Estimate	95% CI	P Value
Main analysis: repeated-measures mixed model after adjusting for baseline covariates and time^a
Other SSRIs	1.00	Reference	--
Paroxetine	0.02	−0.16 to 0.21	.808
First sensitivity analysis: propensity scores adjusted repeated-measures mixed model^b
Other SSRIs	1.00	Reference	--
Paroxetine	0.05	−0.16 to 0.26	.625
Second sensitivity analysis: propensity scores adjusted Poisson regression model^b
Other SSRIs	1.00	Reference	--
Paroxetine	−0.06	−0.23 to 0.12	.527

Abbreviation: SSRIs, selective serotonin reuptake inhibitors.

^aModel adjusted for demographic characteristics such as age, gender, race; behavioral characteristics such as baseline Minimum Data Set Cognition Scale, index of social engagement, depression rating scale, aggressive behavior scale, pain scale; common chronic conditions such as arthritis, cancer, asthma, chronic obstructive pulmonary disease, Parkinson, diabetes, hypertension, stroke, congestive heart failure, other cardiac disorders, schizophrenia, anxiety disorder, manic depression; and use of medications such as antipsychotics, antianxiety, hypnotics, diuretics, and other anticholinergics.

^bModel adjusted for propensity score and time.

Discussion

Paroxetine has strong anticholinergic and sedative properties, which can lead to negative effects on cognition. Anticholinergic medications such as paroxetine are often considered potentially inappropriate for the elderly patients with dementia and cognitive impairment.¹⁷ The present study examined the cognitive effect of paroxetine in elderly residents with depression when compared to other SSRIs. This study focused on elderly residents with depression as they often present with significant cognitive complaints or deficits and are at significant risk for dementia.^4

-7 To the authors’ knowledge, none of the previous studies have evaluated the comparative safety of paroxetine and other SSRIs with regard to cognition in elderly nursing home patients with depression. This study found that although paroxetine is anticholinergic, there is no differential effect on cognition as measured by MDS Cognition Scale when compared to other SSRIs.

Paroxetine is commonly used in elderly nursing home patients with depression.^19
-21 A 1-year double-blind, randomized, parallel-group, multicenter Italian study compared paroxetine and fluoxetine for cognitive functions in nondemented elderly patients with depression. This study did not find cognitive decline with the use of paroxetine and fluoxetine.³⁸ The current study used real-world data and found that there is no significant difference in cognition across the 2 treatment groups. The study results suggest that pharmacologic differences, especially in terms of anticholinergic effects, between paroxetine and other SSRIs do not translate into significant clinical differences with respect to cognition. It is important to note that patients with a dementia diagnosis at baseline were excluded in the present study. Thus, the study cohorts consisted of patients who were at the lower end of the MDS Cognition Scale. It is possible that individuals with dementia would have shown higher sensitivity to the anticholinergic effects of paroxetine than those included in this study. However, a recent study did not find any higher risk of dementia among elderly nursing home patients with depression using paroxetine when compared to the other SSRIs.³⁹ Also, it is possible that the MDS Cognition Scale may not be sensitive to the anticholinergic effect of paroxetine and other SSRIs on cognition in short periods of time in relatively cognitively normal elderly population in the present study. Use of other measures of cognition might provide better insight into the relationship between anticholinergic effect of paroxetine and other SSRIs on cognition. Proust-Lima evaluated the sensitivity of different test in measuring cognitive changes in the aging population. They found that the Mini-Mental State Examination and the Benton Visual Retention Test showed better sensitivity to cognitive changes in low levels of cognition, while the Digit Symbol Substitution Test showed better sensitivity to changes in higher levels of cognition. The Isaacs Set Test shortened at 15 seconds was found to be sensitive to small cognitive changes in all ranges of cognition.⁴⁰ Thus, future study might examine anticholinergic effect of paroxetine and other SSRIs on cognition in elderly nursing home residents with depression, using the tests mentioned above. Also, patients were censored when they switched from paroxetine to other SSRIs and vice versa. This helped in attributing the treatment effect to index treatment and reducing selection bias, but it might have concealed the inherent dynamic nature of antidepressant treatment. Thus, future studies might look at the effect of dynamic nature of use of antidepressant treatment on cognition in elderly nursing home residents with depression.

The lack of significance of the study findings regarding the relative cognitive effect of paroxetine and other SSRIs does not mean that there is no influence of paroxetine on cognition; rather these findings indicate that paroxetine and other SSRIs share similar cognitive profile in depression with respect to measured clinical effects. Additionally, paroxetine is considered as potentially inappropriate for elderly patients with dementia due to the damage to the cholinergic neurons system, and the central adverse effects of anticholinergic medications can further lead to worsening of the condition.^41

-44 Therefore, patient characteristics such as cognitive functioning should be considered by the physician before prescribing any antidepressant therapy. There is also a need for future studies evaluating the long-term safety of these antidepressants in a geriatric population with poor cognitive functioning.

