Pharmacoepidemiology and Drug Safety Journal Impact Factor & Information

Publisher: International Society for Pharmacoepidemiology; International Society of Pharmacovigilance, Wiley

Journal description

The aim of Pharmacoepidemiology and Drug Safety is to provide an international forum for the communication and evaluation of data methods and opinion in the emerging discipline of pharmacoepidemiology. The Journal publishes peer-reviewed reports of original research invited reviews and a variety of guest editorials and commentaries embracing scientific medical statistical and legal aspects of pharmacoepidemiology and post-marketing surveillance of drug safety. Appropriate material in these categories may also be considered for publication as a Brief Communication. Particular areas of interest include: design results analysis and interpretation of post-marketing surveillance and other studies looking at specific drugs populations and outcomes methods for detection and evaluation of drug-associated adverse events assessments of risk versus benefit ratios in drug therapy patterns of drug utilization relationships between pharmacoepidemiology and the formulation and interpretation of regulatory guidelines The Publishers recognize the need of journal users to access articles through a variety of channels and are committed to providing a wide range of options in the future. The Publishers are also aware that new communication media increase the threat to authors' own rights through unauthorized distribution alteration or attribution. To enable the Publishers to make the published version of articles available as widely as possible while protecting authors' rights to be associated with their work it is essential that a Copyright Transfer Agreement form is signed for every article which is to be considered for publication in the journal. The form can be photocopied from the journal or copies can be obtained on request from the Publisher or printed from this Web site. Inclusion of a signed form with the manuscript at the original submission stage will speed up processing and eventual publication of the article.

Current impact factor: 2.94

Impact Factor Rankings

2015 Impact Factor Available summer 2016
2014 Impact Factor 2.939
2013 Impact Factor 3.172
2012 Impact Factor 2.897
2011 Impact Factor 2.528
2010 Impact Factor 2.339
2009 Impact Factor 2.527
2008 Impact Factor 2.516
2007 Impact Factor 2.475
2006 Impact Factor 2.155
2005 Impact Factor 1.773
2004 Impact Factor 2.098
2003 Impact Factor 1.257
2002 Impact Factor 1.092
2001 Impact Factor 1.33
2000 Impact Factor 0.867
1999 Impact Factor 0.848

Impact factor over time

Impact factor

Additional details

5-year impact 3.14
Cited half-life 5.30
Immediacy index 0.57
Eigenfactor 0.02
Article influence 1.21
Website Pharmacoepidemiology and Drug Safety website
Other titles Pharmacoepidemiology and drug safety (Online), Pharmacoepidemiology and drug safety
ISSN 1099-1557
OCLC 44084438
Material type Document, Periodical, Internet resource
Document type Internet Resource, Computer File, Journal / Magazine / Newspaper

Publisher details


  • Pre-print
    • Author can archive a pre-print version
  • Post-print
    • Author cannot archive a post-print version
  • Restrictions
    • 12 months embargo
  • Conditions
    • Some journals have separate policies, please check with each journal directly
    • On author's personal website, institutional repositories, arXiv, AgEcon, PhilPapers, PubMed Central, RePEc or Social Science Research Network
    • Author's pre-print may not be updated with Publisher's Version/PDF
    • Author's pre-print must acknowledge acceptance for publication
    • Non-Commercial
    • Publisher's version/PDF cannot be used
    • Publisher source must be acknowledged with citation
    • Must link to publisher version with set statement (see policy)
    • If OnlineOpen is available, BBSRC, EPSRC, MRC, NERC and STFC authors, may self-archive after 12 months
    • If OnlineOpen is available, AHRC and ESRC authors, may self-archive after 24 months
    • Publisher last contacted on 07/08/2014
    • This policy is an exception to the default policies of 'Wiley'
  • Classification

