Proton pump inhibitors (PPIs) are a class of drugs prescribed

Transcription

1 Proton Pump Inhibitors and Risk of Mild Cognitive Impairment and Dementia Felicia C. Goldstein, PhD,* Kyle Steenland, PhD, Liping Zhao, MSPH, Whitney Wharton, PhD,* Allan I. Levey, MD, PhD,* and Ihab Hajjar, MD, MS* From the *Department of Neurology, School of Medicine, Emory University; Department of Environmental and Occupational Health, School of Public Health, Emory University; Department of Biostatistics and Bioinformatics, School of Public Health, Emory University; and Department of Medicine, School of Medicine, Emory University, Atlanta, Georgia. Address correspondence to Felicia C. Goldstein, Department of Neurology, Emory University, 12 Executive Park Dr. NE, Atlanta, GA DOI: /jgs OBJECTIVES: To examine the risk associated with the use of proton pump inhibitors (PPIs) of conversion to mild cognitive impairment (MCI), dementia, and specifically Alzheimer s disease (AD). DESIGN: Observational, longitudinal study. SETTING: Tertiary academic Alzheimer s Disease Centers funded by the National Institute on Aging. PARTICIPANTS: Research volunteers aged 50 and older with two to six annual visits; 884 were taking PPIs at every visit, 1,925 took PPIs intermittently, and 7,677 never reported taking PPIs. All had baseline normal cognition or MCI. MEASUREMENTS: Multivariable Cox regression analyses evaluated the association between PPI use and annual conversion of baseline normal cognition to MCI or dementia or annual conversion of baseline MCI to dementia, controlling for demographic characteristics, vascular comorbidities, mood, and use of anticholinergics and histamine-2 receptor antagonists. RESULTS: Continuous (always vs never) PPI use was associated with lower risk of decline in cognitive function (hazard ratio (HR) = 0.78, 95% confidence interval (CI) = , P =.005) and lower risk of conversion to MCI or AD (HR = 0.82, 95% CI = , P =.03). Intermittent use was also associated with lower risk of decline in cognitive function (HR = 0.84, 95% CI = , P =.001) and risk of conversion to MCI or AD (HR = 0.82, 95% CI = , P =.001). This lower risk was found for persons with normal cognition or MCI. CONCLUSION: Proton pump inhibitors were not associated with greater risk of dementia or of AD, in contrast to recent reports. Study limitations include reliance on selfreported PPI use and lack of dispensing data. Prospective studies are needed to confirm these results to guide empirically based clinical treatment recommendations. J Am Geriatr Soc 65: , Key words: proton pump inhibitors; cognitive functioning; mild cognitive impairment; Alzheimer s disease Proton pump inhibitors (PPIs) are a class of drugs prescribed to treat gastrointestinal disorders such as duodenal ulcers and gastroesophageal reflux disease by reducing gastric acid secretion. The safety of PPIs with respect to cognitive function, including the risk of dementia and Alzheimer s disease (AD), has recently been questioned. Two studies reported that PPIs were associated with greater risk of incident dementia and AD in persons aged 75 and older, raising concerns about their widespread use in older adults. 1,2 One investigation 1 included 3,076 persons (23% PPI users) enrolled in the multicenter German Study of Aging, Cognition and Dementia in Primary Care Patients. These community-residing individuals were aged 75 and older and were judged not to have dementia at baseline according to a battery of measures from the Structured Interview for Diagnosis of Dementia of Alzheimer type, Multi-infarct Dementia and Dementia of other Aetiology 3 consisting of the Mini-Mental State Examination, the activity of daily living scale, and the Hachinski- Rosen Scale. These measures were repeated at 18-month intervals. The investigators found a greater risk of dementia (hazard ratio (HR) = 1.38, 95% confidence interval (CI) = , P =.02) and AD (HR = 1.44, 95% CI = , P =.04) in PPI users than in nonusers over a follow-up period of 72 months. A subsequent study by the same investigators 2 used the claims data of the largest health insurance company in Germany. The inpatient and outpatient diagnoses of 73,679 individuals (4% PPI users) aged 75 and older with and without a dementia diagnosis at baseline over a 7-year period were examined. JAGS 65: , , Copyright the Authors Journal compilation 2017, The American Geriatrics Society /17/$15.00

