Azithromycin and the risk of cardiovascular death.

BACKGROUND: Although several macrolide antibiotics are proarrhythmic and associated with an increased risk of sudden cardiac death, azithromycin is thought to have minimal cardiotoxicity. However, published reports of arrhythmias suggest that azithromycin may increase the risk of cardiovascular death. METHODS: We studied a Tennessee Medicaid cohort designed to detect an increased risk of death related to short-term cardiac effects of medication, excluding patients with serious noncardiovascular illness and person-time during and shortly after hospitalization. The cohort included patients who took azithromycin (347,795 prescriptions), propensity-score-matched persons who took no antibiotics (1,391,180 control periods), and patients who took amoxicillin (1,348,672 prescriptions), ciprofloxacin (264,626 prescriptions), or levofloxacin (193,906 prescriptions). RESULTS: During 5 days of therapy, patients taking azithromycin, as compared with those who took no antibiotics, had an increased risk of cardiovascular death (hazard ratio, 2.88; 95% confidence interval [CI], 1.79 to 4.63; P<0.001) and death from any cause (hazard ratio, 1.85; 95% CI, 1.25 to 2.75; P=0.002). Patients who took amoxicillin had no increase in the risk of death during this period. Relative to amoxicillin, azithromycin was associated with an increased risk of cardiovascular death (hazard ratio, 2.49; 95% CI, 1.38 to 4.50; P=0.002) and death from any cause (hazard ratio, 2.02; 95% CI, 1.24 to 3.30; P=0.005), with an estimated 47 additional cardiovascular deaths per 1 million courses; patients in the highest decile of risk for cardiovascular disease had an estimated 245 additional cardiovascular deaths per 1 million courses. The risk of cardiovascular death was significantly greater with azithromycin than with ciprofloxacin but did not differ significantly from that with levofloxacin. CONCLUSIONS: During 5 days of azithromycin therapy, there was a small absolute increase in cardiovascular deaths, which was most pronounced among patients with a high baseline risk of cardiovascular disease. (Funded by the National Heart, Lung, and Blood Institute and the Agency for Healthcare Quality and Research Centers for Education and Research on Therapeutics.).

ADHD drugs and serious cardiovascular events in children and young adults.

BACKGROUND: Adverse-event reports from North America have raised concern that the use of drugs for attention deficit-hyperactivity disorder (ADHD) increases the risk of serious cardiovascular events. METHODS: We conducted a retrospective cohort study with automated data from four health plans (Tennessee Medicaid, Washington State Medicaid, Kaiser Permanente California, and OptumInsight Epidemiology), with 1,200,438 children and young adults between the ages of 2 and 24 years and 2,579,104 person-years of follow-up, including 373,667 person-years of current use of ADHD drugs. We identified serious cardiovascular events (sudden cardiac death, acute myocardial infarction, and stroke) from health-plan data and vital records, with end points validated by medical-record review. We estimated the relative risk of end points among current users, as compared with nonusers, with hazard ratios from Cox regression models. RESULTS: Cohort members had 81 serious cardiovascular events (3.1 per 100,000 person-years). Current users of ADHD drugs were not at increased risk for serious cardiovascular events (adjusted hazard ratio, 0.75; 95% confidence interval [CI], 0.31 to 1.85). Risk was not increased for any of the individual end points, or for current users as compared with former users (adjusted hazard ratio, 0.70; 95% CI, 0.29 to 1.72). Alternative analyses addressing several study assumptions also showed no significant association between the use of an ADHD drug and the risk of a study end point. CONCLUSIONS: This large study showed no evidence that current use of an ADHD drug was associated with an increased risk of serious cardiovascular events, although the upper limit of the 95% confidence interval indicated that a doubling of the risk could not be ruled out. However, the absolute magnitude of such an increased risk would be low. (Funded by the Agency for Healthcare Research and Quality and the Food and Drug Administration.).

60-Hz frequency effect on gait in Parkinson's disease with subthalamic nucleus deep brain stimulation.