Publications in this journal

  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To develop the infrastructure to conduct timely active surveillance for safety of influenza vaccines and other medical countermeasures in the Sentinel System (formerly the Mini-Sentinel Pilot), a Food and Drug Administration-sponsored national surveillance system that typically relies on data that are mature, settled, and updated quarterly. Methods: Three Data Partners provided their earliest available ("fresh") cumulative claims data on influenza vaccination and health outcomes 3-4 times on a staggered basis during the 2013-2014 influenza season, collectively producing 10 data updates. We monitored anaphylaxis in the entire population using a cohort design and seizures in children ≤4 years of age using both a self-controlled risk interval design (primary) and a cohort design (secondary). After each data update, we conducted sequential analysis for inactivated (IIV) and live (LAIV) influenza vaccines using the Maximized Sequential Probability Ratio Test, adjusting for data-lag. Results: Most of the 10 sequential analyses were conducted within 6 weeks of the last care-date in the cumulative dataset. A total of 6 682 336 doses of IIV and 782 125 doses of LAIV were captured. The primary analyses did not identify any statistical signals following IIV or LAIV. In secondary analysis, the risk of seizures was higher following concomitant IIV and PCV13 than historically after IIV in 6- to 23-month-olds (relative risk = 2.7), which requires further investigation. Conclusions: The Sentinel System can implement a sequential analysis system that uses fresh data for medical product safety surveillance. Active surveillance using sequential analysis of fresh data holds promise for detecting clinically significant health risks early. Limitations of employing fresh data for surveillance include cost and the need for careful scrutiny of signals. © 2015 The Authors. Pharmacoepidemiology and Drug Safety Published by John Wiley & Sons Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3908
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: Electronic health data are routinely used to conduct studies of cardiovascular disease in the setting of the Veterans Health Administration (VA). Previous studies have estimated the positive predictive value (PPV) of International Classification of Disease, Ninth Revision (ICD-9) codes for acute myocardial infarction (MI), but the sensitivity of these codes for all true events and the accuracy of coding algorithms for prevalent disease status at baseline are largely unknown. Methods: We randomly sampled 180 Veterans from the VA Puget Sound Health Care System who initiated diabetes treatment. The full electronic medical record was reviewed to identify prevalent conditions at baseline and acute MI events during follow-up. The accuracy of various coding algorithms was assessed. Results: Algorithms for previous acute events at baseline had high PPV (previous MI: 97%; previous stroke: 81%) but low sensitivity (previous MI: 38%; previous stroke: 52%). Algorithms for chronic conditions at baseline had high PPV (heart failure: 72%; coronary heart disease [CHD]: 85%) and high sensitivity (heart failure: 90%, CHD: 84%). For current smoking status at baseline, ICD-9 codes with pharmacy data had a PPV of 77% and sensitivity of 73%. The coding algorithm for acute MI events during follow-up had high PPV (80%) and sensitivity (89%) CONCLUSIONS: ICD-9 codes for acute MI events during follow-up had high PPV and sensitivity. The sensitivity of ICD-9 codes for previous acute events at baseline was low, but a composite variable for baseline CHD had good accuracy. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3921
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: The aim of the study is to describe preapproval safety concerns expressed by the European Medicines Agency (EMA) following regulatory review and to compare those concerns with subsequent issuance of postmarket safety communications. Methods: All novel medicines approved by the EMA through the centralized authorization procedure from 2001 to 2010 were included. Preapproval safety concerns were identified through examination of information related to regulatory review publicly available on the EMA's website. Relevant postmarket safety events were identified through Dear Healthcare Professional Communications (DHPCs), including those resulting in a withdrawal, issued by at least one of four leading national regulators of the European Union. Results: Among the 184 novel medicines included, the EMA had expressed at least one preapproval safety concern for 110 (59.8%) of them. Then, at least one safety communication was issued for 53 (28.8%) medicines within the postmarket period of study, totaling 90 DHPCs and 5 withdrawals. Overall, these 95 DHPCs and withdrawals were pertaining to 66 different clinical safety events. The EMA had expressed a preapproval concern consistent with the postmarket safety event for 22.7% (15 of 66). The rate of issuance of a postmarket safety communication was not statistically different between medicines with or without any preapproval safety concern (31.8% vs. 24%, p = 0.25). Conclusions: Preapproval safety concerns are frequently expressed by the EMA following regulatory review. However, when comparing postmarket safety communications with prior concerns, anticipation was low. Our findings emphasize the need to systematically conduct postmarket studies dedicated to safety evaluation. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3910