Strengths and Limitations

The current study used MDS-linked Medicare claims data to examine the comparative cognitive profile of paroxetine and other SSRIs in elderly nursing home residents with depression. This provides several advantages such as population of long-term care residents, long follow-up data, and estimation of treatment effectiveness in real-world settings. Additionally, MDS is a clinically rich database that contains valuable measures including cognition.⁴⁵ This study used a new-user design to address the issue of prevalent user bias. Class-specific analyses helped to control for the indication and selection bias. Use of the repeated-measures mixed model helped to account for correlation among repeated outcome measures, handle missing data, and use of all available data for the analysis.³⁷ This study used 1-year follow-up period. Treatment periods and follow-up intervals frequently used in clinical trials have been of 1 to 3 months, but a longer duration is necessary in order to investigate the long-term effects of antidepressant therapy.

Although the current study revealed important findings, it has some limitations. All the diseases and outcome measures were ascertained using MDS data. These data are submitted by the health-care providers. Incomplete, inaccurate, and erroneous information submitted by the health-care providers and availability of insufficient clinical detail may limit the accuracy of the administrative data.⁴⁶ The MDS Cognition Scale has not been validated to measure or detect drug-induced cognitive impairment/improvement in the elderly population. Exposure to SSRIs was ascertained using Part D claims data. Medications as a component of Part A bundled payments for short postacute nursing home stays were not captured. However, patients using any antidepressant according to MDS records at the baseline cognitive assessment were excluded from the analysis, which enabled a comparison of the cognitive changes with paroxetine compared to other SSRIs among new users, at least among those who continued the SSRI until the next quarterly assessment. Pharmacy claims capture only dispensing data and do not necessarily indicate actual usage by the patient. Dose of paroxetine and other SSRIs was not taken into account. Central anticholinergic effects may be dose dependent and may involve selective muscarinic receptor antagonism in the central nervous system. Also, although use of other anticholinergic drugs was controlled at baseline,¹⁷ drug-related anticholinergic burden was not considered while examining anticholinergic effects of paroxetine and other SSRIs on cognition. Patients with depression often do not fully recover cognitively after a major depressive episode, and neurocognitive symptoms may represent depressive symptomatology, or other factors coexistent with depression, and not necessarily signs of adverse medication effects.^47
-49 There is a possibility of hidden nonobservable bias due to unmeasured confounders which may alter the study findings. However, results of multiple sensitivity analyses supported the main findings. This study was conducted in a geriatric population residing in nursing home settings, thus the results of this study may not be generalizable to other populations.

Conclusions

This retrospective cohort study of elderly nursing home residents with depression revealed that there is no differential effect of paroxetine on cognition when compared to other SSRIs. Although paroxetine is strongly anticholinergic, pharmacologic differences among SSRIs, especially in terms of anticholinergic effects, do not translate into gross clinical differences with respect to cognition. There is also a need for future studies evaluating the long-term safety of these antidepressants in a geriatric population with poor cognitive functioning.

Footnotes

Appendix A

Table A1.

Mean MDS Cognition Scale Score and Sample Size at Different Assessments.

Antidepressant	Assessment	Sample Size	Mean ± SD
Paroxetine (n = 63)	1	63	2.14 ± 2.02
	2	39	2.46 ± 1.92
	3	19	2.16 ± 1.83
Other SSRIs (n = 1018)	1	1018	2.59 ± 2.12
	2	623	2.68 ± 2.14
	3	314	2.58 ± 2.11
	4	20	2.30 ± 1.49

Abbreviations: MDS, minimum data set; SD, standard deviation; SSRIs, selective serotonin reuptake inhibitors.

Appendix B

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by grants from the Agency for Healthcare Research and Quality (AHRQ) (Grant Number: R01HS021264. Principal Investigator: Rajender R. Aparasu). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

References

American Geriatrics Society; American Association for Geriatric Psychiatry. Consensus statement on improving the quality of mental health care in US nursing homes: management of depression and behavioral symptoms associated with dementia. J Am Geriatr Soc. 2003;51(9):1287–1298.

Alexopoulos

Buckwalter

Olin

Martinez

Wainscott

Krishnan

. Comorbidity of late life depression: an opportunity for research on mechanisms and treatment. Biol Psychiatry. 2002;52(6):543–558.

Act OBR. Public Law 100-203. Nursing Home Reform, Part C. 1987.

