Joshua J. Gagne

A combined comorbidity score predicted mortality in elderly patients better than existing scores

OBJECTIVE To develop and validate a single numerical comorbidity score for predicting short- and long-term mortality, by combining conditions in the Charlson and Elixhauser measures. STUDY DESIGN AND SETTING In a cohort of 120,679 Pennsylvania Medicare enrollees with drug coverage through a pharmacy assistance program, we developed a single numerical comorbidity score for predicting 1-year mortality, by combining the conditions in the Charlson and Elixhauser measures. We externally validated the combined score in a cohort of New Jersey Medicare enrollees, by comparing its performance to that of both component scores in predicting 1-year mortality, as well as 180-, 90-, and 30-day mortality. RESULTS C-statistics from logistic regression models including the combined score were higher than corresponding c-statistics from models including either the Romano implementation of the Charlson Index or the single numerical version of the Elixhauser system; c-statistics were 0.860 (95% confidence interval [CI]: 0.854, 0.866), 0.839 (95% CI: 0.836, 0.849), and 0.836 (95% CI: 0.834, 0.847), respectively, for the 30-day mortality outcome. The combined comorbidity score also yielded positive values for two recently proposed measures of reclassification. CONCLUSION In similar populations and data settings, the combined score may offer improvements in comorbidity summarization over existing scores.

Anti-inflammatory drugs and risk of Parkinson disease: a meta-analysis.

Background/Objective: Anti-inflammatory drugs may prevent Parkinson disease (PD) by inhibiting a putative underlying neuroinflammatory process. We tested the hypothesis that anti-inflammatory drugs reduce PD incidence and that there are differential effects by type of anti-inflammatory, duration of use, or intensity of use. Methods: MEDLINE and EMBASE were searched for studies that reported risk of PD associated with anti-inflammatory medications. Random-effects meta-analyses were used to pool results across studies for each type of anti-inflammatory drug. Stratified meta-analyses were used to assess duration- and intensity-response. Results: Seven studies were identified that met the inclusion criteria, all of which reported associations between nonaspirin nonsteroidal anti-inflammatory drugs (NSAIDs) and PD, 6 of which reported on aspirin, and 2 of which reported on acetaminophen. Overall, a 15% reduction in PD incidence was observed among users of nonaspirin NSAIDS (relative risk [RR] 0.85, 95% confidence interval [CI] 0.77–0.94), with a similar effect observed for ibuprofen use. The protective effect of nonaspirin NSAIDs was more pronounced among regular users (RR 0.71, 95% CI 0.58–0.89) and long-term users (RR 0.79, 95% CI 0.59–1.07). No protective effect was observed for aspirin (RR 1.08, 95% CI 0.92–1.27) or acetaminophen (RR 1.06, 95% CI 0.87–1.30). Sensitivity analyses found results to be robust. Conclusions: There may be a protective effect of nonaspirin nonsteroidal anti-inflammatory drug use on risk of Parkinson disease (PD) consistent with a possible neuroinflammatory pathway in PD pathogenesis.

Seizure Outcomes Following the Use of Generic versus Brand-Name Antiepileptic Drugs A Systematic Review and Meta-Analysis

AbstractBackground: The automatic substitution of bioequivalent generics for brand-name antiepileptic drugs (AEDs) has been linked by anecdotal reports to loss of seizure control. Objective: To evaluate studies comparing brand-name and generic AEDs, and determine whether evidence exists of superiority of the brand-name version in maintaining seizure control. Data Sources: English-language human studies identified in searches of MEDLINE, EMBASE and International Pharmaceutical Abstracts (1984 to 2009). Study Selection: Randomized controlled trials (RCTs) and observational studies comparing seizure events or seizure-related outcomes between one brand-name AED and at least one alternative version produced by a distinct manufacturer. Data Extraction: We identified 16 articles (9 RCTs, 1 prospective non-randomized trial, 6 observational studies). We assessed characteristics of the studies and, for RCTs, extracted counts for patients whose seizures were characterized as ‘controlled’ and ‘uncontrolled’. Data Synthesis: Seven RCTs were included in the meta-analysis. The aggregate odds ratio (n = 204) was 1.1 (95% CI 0.9, 1.2), indicating no difference in the odds of uncontrolled seizure for patients on generic medications compared with patients on brand-name medications. In contrast, the observational studies identified trends in drug or health services utilization that the authors attributed to changes in seizure control. Conclusions: Although most RCTs were short-term evaluations, the available evidence does not suggest an association between loss of seizure control and generic substitution of at least three types of AEDs. The observational study data may be explained by factors such as undue concern from patients or physicians about the effectiveness of generic AEDs after a recent switch. In the absence of better data, physicians may want to consider more intensive monitoring of high-risk patients taking AEDs when any switch occurs.