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Triptan medications are serotonin agonists used to treat migraine, a chronic pain condition highly prevalent in women of reproductive age. Data on the safety of triptans during pregnancy are scant. We sought to quantify the association of prenatal triptan exposure on neurodevelopment in 3-year-old children. Methods: Using data from the Norwegian Mother and Child Cohort Study, we used propensity score matching to examine associations between prenatal triptan exposure and psychomotor function, communication, and temperament. We used an external validation study to perform propensity calibration to adjust effect estimates for confounders unmeasured in the main study (migraine severity, type, and maternal attitudes towards medication use). Results: We identified 4204 women who reported migraine headache at baseline, of which 375 (8.9%) reported using a triptan greater than or equal to once during pregnancy. Children with prenatal triptan exposure had 1.37-fold greater unadjusted odds of fine motor problems (95% confidence interval (CI): 1.06-1.77), which decreased after propensity score matching (odds ratio (OR): 1.29, 95%CI 0.97-1.73) and was further attenuated after calibration (OR: 1.25, 95%CI 0.89-1.74). We observed no increased risk for gross motor or communication problems, and no differences in temperament. Adjustment for migraine severity using propensity score calibration had a moderate impact on effect estimates, with percent changes ranging from 2.4% to 50%. Conclusions: Prenatal triptan exposure was not associated with psychomotor function, communication problems, or temperament in 3-year-old children. Adjustment for migraine severity reduced effect estimates and should be considered in future studies of the safety of triptans during pregnancy. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3902
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: The use of oral glucose-lowering therapies with insulin is common, but the cardiovascular effects are largely unknown. Among users of long-acting insulin, we conducted a population-based case-control study to evaluate the incident myocardial infarction (MI) and incident stroke risks associated with the use of sulfonylureas and the use of metformin. Methods: Cases were Group Health Cooperative enrollees with type 2 diabetes who used long-acting insulin at the time of diagnosis with a first MI (n = 413) or first stroke (n = 247) from 1995 to 2010. Controls (n = 443) with type 2 diabetes who used long-acting insulin were matched to cases on age, sex, and calendar year. Sulfonylurea and metformin use was classified as current, past, or never using electronic pharmacy records. MI and stroke diagnoses were validated by medical record review. Analyses were adjusted for potential confounders. Results: Current use of sulfonylureas compared with never use was associated with a higher risk of MI (odds ratio [OR] 1.67; 95% confidence interval [CI], 1.10-2.55) but not stroke (OR 1.22; 95%CI, 0.74-2.00). Current use of metformin compared with never use was associated with a lower risk of stroke (OR 0.54; 95%CI, 0.31-0.95) but not MI (OR 0.77; 95%CI, 0.44-1.33). Past use of sulfonylureas and past use of metformin were not associated with either outcome. Conclusions: Sulfonylureas in combination with long-acting insulin may increase the risk of MI compared with the use of insulin alone. Metformin may be an important cardiovascular disease prevention therapy for patients on insulin therapy. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3914
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To compare discrepancies between in-home interviews and electronic medical records (EMRs) on regularly used prescription drugs among older home care clients. Methods: The participants were home care clients aged 75 years or older living in three Finnish municipalities. In-home interview data on regular prescription drug use from 276 home care clients were compared with EMRs. Agreement between the in-home interview data and EMRs was assessed using Cohen's kappa. Results: A majority (83%, n = 229) of the home care clients had discrepancies between in-home interview data and EMRs, and 40% had discrepancies that could clinically compromise their treatment. Living with a spouse or other family member, use of private health care services, diagnosed asthma/COPD or excessive polypharmacy was associated with having discrepancies. Discrepancies were more common among clients with better functioning and ability to self-manage drug use. Agreement between in-home interview data and EMRs was very good or good for other drug groups, but moderate for opioids, paracetamol, benzodiazepines and benzodiazepine-related drugs and lubricant eye drops, and poor for selective beta-2-adrenoceptor agonists. The most common clinically important discrepancies were psychotropics, opioids and agents acting on the renin-angiotensin system and beta-blocking agents. Conclusions: Eight out of ten home care clients had discrepancies between in-home interview data and EMRs. Of these discrepancies, 40% were clinically important. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3909