OBJECTIVE: Gait dysfunction is common in advancing Parkinson's disease and has a disappointing response to dopamine replacement and subthalamic nucleus deep brain stimulation programming parameters. Low-frequency stimulation, less than 130 Hz in combination with increased voltage, has been shown to decrease freezing episodes and number of steps with little impact on overall performance measured by the Unified Parkinson's Disease Rating Scale. This was in the setting of delivering the same total energy, which required both a change in voltage and frequency. We wanted to determine if the benefit came from low frequency alone. MATERIALS AND METHODS: We enrolled 20 Parkinson's patients who were at least three months in postbilateral subthalamic deep brain stimulation and reported gait changes. Subjects held their Parkinson's medications overnight, and following a baseline evaluation, they were randomly assigned to both 60 and 130 Hz stimulation in a blinded fashion with all other parameters held constant. Each subject was set at each frequency twice during the study, with a 60-min stimulation interval prior to each gait evaluation. RESULTS: There was no significant difference between the two frequencies, with the primary outcome measure of stride length. Two of the 20 patients reported a significant subjective improvement in their gait with no statistical difference in their outcomes. There also was less tremor control at 60 Hz. CONCLUSION: We were unable to demonstrate improved gait with lower frequency stimulation as suggested by prior studies. This may have been because of the decreased energy delivered from the lower frequency and unchanged voltage.

Adherence to transcranial Doppler screening guidelines among children with sickle cell disease.

BACKGROUND: Little is known about adherence to guidelines recommending yearly screening with transcranial Doppler (TCD) ultrasonography to detect stroke risk for children with severe sickle cell disease. The objective was to determine the proportion of children with hemoglobin SS (HbSS) or sickle-ß(0) -thalassemia (HbSß(0) ) aged 2-16 years who received recommended TCD screening from 1997 to 2008, and to identify factors associated with adherence. PROCEDURE: A retrospective cohort study included patients enrolled in Tennessee Medicaid with HbSS or HbSß(0) who received care at the two largest sickle cell centers in Tennessee. The outcome of interest was adherence with guidelines for annual screening TCD's, identified from computer claims and validated through medical record review. The cumulative rate of children who received a TCD per year was calculated using the Kaplan-Meier method. Cox proportional hazards regression was used to examine the association of child, family, and health care use characteristics with receiving a TCD. RESULTS: Among 338 TCD eligible at-risk children, 232 (68.6%) had at least one TCD during the study period. The yearly cumulative incidence of annual TCD's increased from 2.5% in 1997 to 68.3% in 2008. In multivariate models, calendar year, maternal education, and increased number of sickle cell related outpatient visits were associated with an increased rate of receiving a TCD. CONCLUSIONS: Publicly insured children with HbSS or HbSß(0) had increasing adherence with TCD screening guidelines between 1997 and 2008, though 31% had no TCD at all during follow-up. Increasing number of sickle cell related outpatient visits was associated with increasing adherence to screening guidelines.

Increasing use of atypical antipsychotics and anticonvulsants during pregnancy.

PURPOSE: To quantify maternal use of atypical antipsychotics, typical antipsychotics, anticonvulsants, and lithium during pregnancy. METHODS: Tennessee birth and death records were linked to Tennessee Medicaid data to conduct a retrospective cohort study of 296,817 women enrolled in Tennessee Medicaid throughout pregnancy who had a live birth or fetal death from 1985 to 2005. RESULTS: During the study time period, the adjusted rate of use of any study medication during pregnancy increased from nearly 14 to 31 per 1000 pregnancies (ß = 0.08, 95% CI = 0.07, 0.09). Significant increases were reported in use of anticonvulsants alone among mothers with pain and other psychiatric disorders, atypical antipsychotics alone among mothers with bipolar disorders, schizophrenia, unipolar depressive disorders, and other psychiatric disorders, and more than one studied medication for mothers with epilepsy, pain disorders, bipolar disorders, unipolar depressive disorders, and other psychiatric disorders. Significant decreases were reported in use of lithium alone and typical antipsychotics alone for all clinically meaningful diagnosis groups. CONCLUSIONS: There was a substantial increase in use of atypical antipsychotics alone, anticonvulsants alone, and medications from multiple studied categories among Tennessee Medicaid-insured pregnant women during the study period. Further examination of the maternal and fetal consequences of exposure to these medications during pregnancy is warranted.

