Pharmacy data in epidemiological studies: an easy to obtain and reliable tool

Antidepressant Exposure and DNA Methylation: Insights from a Methylome-Wide Association Study

Importance

Understanding antidepressant mechanisms could help design more effective and tolerated treatments.

Objective

Identify DNA methylation (DNAm) changes associated with antidepressant exposure.

Design

Case-control methylome-wide association studies (MWAS) of antidepressant exposure were performed from blood samples collected between 2006-2011 in Generation Scotland (GS). The summary statistics were tested for enrichment in specific tissues, gene ontologies and an independent MWAS in the Netherlands Study of Depression and Anxiety (NESDA). A methylation profile score (MPS) was derived and tested for its association with antidepressant exposure in eight independent cohorts, alongside prospective data from GS.

Setting

Cohorts; GS, NESDA, FTC, SHIP-Trend, FOR2107, LBC1936, MARS-UniDep, ALSPAC, E-Risk, and NTR.

Participants

Participants with DNAm data and self-report/prescription derived antidepressant exposure.

Main Outcomes and Measures

Whole-blood DNAm levels were assayed by the EPIC/450K Illumina array (9 studies, N exposed = 661, N unexposed = 9,575) alongside MBD-Seq in NESDA (N exposed = 398, N unexposed = 414). Antidepressant exposure was measured by self- report and/or antidepressant prescriptions.

Results

The self-report MWAS (N = 16,536, N exposed = 1,508, mean age = 48, 59% female) and the prescription-derived MWAS (N = 7,951, N exposed = 861, mean age = 47, 59% female), found hypermethylation at seven and four DNAm sites (p < 9.42x10 -8 ), respectively. The top locus was cg26277237 ( KANK1, p self-report = 9.3x10 -13 , p prescription = 6.1x10 -3 ). The self-report MWAS found a differentially methylated region, mapping to DGUOK-AS1 ( p adj = 5.0x10 -3 ) alongside significant enrichment for genes expressed in the amygdala, the "synaptic vesicle membrane" gene ontology and the top 1% of CpGs from the NESDA MWAS (OR = 1.39, p < 0.042). The MPS was associated with antidepressant exposure in meta-analysed data from external cohorts (N studies = 9, N = 10,236, N exposed = 661, f3 = 0.196, p < 1x10 -4 ).

Conclusions and Relevance

Antidepressant exposure is associated with changes in DNAm across different cohorts. Further investigation into these changes could inform on new targets for antidepressant treatments.

3 Key Points

Question: Is antidepressant exposure associated with differential whole blood DNA methylation?Findings: In this methylome-wide association study of 16,536 adults across Scotland, antidepressant exposure was significantly associated with hypermethylation at CpGs mapping to KANK1 and DGUOK-AS1. A methylation profile score trained on this sample was significantly associated with antidepressant exposure (pooled f3 [95%CI]=0.196 [0.105, 0.288], p < 1x10 -4 ) in a meta-analysis of external datasets. Meaning: Antidepressant exposure is associated with hypermethylation at KANK1 and DGUOK-AS1 , which have roles in mitochondrial metabolism and neurite outgrowth. If replicated in future studies, targeting these genes could inform the design of more effective and better tolerated treatments for depression.

Agreement in the CARTaGENE cohort between self-reported medication use and claim data

OBJECTIVES

To describe the agreement of self-reported medication use with claim prescription records and to ascertain factors associated with agreement between the two data sources.

METHODS

Baseline data on self-reported medication use was extracted from CARTaGENE, a cohort study in Quebec, Canada, and from the provincial health insurance records (dispensation database) of the same individuals. Kappa statistics were used to estimate concordance beyond chance between the two data sources. Logistic regression models were adjusted to estimate the association between agreement and selected individual's characteristics (sex, age, education, region, income, utilization of health care system, and comorbidities).

RESULTS

Agreement between self-reported medication use and administrative data varied considerably across medication classes (kappa 0.54 for respiratory system and 0.91 for systemic hormonal preparations). Overall, agreement improved when a fixed time window of 90 days was used for exposure measurement. Sex, education level, frequency of health care use and the number of reported medications were associated with agreement.

DISCUSSION

Overall, there was a reasonable agreement between the two data sources, but important variations were found for the different drug classes. These results could be used by researchers to more accurately assess drug exposures using real-world data, which are increasingly important to regulators.

Effects of childhood adversity and cortisol levels on suicidal ideation and behaviour: Results from a general population study

Childhood trauma is known to increase the risk of suicidal ideation and behaviours, and has also been linked to hypothalamic pituitary adrenal (HPA) axis dysregulation measured in cortisol levels. Recent evidence has shown that adverse childhood experiences are associated with lower cortisol reactivity to stress and diminished cortisol levels upon awakening in individuals vulnerable to suicide. The present study aimed to investigate whether less traumatic long term difficulties during childhood produced a similar effect on suicidal ideation/behaviour and cortisol levels in a general population sample. Participants (N = 1094; mean age 53 years, 53.7% female) from a large cohort study completed retrospective measures of long-term difficulties during childhood and adolescent years and a measure of history of suicidal thoughts, plans and actions together with a measure of current psychological distress. 24-hour urinary free cortisol samples were collected over two days. The results showed that experiencing childhood long-term difficulties were associated with 21% higher odds of reporting suicidal thoughts or plans in adulthood. Early childhood and adolescent difficulties were equally important predictors of suicide thoughts and plans. However, childhood difficulties were not found to be associated with adult urinary free cortisol, nor were adulthood suicidal thoughts, plans and behaviour associated with adult urinary free cortisol levels. Future research should explore the extent to which childhood difficulties and stressors are related to other indicators of HPA axis functioning. The current findings have implications for clinicians and for the development of future suicide prevention interventions.

Factors Influencing Self-Reported Medication Use in the Survey of Health Aging and Retirement in Europe (SHARE) Dataset

The validity of self-reported medication use in epidemiological studies is an important issue in healthcare research. Here we investigated factors influencing self-reported medication use for multiple diagnoses in the seventh wave of the Survey of Health Aging and Retirement in Europe (SHARE) dataset in n = 77,261 participants (ages: mean = 68.47, standard deviation = 10.03 years). The influence of mental, physical, and sociodemographic parameters on medication self-report was analyzed with logistic regressions and mediation models. Depression, memory function, and polypharmacy influenced the self-report of medication use in distinct disorders to varying degrees. In addition, sociodemographic factors, knowledge about diagnosis, the presence of several chronic illnesses, and restrictions of daily instrumental activities explained the largest proportion of variance. In the mediation model, polypharmacy had an indirect effect via depression and memory on self-reported medication use. Factors influencing medication self-report vary between different diagnoses, highlighting the complexity of medication knowledge. Therefore, it is essential to assess the individual parameters and their effect on medication behavior. Relying solely on medication self-reports is insufficient, as there is no way to gage their reliability. Thus, self-reported medication intake should be used with caution to indicate the actual medication knowledge and use.

Self-reported medication intake vs information from other data sources such as pharmacy records or medical records: Identification and description of existing publications, and comparison of agreement results for publications focusing on patients with cancer - a systematic review

PURPOSE

To identify and describe publications addressing the agreement between self-reported medication and other data sources among adults and, in a subgroup of studies dealing with cancer patients, seek to identify parameters which are associated with agreement.

METHODS

A systematic review including a systematic search within five biomedical databases up to February 28, 2019 was conducted as per the PRISMA Statement. Studies and agreement results were described. For a subgroup of studies dealing with cancer, we searched for associations between agreement and patients' characteristics, study design, comparison data source, and self-report modality.

RESULTS

The literature search retrieved 3392 publications. Included articles (n = 120) show heterogeneous agreement. Eighteen publications focused on cancer populations, with relatively good agreement identified in those which analyzed hormone therapy, estrogen, and chemotherapy (n = 11). Agreement was especially good for chemotherapy (proportion correct ≥93.6%, kappa ≥0.88). No distinct associations between agreement and age, education or marital status were identified in the results. There was little evaluation of associations between agreement and study design, self-report modality and comparison data source, thus not allowing for any conclusions to be drawn.

CONCLUSION

An overview of the evidence available from validation studies with a description of several characteristics is provided. Studies with experimental design which evaluate factors that might affect agreement between self-report and other data sources are lacking.

Validity Assessment of Self-reported Medication Use for Hypertension, Diabetes, and Dyslipidemia in a Pharmacoepidemiologic Study by Comparison With Health Insurance Claims

Background

Although self-reported questionnaires are widely employed in epidemiologic studies, their validity has not been sufficiently assessed. The aim of this study was to evaluate the validity of a self-reported questionnaire on medication use by comparison with health insurance claims and to identify individual determinants of discordance in the Tsuruoka Metabolomics Cohort Study.

Methods

Participants were 2,472 community-dwellers aged 37 to 78 years from the Tsuruoka Metabolomics Cohort Study. Information on lifestyle and medications was collected through a questionnaire. Sensitivity and specificity were determined using health insurance claims from November 2014 to March 2016, which were used as a standard. Potential determinants of discordance were assessed using multivariable logistic regression.

Results

The self-reported questionnaire on medication use showed high validity. Sensitivity and specificity were 0.95 (95% CI, 0.93–0.96) and 0.97 (95% CI, 0.96–0.98) for antihypertensive medications, 0.94 (95% CI, 0.91–0.97) and 0.98 (95% CI, 0.98–0.99) for diabetes medications, and 0.84 (95% CI, 0.82–0.87) and 0.98 (95% CI, 0.97–0.99) for dyslipidemia medications, respectively. Males without high education and those who currently smoke cigarettes were found to be associated with discordant reporting which affected sensitivity, especially those with medication use for dyslipidemia.