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: When making decisions about medication use in pregnancy, women consult many information sources, including the Internet. The aim of this study was to assess the content of publicly accessible YouTube videos that discuss medication use in pregnancy. Methods: Using 2023 distinct combinations of search terms related to medications and pregnancy, we extracted metadata from YouTube videos using a YouTube video Application Programming Interface. Relevant videos were defined as those with a medication search term and a pregnancy-related search term in either the video title or description. We viewed relevant videos and abstracted content from each video into a database. We documented whether videos implied each medication to be "safe" or "unsafe" in pregnancy and compared that assessment with the medication's Teratogen Information System (TERIS) rating. Results: After viewing 651 videos, 314 videos with information about medication use in pregnancy were available for the final analyses. The majority of videos were from law firms (67%), television segments (10%), or physicians (8%). Selective serotonin reuptake inhibitors (SSRIs) were the most common medication class named (225 videos, 72%), and 88% of videos about SSRIs indicated that they were unsafe for use in pregnancy. However, the TERIS ratings for medication products in this class range from "unlikely" to "minimal" teratogenic risk. Conclusion: For the majority of medications, current YouTube video content does not adequately reflect what is known about the safety of their use in pregnancy and should be interpreted cautiously. However, YouTube could serve as a platform for communicating evidence-based medication safety information. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3911
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: The aim of this study was to describe characteristics and external validity of the German Health Risk Institute (HRI) Database. Methods: The HRI Database is an anonymized healthcare database with longitudinal data from approximately six Mio Germans. In addition to demographic information (gender, age, region of residence), data on persistence of insurants over time, hospitalization rates, mortality rates and drug prescription rates were extracted from the HRI database for 2013. Corresponding national reference data were obtained from official sources. Results: The proportion of men and women was similar in the HRI Database and Germany, but the database population was slightly younger (mean 40.4 vs 43.7 years). The proportion of insurants living in the eastern part of Germany was lower in the HRI Database (10.1% vs 19.7%). There was good accordance to German reference data with respect to hospitalization rates, overall mortality rate and prescription rates for the 20 most often reimbursed drug classes, with the overall burden of morbidity being slightly lower in the HRI database. From insurants insured on 1 January 2009 (N = 6.2 Mio), a total of 70.6% survived and remained continuously insured with the same statutory health insurance until 31 December 2013. This proportion increased to 77.5% if only insurants ≥40 years were considered. Conclusions: There was good overall accordance of the HRI database and the German population in terms of measures of morbidity, mortality and drug usage. Persistence of insurants with the database over time was high, indicating suitability of the data source for longitudinal epidemiological analyses. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3895
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Distributed research networks (DRNs) afford statistical power by integrating observational data from multiple partners for retrospective studies. However, laboratory test results across care sites are derived using different assays from varying patient populations, making it difficult to simply combine data for analysis. Additionally, existing normalization methods are not suitable for retrospective studies. We normalized laboratory results from different data sources by adjusting for heterogeneous clinico-epidemiologic characteristics of the data and called this the subgroup-adjusted normalization (SAN) method. Methods: Subgroup-adjusted normalization renders the means and standard deviations of distributions identical under population structure-adjusted conditions. To evaluate its performance, we compared SAN with existing methods for simulated and real datasets consisting of blood urea nitrogen, serum creatinine, hematocrit, hemoglobin, serum potassium, and total bilirubin. Various clinico-epidemiologic characteristics can be applied together in SAN. For simplicity of comparison, age and gender were used to adjust population heterogeneity in this study. Results: In simulations, SAN had the lowest standardized difference in means (SDM) and Kolmogorov-Smirnov values for all tests (p < 0.05). In a real dataset, SAN had the lowest SDM and Kolmogorov-Smirnov values for blood urea nitrogen, hematocrit, hemoglobin, and serum potassium, and the lowest SDM for serum creatinine (p < 0.05). Conclusion: Subgroup-adjusted normalization performed better than normalization using other methods. The SAN method is applicable in a DRN environment and should facilitate analysis of data integrated across DRN partners for retrospective observational studies. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3893