The impact of beta-blocker therapy on anemia after traumatic brain injury.

BACKGROUND: An increase in endogenous catecholamine levels after traumatic brain injury (TBI) is well described. Animal studies suggest that postinjury anemia is exacerbated by a persistent hyperadrenergic state. This study aims to determine if beta-blocker (BB) exposure affects anemia after TBI. STUDY DESIGN AND METHODS: We reviewed a Level I trauma registry for patients with TBI, examining markers of anemia between patients who received BB with those who did not. RESULTS: A total of 174 patients were exposed to BB (BB+) and 245 were not exposed (BB-). The mean age in the BB+ group was 50 years (vs. 36 years in BB- group, p < 0.001). The mean injury severity score was 33.6 for the BB+ group (vs. 30.8 for BB- group, p = 0.01). While BB+ patients were more likely to receive a transfusion (60.9% vs. 35.1%, p < 0.001), BB+ patients reached their nadir hemoglobin (Hb) at a later day of hospitalization and their rate of decrease in Hb was significantly slower (both p < 0.001). Choosing Hb cutoffs for anemia of both 7 and 10 g/dL, Kaplan-Meier demonstrated a significant delay in time to anemia. CONCLUSION: This study suggests beta-blockade delays anemia after TBI. Elaboration of this effect may demonstrate an additional benefit of beta-blockade after head injury.

Volume-outcome effects for children undergoing resection of renal malignancies.

INTRODUCTION: Adults undergoing oncologic resections at low-volume centers experience increased perioperative morbidity and mortality. The volume-outcome effect has not been extensively studied in pediatric oncologic resections. METHODS: To clarify volume-outcome effects in pediatric oncologic resections, we analyzed resection of renal malignancies in children less than 15 y of age. We conducted a cross-sectional analysis of hospital discharges included in the health care utilization project kids' inpatient database from 1997 to 2009, examining in-hospital operative complications, length of stay (LOS), and inflation-adjusted hospital charges. Hospital volume was expressed as low (n = 1-2), medium (n = 3-4), and high (n > 4) annual volume of resections. RESULTS: One thousand five hundred thirty-eight patients underwent renal malignancy resection. Of these, 527 patients had resection in low-, 422 in medium-, and 589 in high-volume hospitals. Relative to low-volume hospitals, those resected in medium-volume hospitals had an odds ratio of 0.62 (95% confidence interval 0.39-0.99, P = 0.046) for operative complication and those in high-volume hospitals had an odds ratio of 1.02 (95% confidence interval 0.63-1.65, P = 0.95). There was no detectable association with LOS (P = 0.113) or inflation-adjusted charges (P = 0.331). CONCLUSIONS: The number of complications, total charges, and LOS attributable to resection of a childhood renal malignancy did not differ among high-, medium-, or low-operative volume hospitals, although oncologic outcomes could not be determined because of the limited nature of this administrative database.

Trends in the use of antiepileptic drugs among pregnant women in the US, 2001-2007: a medication exposure in pregnancy risk evaluation program study.

BACKGROUND: Little is known about the extent of antiepileptic drug (AED) use in pregnancy, particularly for newer agents. Our objective was to assess whether AED use has increased among pregnant women in the US, 2001-2007. METHODS: We analysed data from the Medication Exposure in Pregnancy Risk Evaluation Program (MEPREP) database, 1 January 2001 to 31 December 2007. We identified liveborn deliveries among women, aged 15-45 years on delivery date, who were members of MEPREP health plans (n=585615 deliveries). Pregnancy exposure to AEDs, determined through outpatient pharmacy dispensing files. Older AEDs were available for clinical use before 1993; other agents were considered newer AEDs. Information on sociodemographic and medical/reproductive factors was obtained from linked birth certificate files. Maternal diagnoses were identified based on ICD-9 codes. RESULTS: Prevalence of AED use during pregnancy increased between 2001 (15.7 per 1000 deliveries) and 2007 (21.9 per 1000 deliveries), driven primarily by a fivefold increase in the use of newer AEDs. Thirteen per cent of AED-exposed deliveries involved a combination of two or more AEDs. Psychiatric disorders were the most prevalent diagnoses, followed by epileptic and pain disorders, among AED users regardless of AED type, year of conception or gestational period. CONCLUSIONS: AED use during pregnancy increased between 2001 and 2007, driven by a fivefold increase in the use of newer AEDs. Nearly one in eight AED-exposed deliveries involved the concomitant use of more than one AED. Additional investigations of the reproductive safety of newer AEDs may be needed.