Wilkins

Mathews

Sheline

. Late life depression with cognitive impairment: evaluation and treatment. Clin Interv Aging. 2009;4:51–57.

Christensen

Griffiths

Mackinnon

Jacomb

. A quantitative review of cognitive deficits in depression and Alzheimer-type dementia. J Int Neuropsychol Soc. 1997;3(6):631–651.

Plassman

Williams

Jr Burke

Holsinger

Benjamin

. Systematic review: factors associated with risk for and possible prevention of cognitive decline in later life. Ann Intern Med. 2010;153(3):182–193.

Richard

Reitz

Honig

. Late-life depression, mild cognitive impairment, and dementia. JAMA Neurol. 2013;70(3):374–382.

Campbell

Boustani

Lane

. Use of anticholinergics and the risk of cognitive impairment in an African American population. Neurology. 2010;75(2):152–159.

Carriere

Fourrier-Reglat

Dartigues

. Drugs with anticholinergic properties, cognitive decline, and dementia in an elderly general population: the 3-city study. Arch Intern Med. 2009;169(14):1317–1324.

10.

Campbell

Boustani

Limbil

. The cognitive impact of anticholinergics: a clinical review. Clin Interv Aging. 2009;4:225–233.

11.

American Medical Directors Association (AMDA). Depression in the Long-term Care Setting. Columbia, MD: 2011.

12.

Gartlehner

Hansen

Morgan

. Comparative benefits and harms of second-generation antidepressants for treating major depressive disorder: an updated meta-analysis. Ann Intern Med. 2011;155(11):772–785.

13.

Gartlehner

Thieda

Hansen

. Comparative risk for harms of second-generation antidepressants: a systematic review and meta-analysis. Drug Saf. 2008;31(10):851–865.

14.

Goodnick

Goldstein

. Selective serotonin reuptake inhibitors in affective disorders—I. Basic pharmacology. J Psychopharmacol. 1998;12(3 suppl B):S5–S20.

15.

Owens

Morgan

Plott

Nemeroff

. Neurotransmitter receptor and transporter binding profile of antidepressants and their metabolites. J Pharmacol Exp Ther. 1997;283(3):1305–1322.

16.

Fujishiro

Imanishi

Onozawa

Tsushima

. Comparison of the anticholinergic effects of the serotonergic antidepressants, paroxetine, fluvoxamine and clomipramine. Eur J Pharmacol. 2002;454(2-3):183–188.

17.

American Geriatrics Society 2015 Beers Criteria Update Expert Panel. American Geriatrics Society 2015 updated Beers Criteria for potentially inappropriate medication use in older adults. J Am Geriatr Soc. 2015;63(11):2227–2246.

18.

Kersten

Molden

Willumsen

Engedal

Bruun Wyller

. Higher anticholinergic drug scale (ADS) scores are associated with peripheral but not cognitive markers of cholinergic blockade. Cross sectional data from 21 Norwegian nursing homes. Br J Clin Pharmacol. 2013;75(3):842–849.

19.

Karkare

Bhattacharjee

Kamble

Aparasu

. Prevalence and predictors of antidepressant prescribing in nursing home residents in the United States. Am J Geriatr Pharmacother. 2011;9(2):109–119.

20.

Chatterjee

Aparasu

Carnahan

Johnson

Chen

. Prevalence of anticholinergic medication use among elderly nursing home residents with depression Value Health. 2014;17(3):A211.

21.

Bali

Aparasu

Johnson

Chen

Carnahan

. Risk of dementia associated with the use of paroxetine among the elderly nursing home patients with depression. Value Health. 2014;17(3):A209.

22.

Biringer

Rongve

Lund

. A review of modern antidepressants effects on neurocognitive function. Curr Psychiatry Rev. 2009;5(3):164–174.

23.

Medicare Standard Analytical Files: Identifiable Data Files. Centers for Medicare & Medicaid Services. http://www.cms.hhs.gov/IdentifiableDataFiles/02_StandardAnalyticalFiles.asp. Accessed June 18, 2015. Updated June 11, 2015.

24.

Hanlon

Donohue

. Medicare Part D data: a valuable tool for pharmacoepidemiology and pharmacoeconomic research. Am J Geriatr Pharmacother. 2010;8(6):483–484.

25.

Morris

Hawes

Fries

. Designing the national resident assessment instrument for nursing homes. Gerontologist. 1990;30(3):293–307.

26.

Morris

Murphy

Nonemaker

Long Term Care Facility Resident Assessment Instrument (RAI) User’s Manual: For Use with Version 2.0 of the Health Care Financing Administration’s Minimum Data Set, Resident Assessment Protocols, and Utilization Guidelines. Health Care Financing Administration; 1995. Baltimore, Maryland.