Conclusions

In this population-based cohort study, we found that the self-reported questionnaire on medication use was a valid measure to capture regular medication users. Sensitivity for dyslipidemia medications was lower than those for the other medications. Type of medication, sex, education years, and smoking status influenced discordance, which affected sensitivity in self-reporting.

New prescription of antihyperglycemic agents among patients with diabetes in Germany: Moderate concordance between health insurance data and self-reports

PURPOSE

To analyze the concordance of new prescription of antihyperglycemic agents between two data sources: patients' self-reports and statutory health insurance (SHI) data among patients with diabetes.

METHODS

Within a cross-sectional study, 494 patients with diabetes were interviewed if and which new prescriptions of diabetes medication they received within the last 3 or 6 months. SHI data for 12 months were linked to cover these periods. For the agreement measurement, SHI data was set as reference, and kappa, positive predictive value (PPV), and sensitivity were calculated for single Anatomical Therapeutic Chemical (ATC) codes and cumulated code groups.

RESULTS

The number of new prescriptions within 3 or 6 months was low, with 5.5% (n = 27) for Metformin/self-report being the highest. Contingency tables were unbalanced and showed large numbers in the no/no-cells. Regarding non-agreement, we found new prescriptions slightly more often in SHI data only than in self-reports only, with insulin and metformin representing an exception. Agreement results were moderate with large confidence intervals (CI). The values for cumulated "all drugs in diabetes" were: kappa = 0.58 (95% CI: 0.51-0.65), PPV = 62.0 (53.4-70.2), sensitivity = 55.6 (47.3-63.6).

CONCLUSIONS

Patients reported a low number of new prescriptions within the last 3 or 6 months. In general we found moderate agreement and in case of non-agreement that self-report no/SHI yes was slightly more frequent than vice versa. These results were based on small case numbers, but could nevertheless be considered when collecting self-reported information on the prescription of antihyperglycemic agents.

Prevalence and Accuracy of Information on CYP2D6, CYP2C19, and CYP2C9 Related Substrate and Inhibitor Co-Prescriptions in the General Population: A Cross-Sectional Descriptive Study as Part of the PharmLines Initiative

BACKGROUND

Drug-drug interaction (DDI) is one of the main contributors to adverse drug reactions and therefore, it is important to study its frequency in the population. We aimed to investigate frequency and concordance on CYP2D6, CYP2C19, and CYP2C9 (CYP2D6/2C19/2C9)-mediated potential DDIs at the Lifelines cohort and linked data from the pharmacy database IADB.nl.

METHODS

As part of the University of Groningen PharmLines Initiative, data were collected on CYP2D6/2C19/2C9-related substrate/inhibitors from entry questionnaires of Lifelines participants and linked information from the pharmacy database IADB.nl. CYP2D6/2C19/2C9 related co-prescriptions were divided based on the type of drugs i.e. chronically used medication (CM) or occasionally used medication (OM). This resulted in the combination of two chronically used drugs (CM-CM), chronically and occasionally used medication (CM-OM), and two occasionally used drugs (OM-OM). To measure the agreement level, cohen's kappa statistics and test characteristics were used. Results were stratified by time window, gender, and age.

RESULTS

Among 80,837 medicine users in the Lifelines, about 1-2 per hundred participants were exposed to a CYP2D6/2C19/2C9-mediated potential DDI. Overall, the overlapping time window of three months produced the highest mean kappa values between the databases i.e. 0.545 (95% CI:0.544-0.545), 0.512 (95% CI:0.511-0.512), and 0.374 (95% CI:0.373-0.375), respectively. CM-CM had a better level of agreement (good) than CM-OM (fair to moderate) and OM-OM combination (poor to moderate). The influence of gender on concordance values was different for different CYPs. Among older persons, agreement levels were higher than for the younger population.

CONCLUSIONS

CYP2D6/2C19/2C9-mediated potential DDIs were frequent and concordance of data varied by time window, type of combination, sex and age. Subsequent studies should rather use a combination of self-reported and pharmacy database information.

Influence of age on real-life effects of doxycycline for acute exacerbations among COPD outpatients: a population-based cohort study

INTRODUCTION

Although bacteria contribute significantly to acute exacerbations of chronic obstructive pulmonary disease (AECOPD), the added value of antibiotics remains controversial, especially in outpatient settings. Age may affect antibiotic effectiveness, but real-world evidence is lacking. We aimed to assess the influence of age on the effectiveness of doxycycline for AECOPD.

METHODS

A retrospective cohort study among outpatients with the first recorded AECOPD treated with oral corticosteroids was conducted using a large pharmacy dispensing database. The primary outcome was treatment failure within 15-31 days after treatment start. Secondary outcome was time to second exacerbation. All analyses were stratified by age groups.

RESULTS

We identified 6300 outpatients with the first AECOPD. 2261 (36%) received doxycycline and 4039 (64%) did not receive any antibiotic (reference group). Overall, there was no difference in treatment failure (adjusted OR: 0.97, 95% CI: 0.84 to 1.12) between two groups. Similarly, no difference in treatment failure was observed in younger groups. However, in patients with advanced age (≥75 years), treatment failure was significantly reduced by doxycycline compared with reference (16% vs 20%, adjusted OR: 0.77, 95% CI: 0.62 to 0.97). Overall, median time to second exacerbation was 169 days (95% CI: 158 to 182 days) in doxycycline group compared with 180 days (95% CI: 169 to 191 days) in reference group (adjusted HR: 1.06, 95% CI: 0.99 to 1.12). Although in older patients there was a trend within 3 months towards longer time of next exacerbation by doxycycline, it did not achieve statistical significance.

CONCLUSIONS

Our findings showed short-term treatment benefit of doxycycline added to oral corticosteroids for chronic obstructive pulmonary disease patients with advanced age. This value remains unclear for persons aged under 75 years in current primary care. Long-term preventive benefits of doxycycline for the next exacerbation were not observed, irrespective of age.

A systematic review of methods for determining cross-sectional active medications using pharmacy databases

PURPOSE

Pharmacy dispensing databases are often used to identify patients' medications at a particular time point, for example to measure prescribing quality or the impact of medication use on clinical outcomes. We performed a systematic review of studies that examined methods to assess medications in use at a specific point in time.

METHODS

Comprehensive literature search to identify studies that compared active medications identified using pharmacy databases to medications identified using nonautomated data sources. Two investigators independently reviewed abstracts and full-text material.

RESULTS

Of 496 studies screened, 29 studies evaluating 50 comparisons met inclusion criteria. Twenty-nine comparisons evaluated fixed look-back period approaches, defining active medications as those filled in a specified period prior to the index date (range 84-730 days). Fourteen comparisons evaluated medication-on-hand approaches, defining active medications as those for which the most recent fill provided sufficient supply to last through the study index date. Sensitivity ranged from 48% to 93% for fixed look-back period approaches and 35% to 97% for medication-on-hand approaches. Interpretation of comparative performance of methods was limited by use of different reference sources, target medication classes, and databases across studies. In four studies with head-to-head comparisons of these methods, sensitivity of the medication-on-hand approach was a median of 7% lower than the corresponding fixed look-back approach.

CONCLUSIONS

The reported accuracy of methods for identifying active medications using pharmacy databases differs greatly across studies. More direct comparisons of common approaches are needed to establish the accuracy of methods within and across populations, medication classes, and databases.

Secondary adherence to non-vitamin-K antagonist oral anticoagulants in patients with atrial fibrillation in Sweden and the Netherlands

OBJECTIVE

There is limited evidence on patients' adherence and the impact of the prescribed dosing regimen in non-vitamin-K oral anticoagulants (NOACs). We aimed to assess secondary adherence to NOACs and to determine the impact of the dosing regimen in patients with atrial fibrillation.

METHODS

Patients using a NOAC between 2009 and 2013 were identified from the nation-wide Swedish Prescribed Drug Register and the Dutch regional IADB.nl database. Patients using a consistent dosage for at least 180 consecutive days were included. Adherence was calculated using the medication possession ratio (MPR) and adjusted for overlapping dates. Adherence was defined as a MPR ≥0.8. Sensitivity analyses were performed using a MPR ≥0.9. Logistic regression was performed to compare secondary adherence and to explore the influence of the dosing regimen.

RESULTS

A total of 5254 Swedish and 430 Dutch NOAC users were included. The mean MPR was 96.0% (SD 7.8%) in Sweden and 95.1% (SD 10.1%) in the Netherlands. Multivariable logistic regression analysis showed that a twice daily regimen had a lower likelihood of being secondary adherent compared to a once daily regimen in Sweden (odds ratio [OR] 0.21 [95% CI 0.12-0.35]).

LIMITATIONS

The influence of selection bias introduced by the inclusion criterion of ≥2 dispensations covering at least 180 days could not be excluded.

CONCLUSIONS

This study demonstrated that secondary adherence was high in this specific setting among patients with at least two initial dispensations of a NOAC covering a minimum of 180 days. The use of NOACs in a once daily regimen showed higher adherence compared to a twice daily regimen.

Exposure measurement error when assessing current glucocorticoid use using UK primary care electronic prescription data

Purpose

To quantify misclassification in glucocorticoid (GC) exposure defined using UK primary care prescription data.