ED crowding and the use of nontraditional beds.

BACKGROUND AND OBJECTIVES: In an effort to compensate for crowding, many emergency departments (EDs) evaluate and treat patients in nontraditional settings such as gurneys in hallways and conference rooms. The impact of this practice on ED evaluation time is unknown. RESEARCH DESIGN AND SUBJECTS: A historical cohort of adult ED visits to an academic hospital between August 1, 2009 and August 1, 2010, was used to evaluate the relationship between ED bed assignment (traditional, hallway, or conference room bed) and mean ED evaluation time, defined as the time spent in an ED bed before admission or discharge. Chief complaints were categorized into the 5 most frequent categories: abdominal/genitourinary, joint/muscle, general (fever, malaise), head/neck, and other. Multiple linear regression and marginal prediction were used to calculate the mean ED evaluation times for each bed type, overall, and by chief complaint category. RESULTS: During the study period, 15 073 patient visits met the inclusion criteria. After adjustment for patient and ED factors, assignments to hallway and conference room beds were associated with increases in a mean ED evaluation time of 13.3 minutes (95% confidence interval, 13.2-13.3) and 10.9 minutes (95% confidence interval, 10.8-10.9), respectively, compared with the traditional bed ED evaluation time. This varied by chief complaint category. CONCLUSIONS: Use of nontraditional beds is associated with increases in mean ED evaluation time; however, these increases are small and may be further minimized by restricting the use of nontraditional beds to patients with specific chief complaints. Nontraditional beds may have a role in improving ED throughput during times of crowding.

Changes to the financial responsibility for juvenile court ordered psychiatric evaluations reduce inpatient services utilization: an interrupted time series study.

BACKGROUND: The purpose of the current study was to evaluate the impact of a July 2008 Tennessee Court of Appeals opinion that shifted financial responsibility for juvenile court ordered psychiatric evaluations from the State to the County. METHODS: We used de-identified administrative data from the Tennessee Department of Mental Health and mid-year population estimates from the U.S. Census Bureau from July 1, 2006 to June 30, 2010, and an interrupted time series design with segmented regression analysis to quantify the impact of the implementation of the Court opinion. RESULTS: In the study period, there were 2,176 referrals for juvenile court ordered psychiatric evaluations in Tennessee; of these, 74.1% were inpatient evaluations. The Court opinion was associated with a decrease of 9.4 (95% C.I. = 7.9-10.8) inpatient and increase of 1.2 (95% C.I. = 0.4-2.1) outpatient evaluations per 100,000 Tennessee youth aged 12 to 19 years per month. CONCLUSIONS: The Court opinion that shifted financial responsibility for juvenile court ordered psychiatric evaluations from the State to the County was associated with a sudden and significant decrease in inpatient psychiatric evaluations, and more modest increase in outpatient evaluations.

Performance of disease risk scores, propensity scores, and traditional multivariable outcome regression in the presence of multiple confounders.

Propensity scores are widely used in cohort studies to improve performance of regression models when considering large numbers of covariates. Another type of summary score, the disease risk score (DRS), which estimates disease probability conditional on nonexposure, has also been suggested. However, little is known about how it compares with propensity scores. Monte Carlo simulations were conducted comparing regression models using the DRS and the propensity score with models that directly adjust for all of the individual covariates. The DRS was calculated in 2 ways: from the unexposed population and from the full cohort. Compared with traditional multivariable outcome regression models, all 3 summary scores had comparable performance for moderate correlation between exposure and covariates and, for strong correlation, the full-cohort DRS and propensity score had comparable performance. When traditional methods had model misspecification, propensity scores and the full-cohort DRS had superior performance. All 4 models were affected by the number of events per covariate, with propensity scores and traditional multivariable outcome regression least affected. These data suggest that, for cohort studies for which covariates are not highly correlated with exposure, the DRS, particularly that calculated from the full cohort, is a useful tool.