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Patients with psoriasis and/or psoriatic arthritis (PsA) are known to have increased cardiovascular morbidity and mortality. Hypertension, an important risk factor for cardiovascular disease, is highly prevalent in patients with psoriasis and/or PsA. The effects of anti-psoriatic medications - including cyclosporine, nonsteroidal anti-inflammatory drugs, and glucocorticoids - on hypertension remain unclear. We examined whether such medication exposure was associated with hypertension in psoriasis patients. Methods: This population-based, nested case-control study analyzed data from an inception psoriasis cohort identified from Taiwan's National Health Insurance Research Database, 2000-2010. A total of 1530 patients with newly diagnosed hypertension and 4542 age- and gender-matched controls were included in the analysis. Conditional logistic regressions were applied to estimate the effects of drug of interest on hypertension. Results: After adjusting for potential confounders, patients with current use of cyclosporine [odds ratio (OR) = 7.13; 95% confidence interval (CI) 1.85-27.49], nonsteroidal anti-inflammatory drugs (OR = 2.2; 95% CI 1.95-2.49), or systemic glucocorticoids (OR = 1.42; 95% CI 1.23-1.64) showed an increased risk of hypertension as compared to those not exposed to these drugs. Moreover, an increasing dose or combined use of nonsteroidal anti-inflammatory drugs and glucocorticoids was associated with increased hypertension risk. The risk of hypertension associated with glucocorticoids, or combined use was greatest among patients aged 49 years or less. Conclusions: The use of cyclosporine, nonsteroidal anti-inflammatory drugs, or glucocorticoid was associated with hypertension in patients with psoriasis and/or PsA. These study results inform physicians on the importance of early identification of hypertension during therapy with such medication. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 11/2015; DOI:10.1002/pds.3890
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: Macrolides have been linked to the occurrence of congenital heart defects, but findings are inconsistent. We therefore aimed to estimate the risk of major congenital malformations (MCMs) after fetal exposure to macrolides, focusing on cardiac malformations. Methods: From the Quebec Pregnancy Cohort (1998-2008), women exposed to a macrolide or penicillin in the first trimester and unexposed women were studied. There were 135 859 pregnancies included; 914 were exposed to azithromycin, 734 to erythromycin, 686 to clarithromycin, and 9106 to penicillin during the first trimester. Cases of MCMs were identified within the first year of life. Results: After adjusting for potential confounders, azithromycin (RR = 1.19, 95%CI: 0.98, 1.44; 120 exposed cases), erythromycin (RR = 0.96, 95%CI: 0.74, 1.24; 66 exposed cases) and clarithromycin use (RR = 1.12, 95%CI: 0.99, 1.42; 79 exposed cases) during the first trimester of pregnancy were not statistically significantly associated with the risk of MCMs; no associations were observed for cardiac malformations. Conclusions: First trimester exposure to any of the macrolides was not associated with an increased risk of overall MCMs or cardiac malformations specifically. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 10/2015; DOI:10.1002/pds.3900
  • [Show abstract] [Hide abstract]
    ABSTRACT: Purpose: To assess antipsychotic prescribing patterns according to insurance coverage type and physician specialty in the outpatient treatment of behavioral disorders (BD) in US youth. Methods: We used 2003-2010 National Ambulatory Medical Care Survey and National Hospital Ambulatory Medical Care Survey data to compare antipsychotic prescribing in the outpatient treatment of BD in youth (6-19 years) according to insurance coverage (public vs. private) and physician specialty (psychiatrist vs. non-psychiatrist) using population-weighted Chi-square and multivariable analyses. Also, we examined co-prescribing of antipsychotics with other psychotropic medication classes. Subgroup analyses were conducted in BD visits with no other clinician-reported psychiatric diagnosis (non-comorbid BD visits). Results: A large majority (71.0%) of BD visits were provided by non-psychiatrists. However, psychiatrists prescribed antipsychotics far more frequently than non-psychiatrists (24.2% vs. 4.6%; adjusted odds ratio (AOR) = 5.1 [95% confidence interval (CI), 2.8-9.2]) in total BD visits as well as in non-comorbid BD visits (18.6% vs. 3.6%; AOR = 5.8 [95% CI, 3.2-10.5]). Antipsychotic prescribing was nearly two-fold greater in visits by publicly insured 6-12 year olds (11.3% vs. 5.8%; AOR = 1.9 [95% CI, 1.1-3.5]) and 13-19 year olds (16.2% vs. 8.9%; AOR = 2.0 [95% CI, 1.1-3.6]) compared with their privately insured counterparts. In more than one-third of antipsychotic-prescribed BD visits, antipsychotics were prescribed concomitantly with ≥2 psychotropic medication classes regardless of age group, insurance coverage, or even in the absence of psychiatric comorbidities. Conclusion: In outpatient visits by youth for BD, antipsychotics were primarily prescribed by psychiatrists, concomitantly, and for the publicly insured. These treatment patterns merit further investigation. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 10/2015; DOI:10.1002/pds.3897