Methods

A cross‐sectional study including patients with rheumatoid arthritis prescribed oral GCs in the past 2 years. Glucocorticoid exposure based on electronic prescription records was compared with participant‐reported GC use captured using a paper diary. Prescription data (containing information about prescriptions issued but no dispensing information) was provided by the Clinical Practice Research Datalink. The following variables were defined: current use and dose of oral GCs and if (and when) participants had received a GC injection. For oral GCs, self‐reported use was taken to represent “true” exposure. A dataset representing a hypothetical population was generated to assess the impact of the misclassification found for current use.

Results

A total of 67 of 78 study participants (86%) were correctly classified as currently on/off oral GCs; 32/38 (84.2%) participants reporting current GC use and 35/40 (87.5%) participants not reporting current use were correctly classified. Estimated values of current dose were imprecise (correlation coefficient 0.46). Concordance between reported and prescribed GC injections was poor (kappa statistic 0.14). Misclassification bias was demonstrated in the hypothetical population: For “true” relative risks of 1.5, 4, and 9, the “observed” relative risks were 1.33, 2.48, and 3.58, respectively.

Conclusions

Misclassification of current use of oral GCs was low but sufficient to lead to significant bias. Researchers should take care to assess the likely impact of exposure misclassification on their analyses.

Concordance assessment of self-reported medication use in the Netherlands three-generation Lifelines Cohort study with the pharmacy database iaDB.nl: The PharmLines initiative

Background

While self-reported data are commonly used as a source of medication use for pharmaco-epidemiological studies, such information is prone to forms of bias. Several previous studies showed that various factors like age, type of drug and data collection method may influence accuracy. We aimed to assess the concordance of the self-reported medication use that was documented at entry to the Lifelines Cohort Study, a three-generation follow-up study in the Netherlands that started in 2006 and included over 167,000 participants.

Materials and methods

As part of the PharmLines Initiative, we collected medication data from the Lifelines participants encoded according to the Anatomical Therapeutic Chemical (ATC) coding scheme and linked the data via Statistics Netherlands to the widely used and representative pharmacy prescription database of the University of Groningen, IADB.nl. Analyses were conducted at second level of ATC coding for all recorded medications as well as a top list of most used medications at drug-specific fifth level. Cohen’s kappa statistics were used to measure the concordance for all participants according to sex and age.

Results

The level of concordance between the two data sources largely differed according to the therapeutic class. Medication used for the cardiovascular system and diabetes, thyroid therapy, bisphosphonates and anti-thrombotic drugs showed a very good agreement (κ>0.75). Medication as needed or prone to stigmatization bias showed a moderate agreement (κ=0.41–0.60), whereas medications used for short periods of time showed a fair agreement (κ=0.0–0.4). Concordance was similar for males and females, but younger adults tended to have lower concordance rates than older adults.

Conclusion

The self-reported method was valid for capturing prevalent chronic medication use at one moment in time, but invalid for medication used for short periods of time. There is no effect of sex on the agreement, and more studies are needed on the influence of age. Future pharmaco-epidemiological studies should preferably combine the two data sources to achieve the highest accuracy of drug exposure rates.

Self-reported medication use validated through record linkage to national prescribing data

OBJECTIVES

Researchers need to be confident about the reliability of epidemiologic studies that quantify medication use through self-report. Some evidence suggests that psychiatric medications are systemically under-reported. Modern record linkage enables validation of self-report with national prescribing data as gold standard. Here, we investigated the validity of medication self-report for multiple medication types.

STUDY DESIGN AND SETTING

Participants in the Generation Scotland population-based cohort (N = 10,244) recruited 2009-2011 self-reported regular usage of several commonly prescribed medication classes. This was matched against Scottish NHS prescriptions data using 3- and 6-month fixed time windows. Potential predictors of discordant self-report, including general intelligence and psychological distress, were studied via multivariable logistic regression.

RESULTS

Antidepressants self-report showed very good agreement (κ = 0.85, [95% confidence interval (CI) 0.84-0.87]), comparable to antihypertensives (κ = 0.90 [CI 0.89-0.91]). Self-report of mood stabilizers showed moderate-poor agreement (κ = 0.42 [CI 0.33-0.50]). Relevant past medical history was the strongest predictor of self-report sensitivity, whereas general intelligence was not predictive.

CONCLUSION

In this large population-based study, we found self-report validity varied among medication classes, with no simple relationship between psychiatric medication and under-reporting. History of indicated illness predicted more accurate self-report, for both psychiatric and nonpsychiatric medications. Although other patient-level factors influenced self-report for some medications, none predicted greater accuracy across all medications studied.

Hypnotics use in children 0-18 months: moderate agreement between mother-reported survey data and prescription registry data

BACKGROUND

Different methods in pharmacoepidemiology can be used to study hypnotic use in children. But neither questionnaire-based data nor prescription records can be considered a "gold standard". This study aimed to investigate the agreement between mother-reported questionnaire-based data and prescription record data for hypnotic drugs in children aged 0-18 months. The agreement was compared to the agreement for a group of antiepileptic drugs.

METHODS

Prescription record data were collected from the Norwegian prescription database for 47,413 children also surveyed in the Norwegian mother and child cohort between 2005 and 2009. Agreement between in the two data sources was calculated using Cohens Kappa. Multinomial logistic regression was used to calculate the effect of sociodemographic variables on discrepancies in data sources.

RESULTS

The agreement between mother-reported and dispensed hypnotics was less than 50% for all hypnotics. Sensitivity of reporting increased with number of filled prescriptions. The agreement of antiepileptic drugs was 92.9% in the same population. Of several sociodemographic factors only paternal educational level and maternal work situation was significantly related to agreement between prescription record and survey data.

CONCLUSION

There was a moderate agreement between reported use and dispensed hypnotic drugs for infants and toddlers. Results indicate that sociodemographic factors play only a minor role in explaining discrepancy.

The concordance between self-reported medication use and pharmacy records in pregnant women

PURPOSE

Several studies have been conducted to assess determinants affecting the performance or accuracy of self-reports. These studies are often not focused on pregnant women, or medical records were used as a data source where it is unclear if medications have been dispensed. Therefore, our objective was to evaluate the concordance between self-reported medication data and pharmacy records among pregnant women and its determinants.

METHODS

We conducted a population-based cohort study within the Generation R study, in 2637 pregnant women. The concordance between self-reported medication data and pharmacy records was calculated for different therapeutic classes using Yule's Y. We evaluated a number of variables as determinant of discordance between both sources through univariate and multivariate logistic regression analysis.

RESULTS

The concordance between self-reports and pharmacy records was moderate to good for medications used for chronic conditions, such as selective serotonin reuptake inhibitors or anti-asthmatic medications (0.88 and 0.68, respectively). Medications that are used occasionally, such as antibiotics, had a lower concordance (0.51). Women with a Turkish or other non-Western background were more likely to demonstrate discordance between pharmacy records and self-reported data compared with women with a Dutch background (Turkish: odds ratio, 1.63; 95% confidence interval, 1.16-2.29; other non-Western: odds ratio, 1.33; 95% confidence interval, 1.03-1.71).

CONCLUSIONS

Further research is needed to assess how the cultural or ethnic differences may affect the concordance or discordance between both medication sources. The results of this study showed that the use of multiple sources is needed to have a good estimation of the medication use during pregnancy.

The burden and management of cytochrome P450 2D6 (CYP2D6)-mediated drug-drug interaction (DDI): co-medication of metoprolol and paroxetine or fluoxetine in the elderly

PURPOSE

Metoprolol and paroxetine/fluoxetine are inevitably co-prescribed because cardiovascular disorders and depression often coexist in the elderly. This leads to CYP2D6-mediated drug-drug interactions (DDI). Because systematic evaluations are lacking, we assessed the burden of metoprolol-paroxetine/fluoxetine interaction in the elderly and how these interactions are managed in Dutch community pharmacies.

METHOD

Dispensing data were collected from the University of Groningen pharmacy database (IADB.nl, 1999-2014) for elderly patients (≥60 years) starting beta-blockers and/or antidepressants. Based on the two main DDI alert systems (G-Standard and Pharmabase), incidences were divided between signalled (metoprolol-fluoxetine/paroxetine) and not-signalled (metoprolol-alternative antidepressants and alternative beta-blockers-paroxetine/fluoxetine) combinations. Incident users were defined as patients starting at least one signalled or a non-signalled combination. G-Standard signalled throughout the study period, whereas Pharmabase stopped after 2005.

RESULTS

A total of 1763 patients had 2039 metoprolol-paroxetine/fluoxetine co-prescriptions, despite DDI alert systems, and about 57.3% were signalled. The number of metoprolol-alternative antidepressant combinations (incidences = 3150) was higher than alternative beta-blocker-paroxetine/fluoxetine combinations (incidences = 1872). Metoprolol users are more likely to be co-medicated with an alternative antidepressant (incidences = 2320) than paroxetine/fluoxetine users (incidences = 1232) are. The number of paroxetine/fluoxetine users co-prescribed with alternative beta-blockers was comparable to those co-medicated with metoprolol (about 50%). Less than 5% of patients received a substitute therapy after using metoprolol-paroxetine/fluoxetine. Most of the metoprolol users (90%) received a low dose (mean DDD = 0.47) regardless whether they were prescribed paroxetine/fluoxetine.

CONCLUSION

Despite the signalling software, metoprolol-paroxetine/fluoxetine combinations are still observed in the elderly population. The clinical impact of these interactions needs further investigation. Copyright © 2017 John Wiley & Sons, Ltd.

Can We Rely on Pharmacy Claims Databases to Ascertain Maternal Use of Medications during Pregnancy?