2 1970 GOLDSTEIN ET AL. SEPTEMBER 2017 VOL. 65, NO. 9 JAGS PPI users had a significantly greater risk of incident dementia than nonusers (HR = 1.44, 95% CI = , P <.001). The finding of a 1.4 times greater risk of dementia with PPI use 1 would mean an increase of 10,000 more new cases of dementia each year in persons aged 75 to 84, 4 although a recent case control study 5 on risk factors for dementia, also conducted in Germany, did not observe greater risk associated with PPI use. That study obtained general practitioner medical record information from a database of individuals aged 70 to 90 with (n = 11,956; 44.3% PPI users) or without (n = 11,956; 45.8% PPI users) a diagnosis of dementia over a 5-year period. The use of PPIs was associated with lower risk of developing dementia (HR = 0.93, 95% CI = ). Proton pump inhibitor use has risen in middle-aged and older adults in the United States, as reported in the National Health and Nutrition Examination Survey, in which the prevalence of prescription PPIs increased significantly from 4.9% to 8.3% in persons aged 40 to 64 from 1999 to The authors of the current study therefore believed that it was important to investigate the association between PPI use and risk of mild cognitive impairment (MCI) and dementia. The sample included individuals enrolled in the National Institutes of Health (NIH), National Institute on Aging (NIA) supported Alzheimer s Disease Centers (ADCs), a nationwide consortium of research sites in the United States. Subjects underwent detailed annual neuropsychological evaluations. The current study examined the risk of incident MCI, dementia, and specifically AD associated with PPI use during followup. The second aim was to examine in a subgroup of persons with MCI at baseline whether PPI use conferred a higher risk of dementia and AD conversion in an already vulnerable group than in persons with normal cognitive function at baseline. All analyses controlled for histamine- 2 receptor antagonists (H2RAs) (cimetidine, ranitidine, famotidine, nizatidine) because these are an alternative treatment for gastric acid related disorders. One study 7 found that use of H2RAs was a risk factor for incident cognitive impairment in a community-residing sample of African Americans aged 65 and older. In contrast, PPIs were not associated with greater risk. Because prior studies did not separately control for H2RA medications, it was important to determine whether they accounted for the conflicting findings. METHODS Participants Information available in the National Alzheimer s Coordinating Center (NACC) database 8 from 33 ADCs from September 2005 through September 2015 was used. Written consent was obtained using forms that the institutional review boards at each site approved. Participants included persons with a baseline diagnosis of normal cognition or MCI according to their ADC clinicians. Clinical diagnosis at each center relies on ADC coding guidelines using not only performance on a core battery of neuropsychological tests, 9 but also the Clinical Dementia Rating 10 score, which provides an index of cognitive and functional status using a structured interview with the participant, a separate interview with the study informant, and a behavioral and neurological examination. A diagnosis of normal cognition requires that neuropsychological test scores be within expectation for age and that persons be able to perform instrumental activities of daily living independently. A diagnosis of MCI requires impairment in a single cognitive domain or multiple domains but also evidence of independence in performing activities of daily living. 11 Criteria for inclusion of cognitively normal controls and participants with MCI in the current study were information about PPI medication use at every visit, aged 50 and older at baseline, and a consistent conversion diagnosis over time (e.g., conversion from normal cognition to MCI but not back to normal cognition at the next visit). PPI Medications Participants were classified as PPI users or nonusers at each annual visit based on information on self- and informant-reported PPIs. PPIs included omeprazole (Prilosec) and omeprazole-sodium bicarbonate (Zegerid), esomeprazole (Nexium), lansoprazole (Prevacid), pantoprazole (Protonix), rabeprazole (AcipHex), and