27.

Cantrell

Eaddy

Shah

Regan

Sokol

. Methods for evaluating patient adherence to antidepressant therapy: a real-world comparison of adherence and economic outcomes. Med Care. 2006;44(4):300–303.

28.

Liu

Chen

Faries

. Adherence and persistence with branded antidepressants and generic SSRIs among managed care patients with major depressive disorder. Clinicoecon Outcomes Res. 2011;3:63–72.

29.

Hartmaier

Sloane

Guess

Koch

. The MDS Cognition Scale: a valid instrument for identifying and staging nursing home residents with dementia using the minimum data set. J Am Geriatr Soc. 1994;42(11):1173–1179.

30.

Cohen-Mansfield

Taylor

McConnell

Horton

. Estimating the cognitive ability of nursing home residents from the minimum data set. Outcomes Manag Nurs Pract. 1999;3(1):43–46.

31.

Gruber-Baldini

Zimmerman

Mortimore

Magaziner

. The validity of the minimum data set in measuring the cognitive impairment of persons admitted to nursing homes. J Am Geriatr Soc. 2000;48(12):1601–1606.

32.

Burrows

Morris

Simon

Hirdes

Phillips

. Development of a minimum data set-based depression rating scale for use in nursing homes. Age Ageing. 2000;29(2):165–172.

33.

Buscemi

Steglitz

Spring

. Factors and predictors of cognitive impairment in the elderly: a synopsis and comment on “systematic review: factors associated with risk for and possible prevention of cognitive decline in later life.” Transl Behav Med. 2012;2(2):126–127.

34.

Daviglus

Plassman

Pirzada

. Risk factors and preventive interventions for Alzheimer disease: state of the science. Arch Neurol. 2011;68(9):1185–1190.

35.

Daviglus

Bell

Berrettini

. NIH state-of-the-science conference statement: preventing Alzheimer’s disease and cognitive decline. NIH Consens State Sci Statements. 2010;27(4):1–30.

36.

Williams

Plassman

Burke

Benjamin

. Preventing Alzheimer’s disease and cognitive decline. Evid Rep Technol Assess. 2010;(193):1–727.

37.

Singer

. Using SAS PROC MIXED to fit multilevel models, hierarchical models, and individual growth models. J Educ Behav Stat. 1998;23(4):323–355.

38.

Cassano

Puca

Scapicchio

Trabucchi

. Paroxetine and fluoxetine effects on mood and cognitive functions in depressed nondemented elderly patients. J Clin Psychiatry. 2002;63(5):396–402.

39.

Bali

Chatterjee

Carnahan

Chen

Johnson

Aparasu

. Risk of dementia among elderly nursing home patients using paroxetine and other selective serotonin reuptake inhibitors. Psychiatr Serv. 2015;66(12):1333–1340.

40.

Proust-Lima

Amieva

Dartigues

Jacqmin-Gadda

. Sensitivity of four psychometric tests to measure cognitive changes in brain aging-population-based studies. Am J Epidemiol. 2007;165(3):344–350.

41.

Tune

. Anticholinergic effects of medication in elderly patients. J Clin Psychiatry. 2001;62(suppl 21):11–14.

42.

Feinberg

. The problems of anticholinergic adverse effects in older patients. Drugs Aging. 1993;3(4):335–348.

43.

Kemper

Steiner

Hicks

Pierce

Iwuagwu

. Anticholinergic medications: use among older adults with memory problems. J Gerontol Nurs. 2007;33(1):21–29; quiz 30-31.

44.

Remillard

. A pharmacoepidemiological evaluation of anticholinergic prescribing patterns in the elderly. Pharmacoepidemiol Drug Saf. 1996;5(3):155–164.

45.

Mor

. A comprehensive clinical assessment tool to inform policy and practice: applications of the minimum data set. Med Care. 2004;42(4 suppl):III50–III59.

46.

Crocco

Castro

Loewenstein

. How late-life depression affects cognition: neural mechanisms. Curr Psychiatry Rep. 2010;12(1):34–38.

47.

Butters

Becker

Nebes

. Changes in cognitive functioning following treatment of late-life depression. Am J Psychiatry. 2000;157(12):1949–1954.

48.

Portella

Marcos

Rami

Navarro

Gasto

Salamero

. Residual cognitive impairment in late-life depression after a 12-month period follow-up. Int J Geriatr Psychiatry. 2003;18(7):571–576.

49.

Nebes

Pollock

Houck

. Persistence of cognitive impairment in geriatric patients following antidepressant treatment: a randomized, double-blind clinical trial with nortriptyline and paroxetine. J Psychiatr Res. 2003;37(2):99–108.