The association between blood pressure and incident Alzheimer disease: a systematic review and meta-analysis.

Background: Many epidemiologic studies have considered the association between blood pressure (BP) and Alzheimer disease, yet the relationship remains poorly understood. Methods: In parallel with work on the AlzRisk online database (www.alzrisk.org), we conducted a systematic review to identify all epidemiologic studies meeting prespecified criteria reporting on the association between hypertension, systolic BP, or diastolic BP and incident Alzheimer disease. When possible, we computed summary measures using random-effects models and explored potential heterogeneity related to age at BP assessment. Results: Eighteen studies reporting on 19 populations met the eligibility criteria. We computed summary relative risks (RR&Sgr;) for 3 measures of BP: hypertension (RR&Sgr; = 0.97 [95% confidence interval = 0.80–1.16]); a 10-mm Hg increase in systolic BP (RR&Sgr; = 0.95 [0.91–1.00]); and a 10-mm Hg increase in diastolic BP (RR&Sgr; = 0.94 [0.85–1.04]). We were unable to compute summary estimates for the association between categories of systolic or diastolic BP and Alzheimer disease; however, there did not appear to be a consistent pattern across studies. After stratifying on age at BP assessment, we found a suggestion of an inverse association between late-life hypertension and Alzheimer disease and a suggestion of an adverse association between midlife diastolic hypertension and Alzheimer disease. Conclusions: Based on existing epidemiologic research, we cannot determine whether there is a causal association between BP and Alzheimer disease. Selection bias and reverse causation may account for the suggested inverse association between late-life hypertension on Alzheimer disease, but, given the expected direction of these biases, they are less likely to account for the suggestion that midlife hypertension increases risk. We advocate continuing systematic review; the AlzRisk database entry on this topic (www.alzrisk.org), which was completed in parallel with this work, will be updated as new studies are published.

Assessing the comparative effectiveness of newly marketed medications: methodological challenges and implications for drug development.

Comparative‐effectiveness research (CER) aims to produce actionable evidence regarding the effectiveness and safety of medical products and interventions as they are used outside of controlled research settings. Although CER evidence regarding medications is particularly needed shortly after market approval, key methodological challenges include (i) potential bias due to channeling of patients to the newly marketed medication because of various patient‐, physician‐, and system‐related factors; (ii) rapid changes in the characteristics of the user population during the early phase of marketing; and (iii) lack of timely data and the often small number of users in the first few months of marketing. We propose a mix of approaches to generate comparative‐effectiveness data in the early marketing period, including sequential cohort monitoring with secondary health‐care data and propensity score (PS) balancing, as well as extended follow‐up of phase III and phase IV trials, indirect comparisons of placebo‐controlled trials, and modeling and simulation of virtual trials.

Effects of Adjusting for Instrumental Variables on Bias and Precision of Effect Estimates

Recent theoretical studies have shown that conditioning on an instrumental variable (IV), a variable that is associated with exposure but not associated with outcome except through exposure, can increase both bias and variance of exposure effect estimates. Although these findings have obvious implications in cases of known IVs, their meaning remains unclear in the more common scenario where investigators are uncertain whether a measured covariate meets the criteria for an IV or rather a confounder. The authors present results from two simulation studies designed to provide insight into the problem of conditioning on potential IVs in routine epidemiologic practice. The simulations explored the effects of conditioning on IVs, near-IVs (predictors of exposure that are weakly associated with outcome), and confounders on the bias and variance of a binary exposure effect estimate. The results indicate that effect estimates which are conditional on a perfect IV or near-IV may have larger bias and variance than the unconditional estimate. However, in most scenarios considered, the increases in error due to conditioning were small compared with the total estimation error. In these cases, minimizing unmeasured confounding should be the priority when selecting variables for adjustment, even at the risk of conditioning on IVs.