Positive predictive value of automated database records for diabetic ketoacidosis (DKA) in children and youth exposed to antipsychotic drugs or control medications: a Tennessee Medicaid Study.

BACKGROUND: Diabetic ketoacidosis (DKA) is a potentially life-threatening complication of treatment with some atypical antipsychotic drugs in children and youth. Because drug-associated DKA is rare, large automated health outcomes databases may be a valuable data source for conducting pharmacoepidemiologic studies of DKA associated with exposure to individual antipsychotic drugs. However, no validated computer case definition of DKA exists. We sought to assess the positive predictive value (PPV) of a computer case definition to detect incident cases of DKA, using automated records of Tennessee Medicaid as the data source and medical record confirmation as a "gold standard." METHODS: The computer case definition of DKA was developed from a retrospective cohort study of antipsychotic-related type 2 diabetes mellitus (1996-2007) in Tennessee Medicaid enrollees, aged 6-24 years. Thirty potential cases with any DKA diagnosis (ICD-9 250.1, ICD-10 E1x.1) were identified from inpatient encounter claims. Medical records were reviewed to determine if they met the clinical definition of DKA. RESULTS: Of 30 potential cases, 27 (90%) were successfully abstracted and adjudicated. Of these, 24 cases were confirmed by medical record review (PPV 88.9%, 95% CI 71.9 to 96.1%). Three non-confirmed cases presented acutely with severe hyperglycemia, but had no evidence of acidosis. CONCLUSIONS: Diabetic ketoacidosis in children and youth can be identified in a computerized Medicaid database using our case definition, which could be useful for automated database studies in which drug-associated DKA is the outcome of interest.

ADHD medications and risk of serious cardiovascular events in young and middle-aged adults.

CONTEXT: More than 1.5 million US adults use stimulants and other medications labeled for treatment of attention-deficit/hyperactivity disorder (ADHD). These agents can increase heart rate and blood pressure, raising concerns about their cardiovascular safety. OBJECTIVE: To examine whether current use of medications prescribed primarily to treat ADHD is associated with increased risk of serious cardiovascular events in young and middle-aged adults. DESIGN, SETTING, AND PARTICIPANTS: Retrospective, population-based cohort study using electronic health care records from 4 study sites (OptumInsight Epidemiology, Tennessee Medicaid, Kaiser Permanente California, and the HMO Research Network), starting in 1986 at 1 site and ending in 2005 at all sites, with additional covariate assessment using 2007 survey data. Participants were adults aged 25 through 64 years with dispensed prescriptions for methylphenidate, amphetamine, or atomoxetine at baseline. Each medication user (n = 150,359) was matched to 2 nonusers on study site, birth year, sex, and calendar year (443,198 total users and nonusers). MAIN OUTCOME MEASURES: Serious cardiovascular events, including myocardial infarction (MI), sudden cardiac death (SCD), or stroke, with comparison between current or new users and remote users to account for potential healthy-user bias. RESULTS: During 806,182 person-years of follow-up (median, 1.3 years per person), 1357 cases of MI, 296 cases of SCD, and 575 cases of stroke occurred. There were 107,322 person-years of current use (median, 0.33 years), with a crude incidence per 1000 person-years of 1.34 (95% CI, 1.14-1.57) for MI, 0.30 (95% CI, 0.20-0.42) for SCD, and 0.56 (95% CI, 0.43-0.72) for stroke. The multivariable-adjusted rate ratio (RR) of serious cardiovascular events for current use vs nonuse of ADHD medications was 0.83 (95% CI, 0.72-0.96). Among new users of ADHD medications, the adjusted RR was 0.77 (95% CI, 0.63-0.94). The adjusted RR for current use vs remote use was 1.03 (95% CI, 0.86-1.24); for new use vs remote use, the adjusted RR was 1.02 (95% CI, 0.82-1.28); the upper limit of 1.28 corresponds to an additional 0.19 events per 1000 person-years at ages 25-44 years and 0.77 events per 1000 person-years at ages 45-64 years. CONCLUSIONS: Among young and middle-aged adults, current or new use of ADHD medications, compared with nonuse or remote use, was not associated with an increased risk of serious cardiovascular events. Apparent protective associations likely represent healthy-user bias.