BACKGROUND

Administrative databases are increasingly used to measure drug exposure in perinatal pharmacoepidemiology. We aimed to estimate the concordance between records of prescriptions filled in pharmacies and self-reported drug use during pregnancy.

METHODS

Data on self-reported medication use were collected at each trimester of pregnancy among a sub-sample from the Organization of Teratology Information Specialists Antidepressants in Pregnancy Cohort. Women were eligible if they were Quebec resident and provided their pharmacist's contact information. Maternal self-reports were compared with prescriptions filled in pharmacies, which are transferred to pharmaceutical services files of Quebec provincial health plan database (Régie de l'asssurance maladie du Québec). Positive and negative predictive values (PPV and NPV) for medications taken chronically (antidepressants, thyroid hormones), acutely (antibiotics), and as needed (antiemetics, asthma medications) were calculated.

RESULTS

Among the 93 participants (mean age = 30.2 ± 3.8 years), 41.9% (n = 39) took at least one antidepressant during pregnancy according to self-reports, and 39.8% (n = 37) according to pharmacy records. Other commonly used drugs were antiemetics (self-reported 22.6%, pharmacy record 24.7%), antibiotics (20.4%, 16.1%), asthma medications (15.1%, 15.1%), and thyroid hormones (10.8%, 8.6%). PPVs and NPVs were: (1) chronic medication: antidepressants PPV = 100% (95% confidence interval [CI], 100-100%), NPV = 96% (95% CI, 92-100%); thyroid hormones PPV = 100% (95% CI, 100-100%), NPV = 98% (95% CI, 95-100%); (2) Acute medication: antibiotics PPV = 87% (95% CI, 70-100%), NPV = 92% (95% CI, 86-98%); (3) as needed medications: antiemetics: PPV = 78% (95% CI, 62-95%), NPV = 96% (95% CI, 91-100%); asthma: PPV = 33% (95% CI, 3-64%), NPV = 99% (95% CI, 97-100%).

CONCLUSION

The high PPV and NPV validate the use of filled prescription data in large databases as a measure of medication exposure. Birth Defects Research 109:423-431, 2017. © 2017 Wiley Periodicals, Inc.

Association between medication prescription for atopic diseases and attention-deficit/hyperactivity disorder

BACKGROUND

Data on the association between atopic diseases and attention-deficit/hyperactivity disorder (ADHD) have been inconclusive.

OBJECTIVE

To assess whether children with drug-treated ADHD are more likely to receive treatment for asthma, allergic rhinitis, or eczema before the start of ADHD medication use compared with controls and to examine the effect of parents receiving medication for ADHD and atopic diseases on ADHD medication use in their offspring.

METHODS

We conducted a retrospective nested case-control study among children (6-12 years of age) using the Groningen University prescription database. Cases were defined as children with at least 2 prescriptions of methylphenidate within 12 months. For each case, 4 controls were matched on age, sex, and regional area code. Parental prescription data were linked to cases and controls to assess the influence of parents receiving medication for ADHD and atopic diseases on ADHD medication use in their offspring.

RESULTS

We identified 4257 cases and 17,028 matched controls. Drug treatment for asthma, allergic rhinitis, and eczema was more common in cases than controls (adjusted odds ratios [aORs], 1.4 [95% confidence interval (CI), 1.3-1.6], 1.4 [95% CI, 1.1-1.8], and 1.3 [95% CI, 1.1-1.5], respectively). Medication for allergic rhinitis and asthma among parents was associated with ADHD treatment in their children (aORs, 1.3 [95% CI, 1.1-1.5] and 1.2 [95% CI, 1.1-1.3], respectively).

CONCLUSION

This study provides further evidence to support the hypothesis that atopic diseases are associated with ADHD. The parental-offspring association suggests a possible genetic and/or environmental component.

Effects of urinary cortisol levels and resting heart rate on the risk for fatal and nonfatal cardiovascular events

BACKGROUND AND AIMS

Higher cortisol levels are associated with cardiovascular mortality in the elderly. It is unclear whether this association also exists in a general population of younger adults and for non-fatal cardiovascular events. Likewise, resting heart rate is associated with cardiovascular mortality, but fewer studies have also considered non-fatal events. The goal of this study was to investigate whether twenty-four-hour urinary cortisol (24-h UFC) levels and resting heart rate (RHR) predict major adverse fatal and non-fatal cardiovascular events (MACE) in the general population.

METHODS

We used data from a subcohort of the PREVEND study, a prospective general population based cohort study with a follow-up of 6.4 years for 24-h UFC and 10.6 years for RHR. Participants were 3432 adults (mean age 49 years, range 28-75). 24-h UFC was collected and measured by liquid chromatography-tandem mass spectrometry. RHR was measured at baseline in a supine position for 10 min with the Dinamap XL Model 9300. Information about cardiovascular events and mortality was obtained from the Dutch national registry of hospital discharge diagnoses and the municipal register respectively.

RESULTS

24-h UFC did not significantly increase the hazard of MACE (hazard ratio = 0.999, 95% confidence interval = 0.993-1.006, p = 0.814). RHR increased the risk for MACE with 17% per 10 extra heart beats per minute (hazard ratio = 1.016, 95% confidence interval = 1.001-1.031, p = 0.036) after adjustment for conventional risk factors.

CONCLUSIONS

In contrast to 24-h UFC, RHR is a risk marker for MACE in the general population.

Medication use during pregnancy, gestational age and date of delivery: agreement between maternal self-reports and health database information in a cohort

Background

Health databases are a promising resource for epidemiological studies on medications safety during pregnancy. The reliability of information on medications exposure and pregnancy timing is a key methodological issue. This study (a) compared maternal self-reports and database information on medication use, gestational age, date of delivery; (b) quantified the degree of agreement between sources; (c) assessed predictors of agreement.

Methods

Pregnant women recruited in a prenatal clinic in Friuli Venezia Giulia (FVG) region, Italy, from 2007 to 2009, completed a questionnaire inquiring on medication use during pregnancy, gestational age and date of delivery. Redeemed prescriptions and birth certificate records were extracted from regional databases through record linkage. Percent agreement, Kappa coefficient, prevalence and bias-adjusted Kappa (PABAK) were calculated. Odds Ratio (OR), with 95 % confidence interval (95 % CI), of ≥1 agreement was calculated through unconditional logistic regression.

Results

The cohort included 767 women, 39.8 % reported medication use, and 70.5 % were dispensed at least one medication. Kappa and PABAK indicated almost perfect to substantial agreement for antihypertensive medications (Kappa 0.86, PABAK 0.99), thyroid hormones (0.88, 0.98), antiepileptic medications (1.00, 1.00), antithrombotic agents (0.70, 0.96). PABAK value was greater than Kappa for medications such as insulin (Kappa 0.50, PABAK 0.99), antihistamines for systemic use (0.50, 0.99), progestogens (0.28, 0.79), and antibiotics (0.12, 0.63). Adjusted OR was 0.48 (95 % CI 0.26; 0.90) in ex- vs. never smokers, 0.64 (0.38; 1.08) in < high school vs. university, 1.55 (1.01; 2.37) in women with comorbidities, 2.25 (1.19; 4.26) in those aged 40+ vs. 30–34 years.

Gestational age matched exactly in 85.2 % and date of delivery in 99.5 %.

Conclusions

For selected medications used for chronic conditions, the agreement between self-reports and dispensing data was high. For medications with low to very low prevalence of use, PABAK provides a more reliable measure of agreement. Maternal reports and dispensing data are complementary to each other to increase the reliability of information on the use of medications during pregnancy. Birth certificates provide reliable data on the timing of pregnancy. FVG health databases are a valuable source of data for pregnancy research.

Electronic supplementary material

The online version of this article (doi:10.1186/s12884-015-0745-3) contains supplementary material, which is available to authorized users.

Clinical medication reviews in elderly patients with polypharmacy: a cross-sectional study on drug-related problems in the Netherlands

Background Knowledge of drug-related problems (DRPs) identified in the medication of home-dwelling elderly patients with polypharmacy has been based predominantly on medication reviews conducted in research settings rather than in daily practice. Objective To evaluate the prevalence of DRPs identified by means of a clinical medication review (CMR) and the implementation rate of proposed interventions in a large group of older patients with polypharmacy in the daily practice of community pharmacies. Setting 318 Dutch community pharmacies. Method A cross-sectional study based on CMR-data of 3807 older patients (≥65 years) with polypharmacy (≥5 drugs) completed between January and August 2012. Data were extracted from community pharmacists’ databases and entailed: year of birth, gender, dispensing data, number and nature of identified DRPs, consultations performed, proposed and implemented interventions. Main outcome measure Prevalence of DRPs, drug classes involved in overtreatment and undertreatment, and proposed and implemented interventions. Results A median of two DRPs (interquartile range 1–4; mean 3.0) was identified per patient. The DRP-categories overtreatment (25.5 %) and undertreatment (15.9 %) were found most frequently. 46.2 % of the proposed interventions to solve DRPs were implemented as proposed, in 22.4 % of cases, the intervention differed from the proposal. In 31.3 % of cases no intervention was implemented. Conclusion By conducting a CMR community pharmacists identified a median of two DRPs in older patients with polypharmacy. Overtreatment and undertreatment accounted for 41.4 % of the DRPs identified. In dealing with DRPs, pharmacists proposed a variety of interventions of which the majority (69.9 %) was either implemented or led to alternative interventions. A set of explicit criteria should be applied during a CMR to solve and prevent DRPs.