Comparative Efficacy and Safety of New Oral Anticoagulants in Patients With Atrial Fibrillation

Background— Dabigatran, an oral thrombin inhibitor, and rivaroxaban and apixaban, oral factor Xa inhibitors, have been found to be safe and effective in reducing stroke risk in patients with atrial fibrillation. We sought to compare the efficacy and safety of the 3 new agents based on data from their published warfarin-controlled randomized trials, using the method of adjusted indirect comparisons. Methods and Results— We included findings from 44 535 patients enrolled in 3 trials of the efficacy of dabigatran (Randomized Evaluation of Long-Term Anticoagulation Therapy [RELY]), apixaban (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation [ARISTOTLE]), and rivaroxaban (Rivaroxaban Once Daily Oral Direct Factor Xa Inhibition Compared With Vitamin K Antagonism for Prevention of Stroke and Embolism Trial in Atrial Fibrillation [ROCKET-AF]), each compared with warfarin. The primary efficacy end point was stroke or systemic embolism; the safety end point we studied was major hemorrhage. To address a lack of comparability between trial populations caused by the restriction of ROCKET-AF to high-risk patients, we conducted a subgroup analysis in patients with a CHADS2 score ≥3. We found no statistically significant efficacy differences among the 3 drugs, although apixaban and dabigatran were numerically superior to rivaroxaban. Apixaban produced significantly fewer major hemorrhages than dabigatran and rivaroxaban. Conclusions— An indirect comparison of new anticoagulants based on existing trial data indicates that in patients with a CHADS2 score ≥3 dabigatran 150 mg, apixaban 5 mg, and rivaroxaban 20 mg resulted in statistically similar rates of stroke and systemic embolism, but apixaban had a lower risk of major hemorrhage compared with dabigatran and rivaroxaban. Until head-to-head trials or large-scale observational studies that reflect routine use of these agents are available, such adjusted indirect comparisons based on trial data are one tool to guide initial therapeutic choices.

Prescription drug use during pregnancy: a population-based study in Regione Emilia-Romagna, Italy

PurposeDrug utilization studies in pregnant women are crucial to inform pharmacovigilance efforts in human teratogenicity. The purpose of this study was to estimate the prevalence of prescription drug use among pregnant women in Regione Emilia-Romagna (RER), Italy.MethodsWe conducted a retrospective prevalence study using data from the RER health care database. Outpatient prescription drug data were reconciled for RER residents who delivered a baby in a hospital between January 1, 2004 and December 31, 2004. Drug data were stratified by trimester of use, pregnancy risk categorization, and anatomical classification.ResultsAmong the 33,343 deliveries identified in 2004, 70% of women were exposed to at least one prescription medication during pregnancy and 48% were exposed to at least one prescription medication after excluding vitamin and mineral products. Many of the most commonly used medications were anti-infectives, such as amoxicillin, fosfomycin, and ampicillin. Nearly 1% of women were exposed to drugs contraindicated (i.e., category X) in pregnancy, including 189 women (0.6%) who received these drugs during the first trimester. Several statin medications were among the most common contraindicated drug exposures.ConclusionA large proportion of women who gave birth in RER in 2004 were exposed to prescription medications. Approximately 1 in 100 women were exposed to contraindicated drugs. The most commonly identified drug exposures can help focus pharmacoepidemiologic efforts in drug-induced birth defects.

