There is growing interest in developing causal inference methods for multi-valued treatments with a focus on pairwise average treatment effects. Here we focus on a clinically important, yet less-studied estimand: causal drug-drug interactions (DDIs), which quantifies the degree to which the causal effect of drug A is altered by the presence versus the absence of drug B. Confounding adjustment when studying the effects of DDIs can be accomplished via inverse probability of treatment weighting (IPTW), a standard approach originally developed for binary treatments and later generalized to multi-valued treatments. However, this approach generally results in biased results when the propensity score model is misspecified. Motivated by the need for more robust techniques, we propose two empirical likelihood-based weighting approaches that allow for specifying a set of propensity score models, with the second method balancing user-specified covariates directly, by incorporating additional, nonparametric constraints. The resulting estimators from both methods are consistent when the postulated set of propensity score models contains a correct one; this property has been termed multiple robustness. In this paper, we derive two multiply-robust estimators of the causal DDI, and develop inference procedures. We then evaluate their finite sample performance through simulation. The results demonstrate that the proposed estimators outperform the standard IPTW method in terms of both robustness and efficiency. Finally, we apply the proposed methods to evaluate the impact of renin-angiotensin system inhibitors (RAS-I) on the comparative nephrotoxicity of nonsteroidal anti-inflammatory drugs (NSAID) and opioids, using data derived from electronic medical records from a large multi-hospital health system.

<p><strong>OBJECTIVE: </strong>To determine antibiotic utilization for NICU infants, as compared to non-NICU infants, in the first 3 years after birth hospital discharge.</p>

<p><strong>STUDY DESIGN: </strong>Retrospective observational study using data from Medicaid Analytic Extract including 667 541 newborns discharged from 2007-2011. Associations between NICU admission and antibiotic prescription were assessed using regression models, adjusting for confounders, and stratified by gestational age and birth weight.</p>

<p><strong>RESULTS: </strong>596 999 infants (89.4%) received ≥1 antibiotic, with a median of 4 prescriptions per 3 person-years (IQR 2-8). Prescribed antibiotics and associated indication were similar between groups. Compared to non-NICU infants (N = 586 227), NICU infants (N = 81 314) received more antibiotic prescriptions (adjusted incidence rate ratio 1.08, 95% confidence interval [CI] (1.08,1.08)). Similar results were observed in all NICU subgroups.</p>

<p><strong>CONCLUSIONS: </strong>Antibiotic utilization in early childhood was higher among infants discharged from NICUs compared to non-NICU infants.</p>

<p><strong>BACKGROUND: </strong>Fluoroquinolones, one of the most commonly prescribed antibiotic classes, have been implicated in cases of central nervous system (CNS) and peripheral nervous system (PNS) adverse events, which highlights the need for epidemiologic studies of the neurological safety of fluoroquinolones.</p>

<p><strong>PURPOSE: </strong>To evaluate the safety of fluoroquinolones with regard to risk of diagnosed neurological dysfunction.</p>

<p><strong>METHODS: </strong>We conducted a propensity score-matched inception cohort study using claims data from a commercially insured population (OptumInsight). Our study included adults prescribed an oral fluoroquinolone or comparator antibiotic between January 2004 and September 2015 for acute bacterial sinusitis, acute bacterial exacerbation of chronic bronchitis, uncomplicated urinary tract infection, or acute bronchitis. Our outcomes were CNS dysfunction, and four separate but complementary PNS dysfunction outcomes. Cox proportional hazards models were estimated after matching on propensity scores fitted using the variables age, sex, epilepsy, hereditary peripheral neuropathy, renal dysfunction, diabetes, gabapentinoid use, statin use, isoniazid use, and chemotherapy use.</p>

<p><strong>RESULTS: </strong>Our cohort contained 976,568 individuals exposed to a fluoroquinolone antibiotic matched 1:1 with a comparator. Matching produced balance (standardized mean difference &lt; 0.1) on all variables included in the propensity score. The hazard ratio associated with fluoroquinolone exposure was 1.08 (95% confidence interval 1.05-1.11) for CNS dysfunction, and 1.09 (95% CI 1.07-1.11) for the most commonly occurring PNS dysfunction outcome.</p>

<p><strong>CONCLUSIONS: </strong>Fluoroquinolone antibiotic use was associated with the development of neurological dysfunction versus comparator antibiotic use in the adult population.</p>

<p><strong>BACKGROUND: </strong>There are few contemporary studies of volume-outcome relationships in pediatric oncology. Children with acute lymphoblastic leukemia (ALL) are treated at a wide variety of hospitals. We investigated if inpatient hospital volume influences outcomes. The objective of this study was to evaluate the relationship between inpatient pediatric and pediatric oncology volume and mortality and intensive care resources (ICU care). We hypothesized an inverse relationship between volume and these outcomes.</p>

<p><strong>PATIENTS AND METHODS: </strong>This was a retrospective cohort study. Patients 0 to 18 years of age in the Pediatric Health Information System or Perspective Premier Database from 2009 to 2011 with ALL were included. Exposures were considered as the average inpatient pediatric and pediatric oncology volume. The primary outcome was inpatient mortality; secondary outcome was need for ICU care.</p>