Cost-effectiveness of extended-release methylphenidate in children and adolescents with attention-deficit/hyperactivity disorder sub-optimally treated with immediate release methylphenidate

Background

Attention-Deficit/Hyperactivity Disorder (ADHD) is a common psychiatric disorder in children and adolescents. Immediate-release methylphenidate (IR-MPH) is the medical treatment of first choice. The necessity to use several IR-MPH tablets per day and associated potential social stigma at school often leads to reduced compliance, sub-optimal treatment, and therefore economic loss. Replacement of IR-MPH with a single-dose extended release (ER-MPH) formulation may improve drug response and economic efficiency.

Objective

To evaluate the cost-effectiveness from a societal perspective of a switch from IR-MPH to ER-MPH in patients who are sub-optimally treated.

Methods

A daily Markov-cycle model covering a time-span of 10 years was developed including four different health states: (1) optimal response, (2) sub-optimal response, (3) discontinued treatment, and (4) natural remission. ER-MPH options included methylphenidate osmotic release oral system (MPH-OROS) and Equasym XL/Medikinet CR. Both direct costs and indirect costs were included in the analysis, and effects were expressed as quality-adjusted life years (QALYs). Univariate, multivariate as well as probabilistic sensitivity analysis were conducted and the main outcomes were incremental cost-effectiveness ratios.

Results

Switching sub-optimally treated patients from IR-MPH to MPH-OROS or Equasym XL/Medikinet CR led to per-patient cost-savings of €4200 and €5400, respectively, over a 10-year treatment span. Sensitivity analysis with plausible variations of input parameters resulted in cost-savings in the vast majority of estimations.

Conclusions

This study lends economic support to switching patients with ADHD with suboptimal response to short-acting IR-MPH to long-acting ER-MPH regimens.

Bilirubin as a potential causal factor in type 2 diabetes risk: a Mendelian randomization study

Circulating bilirubin, a natural antioxidant, is associated with decreased risk of type 2 diabetes (T2D), but the nature of the relationship remains unknown. We performed Mendelian randomization in a prospective cohort of 3,381 participants free of diabetes at baseline (age 28-75 years; women 52.6%). We used rs6742078 located in the uridine diphosphate-glucuronosyltransferase locus as an instrumental variable (IV) to study a potential causal effect of serum total bilirubin level on T2D risk. T2D developed in a total of 210 participants (6.2%) during a median follow-up period of 7.8 years. In adjusted analyses, rs6742078, which explained 19.5% of bilirubin variation, was strongly associated with total bilirubin (a 0.68-SD increase in bilirubin levels per T allele; P < 1 × 10(-122)) and was also associated with T2D risk (odds ratio [OR] 0.69 [95% CI 0.54-0.90]; P = 0.006). Per 1-SD increase in log-transformed bilirubin levels, we observed a 25% (OR 0.75 [95% CI 0.62-0.92]; P = 0.004) lower risk of T2D. In Mendelian randomization analysis, the causal risk reduction for T2D was estimated to be 42% (causal OR for IV estimation per 1-SD increase in log-transformed bilirubin 0.58 [95% CI 0.39-0.84]; P = 0.005), which was comparable to the observational estimate (Durbin-Wu-Hausman χ(2) test, P for difference = 0.19). These novel results provide evidence that an elevated bilirubin level is causally associated with the risk of T2D and support its role as a protective determinant.

Does a cardiovascular event change adherence to statin treatment in patients with type 2 diabetes? A matched cohort design

OBJECTIVE

To be effective, adherence to statin treatment is essential. We assessed the effect of an apparent first cardiovascular event on statin adherence rates in type 2 diabetes patients.

RESEARCH DESIGN AND METHODS

A matched cohort study was conducted among type 2 diabetes patients initiating statin treatment for primary prevention in the Groningen University IADB.nl pharmacy database. Patients who had a drug-treated cardiovascular event (index date) after statin initiation were matched to a reference patient without such an event with similar gender, age at statin initiation, initiation date, follow-up period and adherence level before the event. Adherence rates were measured as percentages of days covered (PDC), and shifts in adherence levels (non-adherent/partially adherent/fully adherent) and rates around the event were evaluated.

RESULTS

We could match 375 of the 855 eligible index patients to a reference patient. Index patients had on average a PDC of 81% after the index date; reference patients had a PDC of 71% (p < 0.001) while both had a PDC of 79% before the index date. Index patients were 4.5 times more likely than reference patients to shift from non-adherent to fully adherent (95% CI 1.1-18.8) and 1.8 times more likely to shift from partially adherent to fully adherent (95% CI 1.2-2.6). In the index group, 26% of patients became more adherent after the first cardiovascular event. In contrast, 20% of patients became less adherent.

LIMITATIONS

Medication proxies were used, which could have caused misclassification. Furthermore, a substantial group of index patients could not be matched to a reference patient due to small ranges in matching criteria.

CONCLUSIONS

The occurrence of a drug-treated cardiovascular event appeared to avert the declining statin adherence rate observed in diabetes patients without such an event. On the other hand, one in five patients became less adherent after the event, indicating that there are still important benefits to achieve.

The association of albuminuria with tubular reabsorption of uric acid: results from a general population cohort

Background

Elevated albuminuria as well as an increased serum uric acid concentration is associated with poor cardiovascular outcome. We questioned whether these 2 variables (albuminuria and serum uric concentration) may be interrelated via tubular uric acid reabsorption.

Methods and Results

Included were 7688 participants of the PREVEND Study, an observational, general population‐based cohort study. Linear regression analyses were used to test associations of baseline albuminuria with baseline serum uric acid concentration and tubular uric acid reabsorption (calculated as [100−fractional uric acid excretion]%). Cox regression analyses were used to study the association of baseline serum uric acid and albuminuria with incident cardiovascular morbidity and mortality. In cross‐sectional analyses, albuminuria was associated positively with serum uric acid concentration, both crude and after adjustment for potential confounders (both P<0.001). Albuminuria was found to be associated positively with tubular uric acid reabsorption, again both crude and after adjustment for potential confounders (both P<0.001). In longitudinal analyses during a median follow‐up of 10.5 years, 702 cardiovascular events occurred. After adjusting for cardiovascular risk factors, both albuminuria and serum uric acid were associated with incident cardiovascular events (Hazard Ratios 1.09 [1.03 to 1.17], P=0.01 and 1.19 [1.09 to 1.30], P<0.001, respectively). A significant interaction between these variables was present (P<0.001), consistent with high serum uric acid being less predictive for cardiovascular morbidity and mortality in the presence of high albuminuria and vice versa.

Conclusions

Albuminuria is strongly associated with tubular uric acid reabsorption, and consequently with serum uric acid concentration. This phenomenon may explain in part why albuminuria is associated with cardiovascular outcome.

A comparison of self-reported oral contraceptive use and automated pharmacy data in perimenopausal and early postmenopausal women

PURPOSE

Oral contraceptive (OC) use can occur throughout a woman's reproductive life span with the potential for long-term impacts on health. To assess potential measurement error in prior OC use, this study compared level of agreement between self-reported prior OC use and pharmacy dispensing data in perimenopausal and/or early postmenopausal women.

METHODS

The study's 1399 women (ages, 45-59 years) were participants in a population-based case-control study of the association between OC use and fracture risk. Episodes of lifetime self-reported OC use (in months) were collected, by telephone interview, for January 1, 2008 through November 25, 2012. Pharmacy fills, back to 1980, were collected from automated data. Agreement was measured using the prevalence-adjusted and bias-adjusted kappa index.

RESULTS

The number of women with OC pharmacy fills was 11% to 45% higher than those who reported OC use during each time period. Between-measures agreement was better for more recent use. Prevalence-adjusted and bias-adjusted kappa index values ranged from 0.88 (95% confidence interval, 0.85-0.90) within 5 years from the reference date to 0.65 (95% confidence interval, 0.59-0.71) within 15 to 20 years.

CONCLUSIONS

For studies designed to assess the long-term effects of OC use, the current results are reassuring in noting moderate agreement between self-reported OC use and pharmacy data for up to 15 to 20 years before the interview.

24 h urinary free cortisol in large-scale epidemiological studies: short-term and long-term stability and sources of variability

BACKGROUND

Function of the hypothalamus-pituitary-adrenal (HPA) axis has been associated with several somatic and psychiatric health problems. The amount of free cortisol excreted in the urine during 24h (24-h UFC) has often been used as a proxy for HPA-axis function. Reference values for 24-h UFC and their stability in the short and long term, as well as sources of variability, are largely lacking.

METHODS

This study was performed in a general population cohort. Participants collected 24-h UFC on two consecutive days (T1), and repeated this collection approximately 2 years later (T2). Cortisol in urine was measured using LC-MS/MS. Height and weight were measured at the research facilities; glomerular filtration rate was estimated using creatinine clearance. Psychological distress (General Health Questionnaire), smoking, alcohol use and exercise were measured by means of questionnaires.

RESULTS

24-h UFC stability on a day-to-day basis was 0.69 (T1, N=1192) and 0.72 (T2, N=963) (both p<0.001). Long-term stability as indicated by correlation between 2-day averages of T1 and T2 was 0.60 (N=972, p<0.001). Multivariable linear regression analysis revealed that 24-h UFC was predicted by urine volume (standardized beta 0.282 (T1, N=1556) and 0.276 (T2, N=1244); both p<0.001) and glomerular filtration rate (standardized beta 0.137 (T1) and 0.179 (T2); both p<0.001), while also sex explained a small part (standardized beta for female sex -0.057 (T1) and -0.080 (T2); both p<0.05).