Comparative Effectiveness of Generic and Brand-Name Statins on Patient Outcomes: A Cohort Study

Context Some patients do not adhere to their prescribed statins and thus do not fully benefit from the decreases in blood lipid levels that these medications provide. Contribution This study found that, compared with those who initiated a brand-name statin, patients who initiated a generic statin had better adherence and fewer occurrences of a composite outcome that included death from any cause plus hospitalization for an acute coronary syndrome or stroke. Caution All patients were Medicare beneficiaries aged 65 years or older with prescription drug coverage. Implication The lower cost of generic statins allowed patients to adhere to the medication better. The Editors Statins are the most frequently prescribed drugs in the United States (1) and are effective in reducing low-density lipoprotein (LDL) cholesterol levels and cardiovascular events (24). Randomized, controlled trials have found that statins reduce the relative risk for major vascular events by 21% for each 1.0-mmol/L (39-mg/dL) reduction in LDL cholesterol level in patients at low risk for vascular disease (3). Patients assigned to statin therapy in trials tended to achieve reductions in LDL cholesterol level of 1.8 mmol/L (70 mg/dL) with doses used regularly in practice (2). However, a large body of evidence suggests that, in routine practice, patients do not fully adhere to statins and therefore may not receive their full benefit (5, 6). Approximately half of patients in ambulatory care settings discontinue statin therapy within 1 year of initiation (68). Medication nonadherence is a complex multifactorial process (9). Among its many determinants, drug cost may be one of the most easily modifiable (10). Reducing patient spending for prescription drugs can improve adherence (11, 12) and, in some cases, clinical outcomes (13). Generic drugs have been shown in small, short, randomized trials(14, 15) to be clinically equivalent to their brand-name counterparts, as required for approval by the U.S. Food and Drug Administration (FDA). They are usually less expensive than brand-name products and have been associated with better adherence (12). However, no study has investigated whether use of generic versus brand-name statins also leads to improved health outcomes (16). We sought to determine whether patients in a large cohort of Medicare beneficiaries were more adherent to therapy after initiating a generic statin versus a brand-name statin and whether this resulted in differences in health outcomes. Methods The study was designed by the authors and approved by the Institutional Review Board at Brigham and Womens Hospital. Study Cohort The study cohort comprised Medicare beneficiaries (aged 65 years) who had prescription drug coverage through either a stand-alone Medicare Part D plan or a retiree drug plan administered by CVS Caremark, a large national pharmacy benefits manager. For each patient, we linked claims for filled prescriptions to diagnostic, health care utilization, and demographic data from Medicare Parts A and B files and enrollment files. The cohort included patients who initiated a statin (lovastatin, pravastatin, or simvastatin) between 2006 and 2008 and had continuous Medicare and CVS Caremark eligibility in the 6 months before initiation. We restricted the cohort to patients initiating these drugs because they were the only statins for which generic versions were available in the United States during the study. Initiation was defined as a new (index) prescription for a study drug with no prescription for any single statin or statin combination product in the preceding 180 days. To maximize the generalizability of this comparative effectiveness study, we did not impose any other exclusion criteria. We classified patients as exposed to a generic or brand-name statin on the basis of the National Drug Code associated with the index prescription claim. We used the FDAs National Drug Code Directory (17) to determine the manufacturer of each drug and the FDAs Approved Drug Products with Therapeutic Equivalence Evaluations publication (18) to determine whether each manufacturers products were approved via a new drug application (brand-name) or abbreviated new drug application (generic). Outcomes and Follow-up The primary outcomes were adherence to the index statin and a composite cardiovascular outcome. Adherence was measured as the proportion of days covered (PDC) by the index statin up to 1 year after the index prescription date. The PDC is calculated by dividing the number of days of medication supplied by the number of days in a given interval (19). For each patient, the denominator interval began on the index date and ended at death, hospitalization, prescription for any other lipid-lowering drug (for example, a different statin or another lipid-lowering agent, such as a fibrate or bile acid sequestrant, although switches between brand-name and generic versions of the index statin were allowed), the end of the study (31 December 2008), or 365 days after the index prescription date, whichever occurred first. The numerator was the sum of the number of days in the interval for which medication was available based on the days supplied by each prescription. The primary clinical outcome comprised hospitalization for an acute coronary