Initiation of rheumatoid arthritis treatments and the risk of serious infections.

OBJECTIVE: In clinical trials of RA patients on traditional DMARDs, the addition of TNF-alpha antagonists increased infections compared with addition of placebo. Our objective was to compare serious infections following initiation of different RA regimens. Prior comparative studies of DMARD initiation have yielded conflicting results. METHODS: We estimated hospitalization rates for infections following initiation of TNF-alpha antagonists, other DMARDs and oral glucocorticoids in Tennessee Medicaid-enrolled RA patients (1995-2005). Exposure time was measured using pharmacy information and infections were identified using validated definitions. Initiation of RA regimens was compared using Cox regression models with MTX as the reference. Sensitivity analyses excluded glucocorticoid users, applied a first exposure carried forward approach, restricted observations to 2002-05 and first episodes of use and explored effects of unmeasured confounders. RESULTS: We identified 28 906 new episodes of medication use, including TNF-alpha antagonists (8%), MTX alone (15%) and glucocorticoids alone (57%). Compared with MTX initiation, TNF-alpha antagonist initiation did not significantly increase the risk of hospitalizations for pneumonia [adjusted hazard ratio (aHR) 1.61; 95% CI 0.85, 3.03] or any infection (aHR 1.31; 95% CI 0.78, 2.19). Initiation of LEF, SSZ or HCQ did not increase serious infections, compared with MTX. Both initiation and concurrent glucocorticoid use were associated with a dose-dependent increase in serious infections. Sensitivity analyses showed consistent results. CONCLUSIONS: Compared with initiation of MTX alone, initiation of TNF-alpha antagonists was not associated with a large increase in the risk of serious infections. Glucocorticoid use was associated with a dose-dependent increase in the risk of these infections.

Black race, sex, and extrapulmonary tuberculosis risk: an observational study.

BACKGROUND: Extrapulmonary tuberculosis is likely a marker of underlying immune compromise. Our objective was to determine race and sex differences in extrapulmonary tuberculosis risk in order to identify the optimal population in which to assess for host factors associated with extrapulmonary tuberculosis. METHODS: We performed an observational study of all tuberculosis cases reported to the Tennessee Department of Health, January 1, 2000 to December 31, 2006. We compared the incidence of extrapulmonary tuberculosis by race and sex. We also examined risk factors associated with extrapulmonary disease among all persons with tuberculosis. RESULTS: Extrapulmonary tuberculosis incidence per 100,000 population was 5.93 in black men, 3.21 in black women, 1.01 in non-black men, and 0.58 in non-black women. Among those with tuberculosis, black women were most likely to have extrapulmonary disease (38.6%), followed by black men (28.1%), non-black women (24.6%) and non-black men (21.1%). In multivariate logistic regression among persons with tuberculosis, black women (OR 1.82 (95% CI 1.24-2.65), p = 0.002), black men (OR 1.54 (95% CI 1.13-2.09, p = 0.006), foreign birth (OR 1.55 (95% CI 1.12-2.14), p = 0.009), and HIV infection (OR 1.45 (95% CI 0.99-2.11), p = 0.06) were associated with extrapulmonary tuberculosis. CONCLUSIONS: Black men and black women had the highest incidence of extrapulmonary tuberculosis, and high odds of extrapulmonary disease among persons with tuberculosis. These data suggest that factors in addition to tuberculosis exposure contribute to extrapulmonary tuberculosis risk in blacks.

Major congenital malformations after first-trimester exposure to ACE inhibitors.