CONCLUSION

24-h UFC is moderately stable both in the short and the long term. The effects of urine volume and glomerular filtration rate on 24-h UFC are much stronger than those of sex.

Maternal high-dose folic acid during pregnancy and asthma medication in the offspring

PURPOSE

Low-dose folic acid supplementation (0.5 mg) taken during pregnancy has been associated with an increased risk for childhood asthma. The effect of high-dose folic acid (5 mg) advised to women at risk for having a child with neural tube defect has not been assessed so far. Our aim was to investigate the effect of dispensed high-dose folic acid during pregnancy and asthma medication in the offspring.

METHODS

We used data from the pregnancy database IADB.nl, which contains pharmacy-dispensing data of mothers and children from community pharmacies in the Netherlands from 1994 until 2011. The dispension of asthma medication in children exposed in utero to high-dose folic acid was compared with children who were not exposed to this high dose. Incidence rate ratios (IRRs) with 95% confidence intervals (CIs) were calculated.

RESULTS

In 2.9% (N = 913) of the 39,602 pregnancies in the database, the mother was dispensed high-dose folic acid. Maternal high-dose folic acid was associated with an increased rate of asthma medication among children: recurrent asthma medication IRR = 1.14 (95%CI: 1.04-1.30) and recurrent inhaled corticosteroids IRR = 1.26 (95%CI: 1.07-1.47). Associations were clustered on the mother and adjusted for maternal age, maternal asthma medication, and dispension of benzodiazepines during pregnancy.

CONCLUSION

Almost 3% of the children were prenatally exposed to high-dose folic acid. This study suggests that supplementation of high-dose folic acid during pregnancy might increase the risk of childhood asthma.

Comparison of delivery strategies for pharmacogenetic testing services

The number and use of pharmacogenetic tests to assess a patient's likelihood of response or risk of an adverse event is expanding across medical specialties and becoming more prevalent. During this period of development and translation, different approaches are being investigated to optimize delivery of pharmacogenetic services. In this paper, we review pre-emptive and point-of-care delivery approaches currently implemented or being investigated and discuss the advantages and disadvantages of each approach. The continued growth in knowledge about the genetic basis of drug response combined with development of new and less expensive testing technologies and electronic medical records will impact future delivery systems. Regardless of delivery approach, the currently limited knowledge of health professionals about genetics generally or PGx specifically will remain a major obstacle to utilization.

The interaction of age and type 2 diabetes on executive function and memory in persons aged 35 years or older

It is generally assumed that type 2 diabetes increases the risk of cognitive dysfunction in old age. As type 2 diabetes is frequently diagnosed before the age of 50, diabetes-related cognitive dysfunction may also occur before the age of 50. Therefore, we investigated the association of type 2 diabetes with cognitive function in people aged 35-82 years. In a cross-sectional study comprising 4,135 participants of the Prevention of Renal and Vascular ENd-stage Disease study (52% men; mean age (SD), 55 (12) years) diabetes was defined according to the criteria of the American Diabetes Association. Executive function was measured with the Ruff Figural Fluency Test (RFFT; worst score, 0 points; best score, 175 points), and memory was measured with the Visual Association Test (VAT; worst score, 0 points; best score, 12 points). The association of diabetes with cognitive function was investigated with multiple linear or, if appropriate, logistic regression analysis adjusting for other cardiovascular risk factors and APOE ε4 carriership. Type 2 diabetes was ascertained in 264 individuals (6%). Persons with diabetes had lower RFFT scores than persons without diabetes: mean (SD), 51 (19) vs. 70 (26) points (p<0.001). The difference in RFFT score was largest at age 35-44 years (mean difference 32 points; 95% CI, 15 to 49; p<0.001) and gradually decreased with increasing age. The association of diabetes with RFFT score was not modified by APOE ε4 carriership. Similar results were found for VAT score as outcome measure although these results were only borderline statistically significant (p≤0.10). In conclusion, type 2 diabetes was associated with cognitive dysfunction, especially in young adults. This was independent of other cardiovascular risk factors and APOE ε4 carriership.

Pharmacist-led intervention study to improve drug therapy in asthma and COPD patients

BACKGROUND

Pharmacists can play an important role in identifying and instructing pulmonary patients on their inhalation techniques in their patient contacts when dispensing inhalation medication. Pharmacy dispensing data can be used to identify inappropriate drug use in asthma and chronic obstructive pulmonary disease (COPD) patients. Recent studies found beneficial effects of pharmacy care services in improving drug adherence of pulmonary patients. However, large-scale and rigorous evaluations on pharmacist-led interventions in community care settings to enhance evidence-based drug treatment in patients with asthma and COPD seems to be lacking and results from studies on pharmacist-led interventions for pharmacotherapy improvements are inconsistent. This study evaluated the effectiveness of pharmacist-led interventions on suboptimal drug use patterns with asthma or COPD medication with substantial numbers of pharmacies and patients involved.

SETTING

A prospective cohort study in a group of community pharmacies (intervention group) with a matched control group of Dutch community pharmacies was conducted between May 2011 and February 2012. Algorithms on 19 potential problems with asthma or COPD medication in a national dispensing database were used to signal patients to pharmacists of the intervention group (IG).

METHODS

IG pharmacists selected patients for comprehensive care by a structured program. Changes in problems were measured during 10 months in selected and all users of asthma and COPD medication of IG pharmacies and in compared to a control pharmacies (CG) without the structured program. Primary outcome was reduction of oral high dosage corticosteroids or antibiotics (HDT). Secondary outcomes were changes in the persistence of 19 potential problems in the IG compared to CG.

RESULTS

In the 107 IG pharmacies, 3,757 patients were selected for comprehensive care from totally 102,497 asthma or COPD patients and compared with 105,507 patients from 105 CG pharmacies. The mean number of HDT decreased in selected IG patients by an additional 0.54 (95 % CI 0.21-0.86) HDT treatments. From the problems with specific COPD and asthma medication, all problems decreased additionally to the CG within the total asthma or COPD population from the IG. Within the selected IG population the following problems decreased additionally for obsolete medication by 35 % (95 % CI 6-54 %), contra-indicated medication by 61 % (95 % CI 38-75 %) and lower use of powder inhalers in elderly patients by 29 % (95 % CI 13-42 %).

CONCLUSION

Community pharmacists actively providing comprehensive pharmacy care could improve effective treatment in asthma and COPD patients and thereby decrease the number of prescriptions for exacerbations for these patients.

Inhaled corticosteroids and the occurrence of oral candidiasis: a prescription sequence symmetry analysis

OBJECTIVES

The primary aim of the study was to gain insight into the relative risk of clinically relevant oral candidiasis following inhaled corticosteroid (ICS) initiation over time. A secondary aim was to analyse the influence of patient characteristics and co-medication on the occurrence of this adverse effect.

METHODS

Drug prescription data from 1994 to 2011 were retrieved from the IADB.nl database. To study the influence of ICS use on occurrence of oral candidiasis, a prescription symmetry analysis was used, including patients using medication for oral candidiasis up to 1 year before or after ICS initiation. The relative risk was calculated by dividing the number of patients receiving medication for oral candidiasis after ICS initiation by the number of patients receiving the same medication before ICS initiation. Sub-analyses were conducted to compare the relative risks at several time points after ICS initiation and to account for therapy persistence by only including chronic users of ICS. A multivariate logistic regression model was used to identify predictive factors.

RESULTS

A total of 52,279 incident users of ICS therapy were identified, of which 1,081 received medication for oral candidiasis up to 1 year before or after ICS initiation. A total of 701 patients received medication for oral candidiasis after ICS initiation, while 361 received these medications in the reversed sequence, resulting in a sequence ratio (SR) of 1.94 (95 % CI 1.71-2.21). In the first 3 months after ICS initiation, the SR was 2.72 (95 % CI 2.19-3.38) and then decreased to 1.47 (95 % CI 1.11-1.95) 9-12 months after ICS initiation. Predictive factors were higher daily dose of ICS and concomitant use of oral corticosteroids.

CONCLUSIONS

This study found a significant and clinically relevant increased number of patients receiving medication for oral candidiasis in the first year after therapy initiation with ICS. Relative risk is highest in the first 3 months, but remains increased up to at least 1 year after ICS initiation. This study stresses the need for patient education and inhalation instruction.

Using spatial analysis to predict health care use at the local level: a case study of type 2 diabetes medication use and its association with demographic change and socioeconomic status

Local health status and health care use may be negatively influenced by low local socio-economic profile, population decline and population ageing. To support the need for targeted local health care, we explored spatial patterns of type 2 diabetes mellitus (T2DM) drug use at local level and determined its association with local demographic, socio-economic and access to care variables. We assessed spatial variability in these associations. We estimated the five-year prevalence of T2DM drug use (2005-2009) in persons aged 45 years and older at four-digit postal code level using the University of Groningen pharmacy database IADB.nl. Statistics Netherlands supplied data on potential predictor variables. We assessed spatial clustering, correlations and estimated a multiple linear regression model and a geographically weighted regression (GWR) model. Prevalence of T2DM medicine use ranged from 2.0% to 25.4%. The regression model included the extent of population ageing, proportion of social welfare/benefits, proportion of low incomes and proportion of pensioners, all significant positive predictors of local T2DM drug use. The GWR model demonstrated considerable spatial variability in the association between T2DM drug use and above predictors and was more accurate. The findings demonstrate the added value of spatial analysis in predicting health care use at local level.

Selection bias in pharmacy-based patient surveys

PURPOSE

To evaluate if there is a selection bias in drug utilization surveys on prescription drugs regularly conducted in Swedish pharmacies, to describe the direction of this potential bias and discuss the implications for the results.