dexlansoprazole (Dexilant, Kapidex). Statistical Analyses Analyses included participants taking PPIs at every visit, those taking PPIs intermittently, and those not taking PPIs at any visit. Multivariable Cox regression analyses using PROC PHREG were performed to evaluate the association between PPI use (always or intermittently vs never) and decline in cognitive status to MCI or dementia in the combined population of those who had normal cognition or MCI at baseline. Follow-up time was the time variable in the regression, beginning with the baseline visit. This was defined as time to decline for those who declined and follow-up time for those who did not decline. A similar regression was run in which the outcome was decline in cognitive status specifically to an underlying AD etiology (conversion to MCI with suspected AD etiology or probable AD), in which MCI due to other suspected etiologies and non-alzheimer s dementias were treated as censored. Finally, separate regressions were run for persons with normal cognition at baseline versus those with MCI at baseline to determine whether one diagnostic group was more vulnerable to the adverse effects of PPIs on risk of decline. Proportional hazard assumptions were tested using an interaction term between PPI use and follow-up time. No violation of the proportional hazard function was found for PPI. Two-tailed statistical significance was set at P <.05. All models controlled for potential confounders, including demographic variables (age at baseline, race, sex, education), vascular comorbidities (self-reported hypertension, diabetes mellitus, heart disease, stroke or transient ischemic attack), mood (depression), and anticholinergic medications and H2RAs. Information on vascular comorbidities and mood were taken from the NACC Uniform Data Set Subject Health History form, which enquires about whether a person has had a diagnosis in the past 2 years of these conditions. Information about medications

3 JAGS SEPTEMBER 2017 VOL. 65, NO. 9 PPIS AND RISK FOR MCI AND DEMENTIA 1971 was taken from the NACC Uniform Data Set Subject Medications form, which enquires about use of prescription and commonly prescribed over-the-counter medications within the 2 weeks before the current visit. The form lists the 100 drugs that NACC participants most commonly report. Additional drugs and their associated DrugIDs are also available at https//alz.washington.edu/member/drug CodeLookUp.html. In the current study, medications were classified into drug classes including H2RAs (cimetidine, ranitidine, famotidine, nizatidine) 7 and those with moderate to severe anticholinergic effects. 12 RESULTS Figure 1 shows the number of participants initially reviewed for study inclusion. Of 12,416 potential participants with baseline diagnoses of normal cognition and MCI, 1,930 were excluded for reasons outlined in Figure 1. Of the remaining 10,486 eligible participants, 884 (8.4%) reported always using PPIs, 1,925 (18.4%) reported intermittent use, and 7,677 (73.2%) reported never using PPIs at any annual follow-up. Table 1 shows the demographic and clinical characteristics of the always, intermittent, and never PPI users. A significantly higher percentage of never PPI users had achieved education levels of high school or more than of those with intermittent PPI use. Always and intermittent PPI users were significantly older than never PPI users. In addition, a significantly higher percentage of always and intermittent PPI users than of never PPI users had heart disease, diabetes mellitus, hypertension, stroke or transient ischemic attack, and depression, and a higher percentage of always and intermittent PPI users took anticholinergic medications. All three groups differed from each other in percentage of H2RA medication use (intermittent>always>never) and number of follow-up visits (intermittent>never>always). Risk of Cognitive Decline and AD Whether persons who did not have a dementia diagnosis at baseline were at greater risk of conversion to a diagnosis of cognitive decline (transition from normal cognition to MCI or dementia; transition from MCI to dementia) was first evaluated as a function of PPI use. Always (HR = 0.78, 95% CI = , P =.005) and intermittent (HR = 0.84, 95% CI = , P <.001) use of PPIs were associated with lower risk of decline in cognitive status after adjusting for