BACKGROUND: Use of angiotensin-converting-enzyme (ACE) inhibitors during the second and third trimesters of pregnancy is contraindicated because of their association with an increased risk of fetopathy. In contrast, first-trimester use of ACE inhibitors has not been linked to adverse fetal outcomes. We conducted a study to assess the association between exposure to ACE inhibitors during the first trimester of pregnancy only and the risk of congenital malformations. METHODS: We studied a cohort of 29,507 infants enrolled in Tennessee Medicaid and born between 1985 and 2000 for whom there was no evidence of maternal diabetes. We identified 209 infants with exposure to ACE inhibitors in the first trimester alone, 202 infants with exposure to other antihypertensive medications in the first trimester alone, and 29,096 infants with no exposure to antihypertensive drugs at any time during gestation. Major congenital malformations were identified from linked vital records and hospitalization claims during the first year of life and confirmed by review of medical records. RESULTS: Infants with only first-trimester exposure to ACE inhibitors had an increased risk of major congenital malformations (risk ratio, 2.71; 95 percent confidence interval, 1.72 to 4.27) as compared with infants who had no exposure to antihypertensive medications. In contrast, fetal exposure to other antihypertensive medications during only the first trimester did not confer an increased risk (risk ratio, 0.66; 95 percent confidence interval, 0.25 to 1.75). Infants exposed to ACE inhibitors were at increased risk for malformations of the cardiovascular system (risk ratio, 3.72; 95 percent confidence interval, 1.89 to 7.30) and the central nervous system (risk ratio, 4.39; 95 percent confidence interval, 1.37 to 14.02). CONCLUSIONS: Exposure to ACE inhibitors during the first trimester cannot be considered safe and should be avoided.

Immortal time bias in critical care research: application of time-varying Cox regression for observational cohort studies.

OBJECTIVES: To examine the bias introduced by using time-fixed methodology to analyze the effects of a time-varying exposure incurred in the intensive care unit. DESIGN: Prospective cohort and Monte Carlo simulation studies. SETTING: Medical and coronary intensive care units in a university hospital. PATIENTS: A total of 224 mechanically ventilated patients. METHODS: Part I was a case study analyzing the association between delirium in the intensive care unit (exposure variable) and outcomes (intensive care unit length of stay and 6-mo mortality) in a prospective cohort study. Part II was a Monte Carlo simulation generating 16,000 data sets wherein the true associations between delirium and outcomes were known before analysis. In both parts, we assessed associations between delirium in the intensive care unit and outcomes (intensive care unit length of stay and mortality), using time-fixed vs. time-varying Cox regression methodology. MEASUREMENTS AND MAIN RESULTS: In the case study, delirium analyzed as a time-fixed variable was associated with a delayed intensive care unit discharge (adjusted hazard ratio = 1.9, 95% confidence interval, 1.3-2.7, p < .001), but no association was noted using a time-varying method (adjusted hazard ratio = 1.1, 95% confidence interval = 0.7-1.6, p = .70). Alternatively, delirium analyzed as a time-fixed variable was not associated with 6-mo mortality (adjusted hazard ratio = 2.9, 95% confidence interval, 0.9-5.0, p = .09), whereas delirium analyzed as a time-varying variable was associated with increased mortality (adjusted hazard ratio = 3.2, 95% confidence interval, 1.4-7.7, p = .008). In the simulation study, time-fixed methods produced erroneous results in 97.1% of the data sets with no true association; time-varying methods produced erroneous results in only 3.7%. Similarly, time-fixed methods produced biased results when a true association was present, whereas time-varying methods produced accurate results. CONCLUSIONS: Studies using a time-fixed analytic approach to understand relationships between exposures and clinical outcomes can result in considerable bias when the variables overlap temporally in occurrence. Those conducting such studies, and clinicians reading them, should ensure that time-varying exposures are correctly analyzed to avoid biased conclusions.

Antibiotics potentially used in response to bioterrorism and the risk of major congenital malformations.