  • Pharmacoepidemiology and Drug Safety 10/2015; DOI:10.1002/pds.3901
  • [Show abstract] [Hide abstract]
    ABSTRACT: Background: A growing body of literature has been produced on the potential role of statins in reducing perioperative cardiac events in patients undergoing non-cardiac surgery. However, evidence remains inconsistent, and little is known about the patterns of perioperative statin use in routine care. Objectives: The objective of this study was to examine patterns of perioperative statin initiation among adults undergoing non-cardiac elective surgery in the USA. Methods: Using data from a large US healthcare insurer, we identified patients aged ≥18 years who underwent moderate-risk to high-risk non-cardiac elective surgery between 2003 and 2012 and initiated statins within 30 days before surgery. We evaluated temporal trends of statin initiation and patient characteristics. In a matched analysis, we assessed the effect of temporal proximity to surgery on the likelihood of statin initiation. Results: Of 460,154 patients undergoing surgery, 5628 (12 per 1000 patients) initiated a statin within 30 days before surgery. Statin initiation increased from 8 per 1000 patients in 2003 to 15 in 2012 (p = 0.0022). The increase was more pronounced among patients undergoing vascular surgery (149 initiators per 1000 patients by the end of 2012) and with Revised Cardiac Risk Index (RCRI) score ≥2 (72 per 1000 patients). Proximity to surgery, in particular vascular surgery, was predictive of statin initiation. Conclusions: Despite the lack of robust evidence, perioperative statin initiation progressively increased from 2003 to 2012, particularly among patients undergoing major vascular surgery and with higher RCRI score. These trends were largely attributable to the initiation of statins in anticipation of non-cardiac surgery rather than routine dyslipidemia treatment. Copyright © 2015 John Wiley & Sons, Ltd.
    Pharmacoepidemiology and Drug Safety 10/2015; DOI:10.1002/pds.3892