potential confounders (Table 2). The NACC database contains information about the suspected etiology of MCI or dementia, allowing whether PPI use was associated with risk of probable AD to be examined. PPI use in the always user group (Table 2) was associated with lower risk of conversion to MCI or dementia due to AD (HR = 0.82, 95% CI = , P =.03), as was intermittent use of PPIs (HR = 0.82, 95% CI = , P <.001), after adjusting for potential confounders. Figure 2 shows the adjusted survival curves of cognitive decline to MCI or dementia; always PPI users had the highest survival rate, with intermittent users in the middle and never PPI users in the lowest group. Always and intermittent users had rates of survival that were higher than and comparable with those of never PPI users for the transition to MCI or AD (Figure S1A,B). Subgroup Analyses as a Function of Baseline Cognitive Status Persons with Normal Cognition Persons diagnosed with normal cognition at baseline (n = 7,404: 613 (8.3%) always users, 1,351 (18.2%) intermittent users, 5,440 (73.5%) never users) were examined separately to determine their relative risk of conversion to cognitive impairment (MCI or dementia) with or without Total sample with normal cogni on or MCI at Baseline (n=12,416) Ineligible (n=1,930) Baseline age <50 years (n= 240) Unstable diagnosis (n=1,523)* Transi on to impaired cogni on but not MCI (n=167) * refers to persons who did not have a consistent conversion diagnosis over me. Eligible for analysis (n=10,486) Normal cogni on (n=7,404) MCI (n=3,082) Figure 1. Inclusion and exclusion criteria. [Color figure can be viewed at wileyonlinelibrary.com]

4 1972 GOLDSTEIN ET AL. SEPTEMBER 2017 VOL. 65, NO. 9 JAGS Table 1. Baseline Characteristics of Participants According to Proton Pump Inhibitor (PPI) Use Variable Always PPI Users, n = 884 Intermittent PPI Users, n = 1,925 Never PPI Users, n = 7,677 P-Value Age, mean SD a b a,b <.001 Female, n (%) 522 (59.0) 1,168 (60.7) 4,756 (62.0).18 African American, n (%) 92 (11.1) 249 (13.7) 971 (13.5).14 High school or above, n (%) 833 (94.4) 1,793 (93.4) a 7,318 (95.7) a <.001 Heart disease, n (%) 289 (32.8) a 588 (30.8) b 1,801 (23.6) a,b <.001 Diabetes mellitus, n (%) 131 (14.8) a 285 (14.9) b 795 (10.4) a,b <.001 Hypertension, n (%) 543 (61.7) a 1,144 (59.6) b 3,646 (47.6) a,b <.001 Depression, n (%) 239 (27.1) a 519 (27.0) b 1,586 (20.7) a,b <.001 Stroke, transient ischemic attack, n (%) 84 (9.6) a 190 (9.9) b 520 (6.8) a,b <.001 Histamine-2 receptor antagonist use, n (%) 106 (12.0) a 440 (22.9) a 651 (8.5) a <.001 Anticholinergic medication use, n (%) 212 (24.0) a 491 (25.5) b 1,034 (13.5) a,b <. 001 Number of visits, median (interquartile range) 3.0 ( ) a 5.0 ( ) a 4.0 ( ) a <.001 A common superscript indicates a significant difference between groups, P <.05. Table 2. Conversion of Baseline Normal Cognition to Mild Cognitive Impairment (MCI) or Dementia and Conversion of Baseline MCI to Dementia Hazard Ratio (95% Confidence Interval), P-Value Parameter Cognitive Decline to MCI or Dementia from Any Cause, n = 10,486 Cognitive Decline to MCI or Alzheimer s Disease, n = 10,156 Always vs never proton pump inhibitor users 0.78 ( ), ( ),.026 Intermittent vs never proton pump inhibitor users 0.84 ( ), ( ), <.001 Histamine-2 receptor antagonist blocker user 0.80 ( ), < ( ), <.001 Anticholinergic medication user 1.03 ( ), ( ),.59 Age at baseline 1.04 ( ), < ( ), <.001 Female 0.64 ( ), < ( ), <.001 African American 0.78 ( ), < ( ),.002 High school or above at baseline 0.78 ( ), ( ), <.001 Heart disease at baseline 1.09 ( ), ( ),.44 Diabetes mellitus at baseline 1.18 ( ), ( ),.026 Hypertension at baseline 1.01 ( ), ( ),.34 Depression at baseline 2.66 ( ), < ( ), <.001 Stroke or transient ischemic attack at baseline 1.25 ( ), ( ),.006 AD etiology. Table S1 shows the relative risk of conversion to MCI or dementia of any etiology (left side) and to an AD etiology (right side), controlling for potential confounders. PPI use in the always user group was associated with lower risk of transition to MCI or dementia with any etiology (HR = 0.73, 95% CI = , P =.03) but was not a significant risk factor for an AD etiology (HR = 0.74, CI , P =.08). Intermittent PPI use was not a significant risk factor for decline in cognitive status due to any etiology (HR = 0.87, 95% CI = , P =.07) or AD etiology (HR = 0.86, 95% CI = , P =.13). Persons with MCI Of 3,082 persons with MCI at baseline, 271 (8.8%) were always PPI users, 574 (18.6%) were intermittent users, and 2,237 (72.6%) were never users. The relative risk of conversion to dementia of any etiology (HR = 0.86, 95% CI = , P =.03) and of an AD etiology (HR = 0.83, 95% CI = , P =.01) was significantly lower in intermittent PPI users, whereas the risk of dementia of any etiology (HR = 0.83, 95% CI = , P =.08) or of an AD etiology (HR = 0.97, 95% CI = , P =.78) was not significantly lower in those who always used PPIs (Table S2). Association Between Age and PPI Use The sample was restricted to persons aged 75 and older because prior studies 1,2,5 investigated the risk of cognitive decline in persons in the same age group. The risk of cognitive decline in individuals without dementia but with normal cognition or MCI at baseline (n = 4,562) was significantly lower in intermittent PPI users (HR = 0.85, 95% CI = , P =.009), with a similar trend in always users (HR = 0.81, 95% CI = , P =.049). DISCUSSION The results of this study do not confirm recent reports 1,2 that the use of PPIs is linked to greater risk of dementia and AD. In this dataset, individuals without dementia at

5 JAGS SEPTEMBER 2017 VOL. 65, NO. 9 PPIS AND RISK FOR MCI AND DEMENTIA 1973 S u r v i v a l P r o b a b i l i t y Years to Conversion Always Intermittent Never Figure 2. Adjusted survival curves of decline in baseline cognition to mild cognitive impairment (MCI) or dementia. baseline who were taking PPIs at every follow-up occasion or were intermittently taking PPIs had lower risk of cognitive decline than those were not taking PPIs. In addition, consistent and intermittent use of PPIs was associated with lower risk of cognitive decline with a suspected AD etiology. When the sample was further divided into those with normal cognition or MCI, the same finding of lower risk for cognitive decline was observed for both groups taking PPIs. A strength of this study was the well-phenotyped sample of individuals that relied on diagnoses of cognitive status by a team of experienced clinicians in academic medical centers. ADC clinicians rely on a comprehensive neuropsychological battery and an independent interview with a study partner to obtain information about a person s functional status before making a diagnosis. Two of the previous investigations 1,2 in individuals in primary care relied on administrative databases using the diagnoses of a diverse group of practitioners presumably with varying degrees of experience in diagnosing dementia. Studies performed in the United States indicate that cognitive impairment and dementia are underdiagnosed and underreported in individuals in primary care. 13,14 Another strength was the availability of a broad age range of individuals; prior studies were limited to persons in their mid 70 s. 1,2,5 H2RA use was also controlled for, which was not done in prior studies. 1,2 Other investigations included polypharmacy, defined as taking five or more drugs, as a covariate instead. This may or may not have included H2Ras, which are an older class of drugs used to treat gastrointestinal disorders and are still prescribed. One of these studies 7 found that use of these medications more than doubled the risk of incident cognitive impairment in community-residing African Americans, whereas there was no greater risk with PPIs. Failure of prior studies to control for H2RAs could have produced varying results depending on the mix of such medications in their polypharmacy category. The current study, unlike the previous study, 7 found that H2RAs decreased the risk of MCI and dementia, including that with AD as the suspected etiology. The current study sample was more heterogeneous in age and race, which may have contributed to the differences in findings. Another study 15 more recently found that H2RA use was not a risk factor for dementia or AD in Adult Changes in Thought (ACT) study participants. Strengths included the ability to evaluate exposure over a decade before enrollment in ACT and detailed information using pharmacy records that allowed them to examine dose-response relationships as risk factors. The current study findings replicate the results of a recent study, 5 but they conflict with those of two others 1,2 demonstrating a detrimental effect of PPIs. The conflicting findings from these two studies cannot be attributed to differences in the classes of PPI medications used for analyses, because they were the same in the current study. Information concerning dosage and schedule of PPI use was not available in the NACC database. This lack of dispensing data is also a limitation of these other recent published reports. 