<p><strong>RESULTS: </strong>The included population comprised 3350 patients from 75 hospitals. The inpatient mortality rate was 0.86% (95% confidence interval, 0.58%-1.2%). In the unadjusted analysis, mortality increased as pediatric oncology volume increased from low (0%) to high volume (1.3%) (P&nbsp;= .009). The small number of deaths precluded multivariable analysis of this outcome. Pediatric and pediatric oncology volume was not associated with ICU care when we controlled for potential confounders.</p>

<p><strong>CONCLUSION: </strong>Induction mortality was low. We did not observe an inverse relationship between volume and mortality or ICU care. This suggests that in a modern treatment era, treatment at a low-volume center might not be associated with increased mortality or ICU care in the first portion of therapy. This relationship should be evaluated in other oncology populations with higher mortality rates and with longer-term outcomes.</p>

<p><strong>OBJECTIVE: </strong>The authors sought to determine whether use of methylphenidate in adults is associated with elevated rates of serious cardiovascular events compared with rates in nonusers.</p>

<p><strong>METHOD: </strong>This was a cohort study of new users of methylphenidate based on administrative data from a five-state Medicaid database and a 14-state commercial insurance database. All new methylphenidate users with at least 180 days of prior enrollment were identified. Users were matched on data source, state, sex, and age to as many as four comparison subjects who did not use methylphenidate, amphetamines, or atomoxetine. A total of 43,999 new methylphenidate users were identified and matched to 175,955 nonusers. Events of primary interest were 1) sudden death or ventricular arrhythmia, 2) stroke, 3) myocardial infarction, and 4) a composite endpoint of stroke or myocardial infarction.</p>

<p><strong>RESULTS: </strong>The age-standardized incidence rate per 1,000 person-years of sudden death or ventricular arrhythmia was 2.17 (95% CI=1.63-2.83) in methylphenidate users and 0.98 (95% CI=0.89-1.08) in nonusers, for an adjusted hazard ratio of 1.84 (95% CI=1.33-2.55). Dosage was inversely associated with risk. Adjusted hazard ratios for stroke, myocardial infarction, and the composite endpoint of stroke or myocardial infarction did not differ statistically from 1.</p>

<p><strong>CONCLUSIONS: </strong>Although initiation of methylphenidate was associated with a 1.8-fold increase in risk of sudden death or ventricular arrhythmia, the lack of a dose-response relationship suggests that this association may not be a causal one.</p>

<p>International Classification of Diseases, 9th Revision (ICD-9) code(s) for neuroblastoma do not exist, preventing identification of these patients in administrative databases. To overcome this challenge, a three-step algorithm, using ICD-9 codes, exclusion criteria, and manual review of chemotherapy billing data, was utilized to assemble a high-risk neuroblastoma cohort (n = 952) from the Pediatric Health Information System (PHIS) Database and validated at a single institution [sensitivity 89.1%; positive predictive value (PPV) 96.1%]. This cohort provides a data source for future comparative effectiveness and clinical epidemiology studies in high-risk neuroblastoma patients.</p>

<p><strong>MAIN OBJECTIVE: </strong>To compare the incidence rates of serious cardiovascular events in adult initiators of amphetamines or atomoxetine to rates in non-users.</p>

<p><strong>METHODS: </strong>This was a retrospective cohort study of new amphetamines (n=38,586) or atomoxetine (n=20,995) users. Each medication user was matched to up to four non-users on age, gender, data source, and state (n=238,183). The following events were primary outcomes of interest 1) sudden death or ventricular arrhythmia, 2) stroke, 3) myocardial infarction, 4) a composite endpoint of stroke or myocardial infarction. Cox proportional hazard regression was used to calculate propensity-adjusted hazard ratios for amphetamines versus matched non-users and atomoxetine versus matched non-users, with intracluster dependence within matched sets accounted for using a robust sandwich estimator.</p>

<p><strong>RESULTS: </strong>The propensity-score adjusted hazard ratio for amphetamines use versus non-use was 1.18 (95% CI: 0.55-2.54) for sudden death/ventricular arrhythmia, 0.80 (95% CI: 0.44-1.47) for stroke, 0.75 (95% CI: 0.42-1.35) for myocardial infarction, and 0.78 (95% CI: 0.51-1.19) for stroke/myocardial infarction. The propensity-score adjusted hazard ratio for atomoxetine use versus non-use was 0.41 (95% CI: 0.10-1.75) for sudden death/ventricular arrhythmia, 1.30 (95% CI: 0.52-3.29) for stroke, 0.56 (95% CI: 0.16-2.00) for myocardial infarction, and 0.92 (95% CI: 0.44-1.92) for stroke/myocardial infarction.</p>

<p><strong>CONCLUSIONS: </strong>Initiation of amphetamines or atomoxetine was not associated with an elevated risk of serious cardiovascular events. However, some of the confidence intervals do not exclude modest elevated risks, e.g. for sudden death/ventricular arrhythmia.</p>