This study was designed to assess the association between pregnancy-related exposures to antibiotics recommended for use in the event of a bioterrorism attack and major congenital malformations. A retrospective cohort study included 30 049 infants from Tennessee Medicaid born between 1985 and 2000 identified from computerised state databases. Infants with fetal exposures to ciprofloxacin, azithromycin, doxycycline and amoxicillin (antibiotics recommended for potential bioterrorism attacks) (n = 24 521) and erythromycin (included as a positive control) (n = 2128) were compared with infants with no fetal exposure to any antibiotics (n = 3400). Major congenital malformations identified from computerised records were confirmed through medical record review. Overall, 869 (2.9%) of infants in the cohort had a confirmed major congenital malformation, with major malformations ranging from 2.5% to 3.0% among the antibiotic-specific exposure groups. No increased risk was present in multivariable analyses for any malformations and for malformations of specific organ systems. In conclusion, these data suggest that ciprofloxacin, azithromycin, doxycycline or amoxicillin use by pregnant women should not result in a greater incidence of overall major congenital malformations in infants whose mothers take these medications, though a large increase in risk cannot be ruled out.

Non-aspirin NSAIDs, cyclooxygenase-2 inhibitors and risk for cardiovascular events-stroke, acute myocardial infarction, and death from coronary heart disease.

PURPOSE: To determine if certain non-steroidal anti-inflammatory drugs (NSAIDs) are associated with increased risk of cardiovascular events: acute myocardial infarction (AMI), stroke, and death from coronary heart disease (CHD). METHODS: We conducted a retrospective cohort study of Tennessee Medicaid enrollees aged 35-94 years between 1 January 1999 and 31 December 2005. Eligible persons were non-institutionalized, had continuous enrollment, and had no serious illness prior to cohort entry. Exposure to celecoxib, rofecoxib, valdecoxib, ibuprofen, naproxen, diclofenac, and indomethacin was studied. The outcome was hospitalization for AMI, stroke, or death from CHD among those with and without a history of cardiovascular disease (CVD). Adjusted hazard ratios (aHR) and 95% confidence intervals (95% CI) are reported. RESULTS: There were 610 001 persons in the final cohort and 14% had a baseline history of CVD. In those without CVD (N = 525 249) there were 1 566 678 person-years of follow-up and 12 184 events. In this group, non-users had 7.90 events/1000 person-years. Events/1000 person-years were 10.41 for current use of celecoxib (aHR 1.00, 95% CI 0.89-1.13), 10.91 for rofecoxib (aHR 1.21, 95% CI 1.07-1.37), 12.46 for valdecoxib (aHR 1.30 95% CI 1.04-1.61), and 13.25 for indomethacin (aHR 1.36, 95% CI 1.11-1.66) compared to non-users. Among patients with a past history of CVD (N = 84 752) there were 397 977 person-years of follow-up and 10 248 events. Non-users had 28.30 events/1000 person-years. Among those with CVD, rofecoxib use was associated with increased event rate (30.28 events/1000 person-years [aHR 1.21, 95% CI 1.08-1.37]) and naproxen was associated with a decreased event rate (22.66 events/1000 person-years [aHR 0.88, 95% CI 0.79-0.99]). Among new users, the results were similar except risk among naproxen users was no longer different than non-users. CONCLUSIONS: We found an increased risk of cardiovascular events among all and new current users of rofecoxib, valdecoxib, and indomethacin in patients with no history of CVD. Among patients with CVD, all and new current rofecoxib use was associated with an increased risk of a cardiovascular event.

Use of disease risk scores in pharmacoepidemiologic studies.

Automated databases are increasingly used in pharmacoepidemiologic studies. These databases include records of prescribed medications and encounters with medical care providers from which one can construct very detailed surrogate measures for both drug exposure and covariates that are potential confounders. Often it is possible to track day-by-day changes in these variables. However, while this information is often critical for study success, its volume can pose challenges for statistical analysis. One common approach is the use of propensity scores. An alternative approach is to construct a disease risk score. This is analogous to the propensity score in that it calculates a summary measure from the covariates. However, the disease risk score estimates the probability or rate of disease occurrence conditional on being unexposed. The association between exposure and disease is then estimated adjusting for the disease risk score in place of the individual covariates. This review describes the use of disease risk scores in pharmacoepidemiologic studies, and includes a brief discussion of their history, a more detailed description of their construction and use, a summary of simulation studies comparing their performance vis-á-vis traditional models, a comparison of their utility with that of propensity scores, and some further topics for future research.