1,2,5 Another limitation concerns a reliance on PPI use based on self-report. This may have resulted in misclassification bias due to persons forgetting their medications, especially in those with MCI. This bias could have occurred as well in the finding of one of the previous studies of detrimental associations between PPIs and cognition because they relied on interviews. 1 All studies, including the current one, lacked information on adherence to PPIs. Finally, all available information was used in the analyses; the missed visits were not considered. For example, a participant who was cognitively normal at baseline may have skipped a Year 2 follow-up. If there were records of PPI use at all the available visits, that participant was considered a PPI user at Year 2. If the diagnosis was MCI at Year 3, conversion would have been classified at Year 3 despite the lack of Year 2 follow-up information. This could have led to misclassification of PPI use or time to diagnosis. One of the previous studies 1 used a last observation carried forward approach, which could have led to similar misclassification. Other risk factors for dementia and AD were replicated across all studies, including older age, depression, and diagnoses of diabetes mellitus and stroke. As in all observational pharmacoepidemiological studies, indication bias may be skewing these results to suggest that PPIs are not detrimental to cognition. The current analysis found that PPI users were at higher risk of dementia due to their higher frequency of cardiovascular disease, hypertension, diabetes mellitus, and depression. Hence, it is unlikely that the observation was related to a bias to prescribing PPIs to healthier individuals. It is possible that the PPI users, because of the greater frequency of cardiovascular risk factors, received better health care than non-ppi users. This, in turn, could have reduced their risk of dementia. The current findings do not support that PPIs are associated with greater risk of dementia despite mechanisms proposed as to why they should be. It has been hypothesized that PPIs are associated with higher beta amyloid levels, which are involved in the pathogenesis of AD. Higher beta amyloid levels were observed in an amyloid cell model and in mice after PPI treatment. 16 A human clinical trial demonstrating changes in cognitive functioning and cerebrospinal fluid and neuroimaging biomarkers of AD linked to PPI use would support these findings.

6 1974 GOLDSTEIN ET AL. SEPTEMBER 2017 VOL. 65, NO. 9 JAGS Another potential mechanism involves the role of vitamin B 12, given findings of an association between use of PPIs and vitamin B 12 deficiency. 17 Certain PPIs, including lansoprazole and omeprazole, have been found to cross the blood brain barrier, demonstrating that they directly affect the brain. 18,19 Caution needs to be exercised when speculating about the effect of PPIs on brain functioning until a randomized, prospective clinical trial elucidates the effect of PPIs on cognition. ACKNOWLEDGMENTS Conflict of Interest: The authors have no disclosures. Author Contributions: Dr. Goldstein had full access to all the data in this study and takes responsibility for the integrity of the data and accuracy of the analyses. Goldstein, Steenland, Hajjar: Study concept and design. Goldstein, Steenland: Drafting of manuscript. All authors: Acquisition, analysis, or interpretation of data; critical revision of manuscript for important intellectual content. Steenland, Zhao: Statistical analysis. Financial Disclosure: Funded by the Emory Alzheimer s Disease Research Center (NIH-NIA 5 P50 AG025688). The NACC database is funded by NIA/NIH Grant U01 AG NACC data are contributed by the NIAfunded ADCs: P30 AG (PI Eric Reiman, MD), P30 AG (PI Neil Kowall, MD), P50 AG (PI Scott Small, MD), P50 AG (PI Allan Levey, MD, PhD), P50 AG (PI Todd Golde, MD, PhD), P30 AG (PI Andrew Saykin, PsyD), P50 AG (PI Marilyn Albert, PhD), P50 AG (PI Bradley Hyman, MD, PhD), P50 