METHODS

Age and gender distributions within patient survey samples from drug utilization surveys conducted during 2006-2010 are compared to the age and gender distribution of all Swedish patients, receiving the same drug or drugs, as given by the Swedish Prescribed Drug Register. The differences between the proportions of patients within the age and gender segments of each pair of survey/register data were calculated.

RESULTS

In 25 (81%) out of 31 included surveys, patients aged 75 years or older are significantly underrepresented, as they are less likely to visit the pharmacy to collect their prescription drugs themselves and thus disqualify for the interviews. Data on women show similar results as overall survey data, whereas the underrepresentation of the oldest age group among men appears in a lower proportion of the surveys, 67%. The general consequence is a selection towards a healthier survey sample, but the consequences in the individual surveys vary, depending on what drug is being studied.

CONCLUSION

Pharmacy-based patient surveys provide a convenient data collection method for patient self-reported data, but patients aged 75 years or older are consistently underrepresented. In surveys where this may influence the main research question, data should also be collected with other methods reaching the oldest patients.

Agreement between patient interview data on prescription medication use and pharmacy records in those aged older than 50 years varied by therapeutic group and reporting of indicated health conditions

OBJECTIVES

To estimate the agreement between interview-ascertained medication use and pharmacy records among the population aged older than 50 years, and to identify patient-level predictors of discordance.

STUDY DESIGN AND SETTING

The Irish Longitudinal study on Ageing is representative of community-dwelling adults aged 50 years and older in Ireland. Interview-ascertained medication data from 2,621 participants were linked to pharmacy dispensing records. The kappa statistics measured the agreement between the two sources for 19 therapeutic classes. Logistic regression assessed the effect of patient-level characteristics on survey under- and overreporting of regularly dispensed medications.

RESULTS

Agreement was good or very good (κ=0.64-0.86) for 15 medication classes, and moderate or poor for antiinflammatory and antirheumatic products (κ=0.54), analgesics (κ=0.50), psycholeptics (κ=0.59), and ophthalmologicals (κ=0.37). Not reporting an indicated health condition, less frequent dispensing, older age, and more medications regularly dispensed were associated with survey underreporting, but results varied by therapeutic class. Memory and cognition were not associated with discordance.

CONCLUSION

Ascertaining medication use via patient interview seems a valid method for most medication classes and also captures nonprescription and supplement use. However, medications applied topically and as needed may be underreported. The source of medication data should be carefully considered when performing pharmacoepidemiological studies.

Antibiotics prescribed before, during and after pregnancy in the Netherlands: a drug utilization study

PURPOSE

To describe the prescription of antibiotics before, during and after pregnancy, and the trends over a 16-year period in the Netherlands, and to determine whether they were prescribed according to national guidelines.

METHODS

The IADB (http://iadb.nl) contains prescriptions dispensed by community pharmacies in the Netherlands. We extracted information on 18,873 pregnancies for 14,969 women between 1994 and 2009, focusing on antibiotics dispensed in the four trimesters before conception to two trimesters after birth (nine trimesters in total). We calculated trends in prescription rates during pregnancy and over time, and also compared the prescription of antibiotics in the Netherlands with safety category based on the Australian Drug Evaluation Committee.

RESULTS

During pregnancy 20.8% of the women were prescribed at least one antibiotic. The 'beta-lactam antibacterials/penicillins' group and the specific antibiotic amoxicillin were most commonly prescribed in the nine trimesters covered. The prescription rate of the 'other antibacterials' group during pregnancy increased over the years, in contrast to that of the 'sulphonamides/trimethoprim' group, which decreased. In total, 2.0% of pregnancies were exposed to a 'potentially harmful' antibiotic and 0.8% to a 'harmful' antibiotic. Compared with the period before conception, 'safe' antibiotics were prescribed more often during pregnancy than the other groups.

CONCLUSIONS

One in five women was prescribed at least one antibiotic during pregnancy in the Netherlands, which is comparable with rates in other European countries. Our results suggest that antibiotics appear to be prescribed to pregnant women generally in accordance with national recommendations.

Role of HDL cholesterol and estimates of HDL particle composition in future development of type 2 diabetes in the general population: the PREVEND study

BACKGROUND AND AIMS

High-density lipoproteins (HDLs) may directly stimulate β-cell function and glucose metabolism. We determined the relationships of fasting high-density lipoprotein cholesterol (HDL-C), plasma apolipoprotein (apo) A-I and apoA-II, and HDL-C-to-apoA-I and HDL-C-to-apoA-II ratios, as estimates of HDL particle composition, with incident type 2 diabetes mellitus.

METHODS

A prospective study was carried out in the Prevention of Renal and Vascular End-Stage Disease (PREVEND) cohort after exclusion of subjects with diabetes at baseline (n = 6820; age, 28-75 years). The association of HDL-related variables with incident type 2 diabetes was determined by multivariate logistic regression analyses.

RESULTS

After a median follow-up of 7.7 years, 394 incident cases of type 2 diabetes mellitus were ascertained (5.8%). After adjustment for age, sex, family history of diabetes, body mass index, hypertension, alcohol, and smoking, odd ratios (ORs) for diabetes were 0.55 (95% confidence interval [CI], 0.47-0.64; P < .001), 0.81 (0.71-0.93; P = .002), 0.02 (0.01-0.06; P < .001), and 0.03 (0.01-0.060; P < .001) per 1-SD increase in HDL-C and apoA-I and in the HDL-C-to-apoA-I and the HDL-C-to-apoA-II ratios, respectively. In contrast, apoA-II was not related to incident diabetes (OR = 1.02; 95% CI, 0.90-1.16; P=0.71). The relationships of HDL-C and the ratios of HDL-C to apoA-I and HDL-C to apoA-II remained significant after further adjustment for baseline glucose and triglycerides (OR(HDL) = 0.74 [95% CI, 0.61-0.88], OR(HDL/APO A-I) = 0.14 [0.04-0.44], and OR(HDL/APOA-II) = 0.12 [0.04-0.36]; all P ≤ .001).

CONCLUSIONS

Higher HDL-C, as well as higher HDL-C-to-apoA-I and HDL-C-to-apoA-II ratios are strongly and independently related to a lower risk of future type 2 diabetes.

Association between anxiety but not depressive disorders and leukocyte telomere length after 2 years of follow-up in a population-based sample

BACKGROUND

Telomere length is considered an emerging marker of biological aging. Depression and anxiety are associated with excess mortality risk but the mechanisms remain obscure. Telomere length might be involved because it is associated with psychological distress and mortality. The aim of this study was to test whether anxiety and depressive disorders predict telomere length over time in a large population-based sample. Method All analyses were performed in a longitudinal study in a general population cohort of 974 participants. The Composite International Diagnostic Interview (CIDI) was used to measure the presence of anxiety and depressive disorders. Telomere length was measured using monochrome multiplex polymerase chain reaction (PCR) at approximately 2 years of follow-up. We used linear multivariable regression models to evaluate the association between anxiety and depressive disorders and telomere length, adjusting for adverse life events, lifestyle factors, educational level and antidepressant use.

RESULTS

The presence of anxiety disorders predicted shorter telomeres at follow-up (β = -0.073, t = -2.302, p = 0.022). This association was similar after controlling for adverse life events, lifestyle factors, educational level and antidepressant use (β = -0.077, t = -2.144, p = 0.032). No association was found between depressive disorders and shorter telomeres at follow-up (β = 0.010, t = 0.315, p = 0.753).

CONCLUSIONS

This study found that anxiety disorders predicted shorter telomere length at follow-up in a general population cohort. The association was not explained by adverse life events, lifestyle factors, educational level and antidepressant use. How anxiety disorders might lead to accelerated telomere shortening and whether this might be a mediator explaining the excess mortality risk associated with anxiety deserve further investigation.

Isolated microalbuminuria indicates a poor medical prognosis

BACKGROUND

Microalbuminuria is often regarded as a sign of end-organ damage due to diabetes and/or hypertension, and as such to be associated with an increased risk for cardiovascular events. It has been questioned whether isolated microalbuminuria, that is microalbuminuria in the absence of a cardiovascular disease (CVD) history, hypertension and diabetes has clinical relevance.

METHODS

Included were 8356 subjects who participated in the first four screening rounds of the PREVEND study, a prospective, community-based, observational cohort study. Isolated microalbuminuria was defined as microalbuminuria (30-300 mg/24 h), in the absence of a CVD history, hypertension (blood pressure<140/90 mmHg, not using blood pressure-lowering drugs) and diabetes (fasting glucose<7.0 mmol/L, not using glucose-lowering drugs).

RESULTS

Three hundred subjects met the definition of isolated microalbuminuria, in which 2250 person-years of follow-up were available. In subjects with isolated microalbuminuria, the incidence rates of cardiovascular events and mortality, hypertension and diabetes were 15.3, 28.9 and 8.9 per 1000 person-year follow-up, respectively. Subjects with isolated microalbuminuria had an increased risk for cardiovascular events and mortality [crude HR 2.23 (1.63-3.07); P<0.001], hypertension [OR 1.95 (1.47-2.59); P<0.001] and diabetes [OR 4.69 (2.92-7.51); P<0.001] compared with subjects without microalbuminuria, CVD history, hypertension and/or diabetes. This increased risk remained significant after adjustment for age and gender. The relative risk held by isolated microalbuminuria was similar to the relative risk held by microalbuminuria in subjects that did have a CVD history, hypertension and/or diabetes.