syndrome or stroke and all-cause mortality. We also examined each of these outcomes separately. We used a validated claims-based definition for each outcome, with positive predictive values ranging from 86% to 96% (2022). In the primary analysis, we followed patients from the day after index drug initiation until an occurrence of an event of interest, the end of the study (31 December 2008), or 365 days after initiation, whichever came first. Covariates We measured potential confounders in the 180-day baseline period preceding each patients index date. Demographic variables included age, sex, and race. Health service utilization variables included the number of unique drugs dispensed, number of hospitalizations, number of cardiovascular diagnoses, number of days in the hospital, number of physician office visits, and number of physician office visits with cardiovascular diagnoses. In addition to a comorbidity score that captured patients general health status (23), we determined whether patients had health care encounters with diagnoses for specific cardiovascular conditions (such as atrial fibrillation, congestive heart failure, or peripheral vascular disease) and other disorders (such as chronic obstructive pulmonary disease, diabetes, musculoskeletal conditions, and endocrine disease). Furthermore, we determined whether patients initiated statin treatment for primary or secondary prevention, with the latter defined as having been hospitalized for an acute coronary syndrome in the baseline period. We also measured use of preventive services, including screening mammography and vaccinations, to account for healthy-user effects (24, 25) and ascertained proxies of frailty, such as use of supplemental oxygen, to account for the propensity to stop preventive medications in patients who are very ill (26). Finally, we geocoded patients street addresses and linked them to U.S. census data at the block group level, which is the lowest level for which data are publicly available. We identified the unemployment rate and the median household income in each patients census block group as proxies for socioeconomic status (SES). Statistical Analysis We used propensity score matching (27) to mitigate confounding due to different characteristics between the brand-name and generic groups. The propensity score, which was estimated with a logistic regression model, was defined as a patients probability of receiving a generic statin versus a brand-name statin and was conditional on measured baseline covariates. We matched generic and brand-name drug recipients by using a nearest-neighbor algorithm and within calipers of 0.05 units on the propensity score scale in the primary analysis. Because the cohort included many more generic than brand-name drug recipients, we matched each patient in the brand-name group to as many patients as possible in the generic group with similar propensity scores within the specified caliper. We matched brand-name drug recipients only to recipients of the generic version of the same product (for example, brand-name and generic simvastatin) to compare patients who had initiated molecularly identical drugs. This ensured that differences in outcome rates between treatment groups could be attributed to the generic versus brand-name status rather than to differences among the 3 statins. To assess the performance of the propensity score matching process, we evaluated balance in each baseline covariate and overlap in propensity score distributions between treatment groups before and after matching. We used Cox proportional hazards models to estimate hazard ratios (HRs) and 95% CIs. To account for the variable ratio matching, the Cox models were stratified by matching set. We also estimated rate differences. We performed several prespecified secondary, sensitivity, and subgroup analyses to assess the validity of our study assumptions. We altered our primary analysis by shortening (to 90 days) and lengthening (to 720 days) the maximum follow-up time, excluding events that occurred within 30 and 60 days after index drug initiation, and performing 1:1 fixed-ratio matching on the propensity score. Furthermore, outcome event rates were compared between recipients of the generic and brand-name versions of each drug separately. The primary analysis was also repeated separately for primary prevention and secondary prevention patients. We conducted an on treatment analysis in which we censored patients when they discontinued statin therapy, defined as a gap of more than 30 days without filling a statin prescription beyond the number of days supplied by the last prescription or switching to another lipid-lowering

Prevalence and predictors of potential drug–drug interactions in Regione Emilia-Romagna, Italy

Background and objective:  Drug–drug interactions (DDIs) are preventable medication errors associated with potentially serious adverse events and death. Several studies have examined the prevalence of potential DDIs among ambulatory patients in various countries. Limited recent data on the prevalence of potential DDIs in Italy are available in the medical literature. The objective of this study was to estimate the prevalence of clinically important potential DDIs among the approximately 4 million residents of Regione Emilia‐Romagna (RER), Italy, and to examine possible predictors of potential DDI exposure.