AG (PI Ronald Petersen, MD, PhD), P50 AG (PI Mary Sano, PhD), P30 AG (PI Steven Ferris, PhD), P30 AG (PI M. Marsel Mesulam, MD), P30 AG (PI Jeffrey Kaye, MD), P30 AG (PI David Bennett, MD), P50 AG (PI Victor Henderson, MD, MS), P30 AG (PI Charles DeCarli, MD), P50 AG (PI Frank LaFerla, PhD), P50 AG (PI Marie-Francoise Chesselet, MD, PhD), P50 AG (PI Douglas Galasko, MD), P50 AG (PI Bruce Miller, MD), P30 AG (PI Russell Swerdlow, MD), P30 AG (PI Linda Van Eldik, PhD), P30 AG (PI John Trojanowski, MD, PhD), P50 AG (PI Oscar Lopez, MD), P50 AG (PI Helena Chui, MD), P30 AG (PI Roger Rosenberg, MD), P50 AG (PI Thomas Montine, MD, PhD), P50 AG (PI Sanjay Asthana, MD, FRCP), P50 AG (PI John Morris, MD), and P50 AG (PI Stephen Strittmatter, MD, PhD). REFERENCES 1. Haenisch B, von Holt K, Wiese B et al. Risk of dementia in elderly patients with the use of proton pump inhibitors. Eur Arch Psychiatry Clin Neurosci 2015;265: Gomm W, Von Holt K, Thome F et al. Association of proton pump inhibitors with risk of dementia: A pharmacoepidemiological claims data analysis. JAMA Neurol 2016;73: Zaudig M, Hiller W. SIDAM Strukturiertes Interview f ur die Diagnose einer Demenz vom Alzheimer Typ, der Multi-Infarkt- (oder vaskul aren) Demenz und Demenzen anderer Atiologie nach DSM-III-R, DSM-IV und ICD-10 (SIDAM-Handbuch). Bern, Switzerland: Huber, Kuller LH. Do proton pump inhibitors increase the risk of dementia? JAMA Neurol 2016;73: Booker A, Jacob LE, Rapp M et al. Risk factors for dementia diagnosis in German primary care practices. Int Psychogeriatr 2016;28: Kantor ED, Rehm CD, Haas JS et al. Trends in prescription drug use among adults in the United States from JAMA 2015;314: Boustani M, Hall KS, Lane KA et al. The association between cognition and histamine-2 receptor antagonists in African Americans. J Am Geriatr Soc 2007;55: Beekly DL, Ramos EM, van Belle G et al. The National Alzheimer s Coordinating Center (NACC) Database. Alzheimer Dis Assoc Disord 2004;18: Weintraub S, Salmon D, Mercaldo N et al. The Alzheimer s Disease Centers Uniform Data Set (UDS): The Neuropsychologic Test Battery. Alzheimer Dis Assoc Disord 2009;23: Morris JC. The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology 1993;43: Winblad B, Palmer K, Kivipelto M et al. Mild cognitive impairmentbeyond controversies, towards a consensus: Report of the International Working Group on Mild Cognitive Impairment. J Intern Med 2004;256: Cai X, Campbell N, Kahn B et al. Long-term anticholinergic use and the aging brain. Alzheimer Dement 2013;9: Bradford A, Kunik ME, Schulz P et al. Missed and delayed diagnosis of dementia in primary care: Prevalence and contributing factors. Alzheimer Dis Assoc Disord 2009;23: McCarten JR, Anderson P, Kuskowski MA et al. Finding dementia in primary care: The results of a clinical demonstration project. J Am Geriatr Soc 2012;60: Gray SL, Walker R, Dublin S et al. Histamine-2 receptor antagonist use and incident dementia in an older cohort. J Am Geriatr Soc 2011;59: Badiola N, Alcalde V, Pujol A et al. The proton-pump inhibitor lansoprazole enhances amyloid beta production. PLoS ONE 2013;8:e Lam JR, Schneider JL, Zhao W et al. Proton pump inhibitor and histamine 2 receptor antagonist use and vitamin B12 deficiency. JAMA 2013;310: Cheng FO, Ho YF, Hung LC et al. Determination and pharmacokinetic profile of omeprazole in rat blood, brain, and bile by microdialysis and high-performance liquid chromatography. J Chromatogr 2002;949: Rojo LE, Alzate-Morales J, Saavedra IN et al. Selective interaction of lansoprazole and astemizole with tau polymers: Potential new clinical use in diagnosis of Alzheimer s disease. J Alzheimer Dis 2010;19: SUPPORTING INFORMATION Additional Supporting Information may be found in the online version of this article: Figure S1. Decline in baseline cognition to mild cognitive impairment or suspected Alzheimer s disease. Figure S2. Adjusted survival curves of decline in baseline cognition to (A) mild cognitive impairment (MCI) or dementia (B) (MCI) or suspected Alzheimer s disease. Table S1. Conversion of Baseline Normal Cognition to Mild Cognitive Impairment (MCI) or Dementia Table S2. Conversion of Baseline Mild Cognitive Impairment (MCI) to Dementia Please note: Wiley-Blackwell is not responsible for the content, accuracy, errors, or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.