CONCLUSIONS

Isolated microalbuminuria indicates a poor prognosis and warrants medical attention.

Peroxiredoxin 4, a novel circulating biomarker for oxidative stress and the risk of incident cardiovascular disease and all-cause mortality

BACKGROUND

Oxidative stress has been suggested to play a key role in the development of cardiovascular disease (CVD). The aim of our study was to investigate the associations of serum peroxiredoxin 4 (Prx4), a hydrogen peroxide-degrading peroxidase, with incident CVD and all-cause mortality. We subsequently examined the incremental value of Prx4 for the risk prediction of CVD compared with the Framingham risk score (FRS).

METHODS AND RESULTS

We performed Cox regression analyses in 8141 participants without history of CVD (aged 28 to 75 years; women 52.6%) from the Prevention of Renal and Vascular End-stage Disease (PREVEND) study in Groningen, The Netherlands. Serum Prx4 was measured by an immunoluminometric assay in baseline samples. Main outcomes were: (1) incident CVD events or CVD mortality and (2) all-cause mortality during a median follow-up of 10.5 years. In total, 708 participants (7.8%) developed CVD events or CVD mortality, and 517 participants (6.3%) died. Baseline serum Prx4 levels were significantly higher in participants with incident CVD events or CVD mortality and in those who died than in participants who remained free of outcomes (both P<0.001). In multivariable models with adjustment for Framingham risk factors, hazard ratios were 1.16 (95% CI 1.06 to 1.27, P<0.001) for incident CVD events or CVD mortality and 1.17 (95% CI 1.06 to 1.29, P=0.003) for all-cause mortality per doubling of Prx4 levels. After the addition of Prx4 to the FRS, the net reclassification improvement was 2.7% (P=0.01) using 10-year risk categories of CVD.

CONCLUSIONS

Elevated serum Prx4 levels are associated with a significantly higher risk of incident CVD events or CVD mortality and all-cause mortality after adjustment for clinical risk factors. The addition of Prx4 to the FRS marginally improved risk prediction of future CVD.

Validity of chronic drug exposure presumed from repeated patient interviews varied according to drug class

OBJECTIVE

To evaluate the validity of chronic drug exposure presumed from cross-sectional interviews taking reimbursement data as reference.

STUDY DESIGN AND SETTING

The study concerned 2,985 elderly persons of the French Three-City cohort (1) who were interviewed on current drug use 2 and 4 years after inclusion and (2) whose reimbursement data were obtained from the main health care insurance system. Validity (sensitivity, Se; specificity, Sp; positive predictive value, PPV; negative predictive value, NPV) of chronic exposure presumed from follow-up interviews was investigated taking two exposure definitions from reimbursements as reference for the period between interviews: at least 80% coverage with and without a maximal time between reimbursements of 60 days.

RESULTS

Using 80% coverage as reference, validity of interview data was substantial for cardiovascular and antithrombotic drugs (Se, 85.3-95.4%; Sp, 67.1-97.6%; PPV, 65.9-86.6%; NPV, 93.3-99.3%). For benzodiazepines, nonsteroidal anti-inflammatory drugs, or analgesics, validity was low especially owing to PPVs (15.8-51.4%).

CONCLUSION

Using reported use at cross-sectional interviews as a proxy for chronic exposure between interviews was valid for drugs used regularly but not so for drugs used more irregularly.

Sex differences in the association between plasma copeptin and incident type 2 diabetes: the Prevention of Renal and Vascular Endstage Disease (PREVEND) study

AIMS/HYPOTHESIS

Vasopressin plays a role in osmoregulation, glucose homeostasis and inflammation. Therefore, plasma copeptin, the stable C-terminal portion of the precursor of vasopressin, has strong potential as a biomarker for the cardiometabolic syndrome and diabetes. Previous results were contradictory, which may be explained by differences between men and women in responsiveness of the vasopressin system. The aim of this study was to evaluate the usefulness of copeptin for prediction of future type 2 diabetes in men and women separately.

METHODS

From the Prevention of Renal and Vascular Endstage Disease (PREVEND) study, 4,063 women and 3,909 men without diabetes at baseline were included. A total of 208 women and 288 men developed diabetes during a median follow-up of 7.7 years.

RESULTS

In multivariable-adjusted models, we observed a stronger association of copeptin with risk of future diabetes in women (OR 1.49 [95% CI 1.24, 1.79]) than in men (OR 1.01 [95% CI 0.85, 1.19]) (p (interaction) < 0.01). The addition of copeptin to the Data from the Epidemiological Study on the Insulin Resistance Syndrome (DESIR) clinical model improved the discriminative value (C-statistic,+0.007, p = 0.02) and reclassification (integrated discrimination improvement [IDI] = 0.004, p < 0.01) in women. However, we observed no improvement in men. The additive value of copeptin in women was maintained when other independent predictors, such as glucose, high sensitivity C-reactive protein (hs-CRP) and 24 h urinary albumin excretion (UAE), were included in the model.

CONCLUSIONS/INTERPRETATION

The association of plasma copeptin with the risk of developing diabetes was stronger in women than in men. Plasma copeptin alone, and along with existing biomarkers (glucose, hs-CRP and UAE), significantly improved the risk prediction for diabetes in women.

Cholesteryl Ester Transfer Protein (CETP) genotype and cognitive function in persons aged 35 years or older

Common polymorphisms of the Cholestryl Ester Transfer Protein (CETP) gene may predict lower risk of cognitive decline. We investigated the association of cognitive function with CETP genotype in a population-based cohort of 4135 persons aged 35-82 years. Cognitive function was measured with the Ruff Figural Fluency Test (RFFT; worst score, 0 points; best score, 175 points) and CETP I405V and Taq1B genotypes were determined by polymerase chain reaction. RFFT score was not associated with I405V genotype in persons aged 35-64 years. Remarkably, beyond age 65, homozygous valine carriers had higher RFFT scores than heterozygous carriers and noncarriers: RFFT (SD), 52 (21), 49 (18), and 47 (17) points, respectively (p = 0.005). There also was a statistically significant interaction between I405V genotype and age. Beyond age 65, the difference between homozygous valine carriers and noncarriers increased by 0.11 point per year (p = 0.005). RFFT score was not associated with Taq1B genotype. In conclusion, CETP I405V valine homozygosity was associated with better cognitive function in persons aged 65 years or older.

Comparison of anti-diabetic drug prescribing in children and adolescents in seven European countries

AIMS

The aim of this study was to compare the prevalence of diabetes in children across seven European countries, when using prescribing of anti-diabetics as a proxy for diabetes. A secondary aim was to assess the potential for collaboration between countries using different databases in diabetes research.

METHODS

Data were obtained from population-based clinical databases in seven European countries. The study population comprised children aged 0-18 years. Prescriptions were categorized using the Anatomic Therapeutic Chemical (ATC) classification. The one-year user prevalence in 2008 was calculated for each country and stratified by age and sex.

RESULTS

We studied a total of 5.8 million children and adolescents. The prevalence of insulin prescribing varied between 1.1 and 3.5 per 1000 population, being highest in Sweden and lowest in Italy. In all countries, novel insulin analogues were the most commonly used insulins. The prevalence of oral anti-diabetic prescribing ranged from 0.08 per 1000 individuals in Sweden and Germany to 0.21 per 1000 population in the UK. Overall, the absolute number of oral anti-diabetic users was very low.

CONCLUSION

This study shows that there is a varying frequency of type 1 diabetes in children and adolescents across Europe. We also demonstrated that it is possible to obtain similar information from different clinical databases within Europe, which would allow continuous monitoring of type 1 diabetes. Owing to the lack of indications in most of the databases, this approach is less suitable for type 2 diabetes.

Implications of the accuracy of MEPS prescription drug data for health services research

This paper assesses the quality of the Medical Expenditure Panel Survey (MEPS) drug data and the impact that misreporting prescription drug data has on descriptive and behavioral analyses. It does this by matching MEPS participants with Medicare Part D coverage during the period 2006-2007 to their Part D claims data. In the validation sample, the number of drug fills and total expenditures are reasonably accurate compared with claims. Household respondents tended to underreport the number of different drugs taken, but tended to overreport the number of fills of each drug. Behavioral analyses of the determinants of medication use and expenditures were largely unaffected because underreporting cut across most sociodemographic groups.

The association of APOE genotype with cognitive function in persons aged 35 years or older

APOE genotype is associated with the risk of Alzheimer's disease. In the present study, we investigated whether APOE genotype was associated with cognitive function in predominantly middle-aged persons. In a population-based cohort of 4,135 persons aged 35 to 82 years (mean age (SD), 55 (12) years), cognitive function was measured with the Ruff Figural Fluency Test (RFFT; worst score, 0 points; best score, 175 points). APOE genotype (rs429358 and rs7412) was determined by polymerase chain reaction. The mean RFFT score (SD) of the total cohort was 69 (26) points. Unadjusted, the mean RFFT score in homozygous APOE ε4 carriers was 4.66 points lower than in noncarriers (95% confidence interval, -9.84 to 0.51; p = 0.08). After adjustment for age and other risk factors, the mean RFFT score in homozygous APOE ε4 carriers was 5.24 points lower than in noncarriers (95% confidence interval, -9.41 to -1.07; p = 0.01). The difference in RFFT score was not dependent on age. There was no difference in RFFT score between heterozygous APOE ε4 carriers and noncarriers. The results indicated that homozygous APOE ε4 carriers aged 35 years or older had worse cognitive function than heterozygous carriers and noncarriers.