Drivers of treatment patterns in patients with chronic lymphocytic leukemia stopping ibrutinib or idelalisib therapies.

Novel therapies including kinase inhibitors (KI) have led to high and durable response in patients with chronic lymphocytic leukemia (CLL), however, some patients stop therapy. This study evaluates reasons for treatment changes among CLL patients who stopped KI in real-world practice. Sixty-nine US oncologists/hematologists provided patient-level data abstracted from charts of CLL adult patients who initiated a KI and later (1) switched to another anti-neoplastic regimen (Switched cohort), (2) discontinued the KI and remained untreated (Discontinued cohort), or (3) restarted the same KI after an interruption of ≥60 days (Restarted cohort). Demographics, clinical/treatment characteristics, and reasons for stopping, restarting, and switching the KI therapy were described. In the Switched cohort, reasons for stopping included disease progression (72.5%), low/no disease activity (3.9%), adverse event [AE]/ intolerance/comorbidity (15.7%), and planned cellular therapies (7.9%). In the Discontinued cohort, approximately half (46.0%) of patients stopped KI therapy because they were terminally ill/died, or were moved to best supportive care - these patients were older, had more severe disease, and high comorbidity burden. The other half (54.0% of patients) stopped due to low/no disease activity (24.0%), AEs/toxicity (12.0%), or patient-requested drug holiday (18.0%). In the Restarted cohort, the most common reasons for stopping KIs were patient request (37.3%), AEs/intolerance (31.4%), and economic reasons (10%). Patients restarted when disease progressed (60.8%) or when they recovered from the AE (33%). Reasons for KI stop and subsequent treatment patterns were varied and multifactorial, suggesting heterogeneous disease management and a need for more evidence around supporting strategies and physician education.

Predicting Acute Exacerbations in Chronic Obstructive Pulmonary Disease.

BACKGROUND: With increasing health care costs that have outpaced those of other industries, payers of health care are moving from a fee-for-service payment model to one in which reimbursement is tied to outcomes. Chronic obstructive pulmonary disease (COPD) is a disease where this payment model has been implemented by some payers, and COPD exacerbations are a quality metric that is used. Under an outcomes-based payment model, it is important for health systems to be able to identify patients at risk for poor outcomes so that they can target interventions to improve outcomes. OBJECTIVE: To develop and evaluate predictive models that could be used to identify patients at high risk for COPD exacerbations. METHODS: This study was retrospective and observational and included COPD patients treated with a bronchodilator-based combination therapy. We used health insurance claims data to obtain demographics, enrollment information, comorbidities, medication use, and health care resource utilization for each patient over a 6-month baseline period. Exacerbations were examined over a 6-month outcome period and included inpatient (primary discharge diagnosis for COPD), outpatient, and emergency department (outpatient/emergency department visits with a COPD diagnosis plus an acute prescription for an antibiotic or corticosteroid within 5 days) exacerbations. The cohort was split into training (75%) and validation (25%) sets. Within the training cohort, stepwise logistic regression models were created to evaluate risk of exacerbations based on factors measured during the baseline period. Models were evaluated using sensitivity, specificity, and positive and negative predictive values. The base model included all confounding or effect modifier covariates. Several other models were explored using different sets of observations and variables to determine the best predictive model. RESULTS: There were 478,772 patients included in the analytic sample, of which 40.5% had exacerbations during the outcome period. Patients with exacerbations had slightly more comorbidities, medication use, and health care resource utilization compared with patients without exacerbations. In the base model, sensitivity was 41.6% and specificity was 85.5%. Positive and negative predictive values were 66.2% and 68.2%, respectively. Other models that were evaluated resulted in similar test characteristics as the base model. CONCLUSIONS: In this study, we were not able to predict COPD exacerbations with a high level of accuracy using health insurance claims data from COPD patients treated with bronchodilator-based combination therapy. Future studies should be done to explore predictive models for exacerbations. DISCLOSURES: No outside funding supported this study. Samp is now employed by, and owns stock in, AbbVie. The other authors have nothing to disclose. Study concept and design were contributed by Joo and Pickard, along with the other authors. Samp and Lee performed the data analysis, with assistance from the other authors. Samp wrote the manuscript, which was revised by Schumock and Calip, along with the other authors.

Risk of Cardiovascular and Cerebrovascular Events in COPD Patients Treated With Long-Acting ß2-Agonist Combined With a Long-Acting Muscarinic or Inhaled Corticosteroid.

BACKGROUND: The recent approval of several fixed-dose combination long-acting ß2-agonist (LABA) and long-acting muscarinic antagonist (LAMA) products has increased the use of dual bronchodilators in the treatment of chronic obstructive pulmonary disease (COPD). Understanding the comparative safety of this combination is important for informing treatment decisions. OBJECTIVE: To compare the risk of cardiovascular and cerebrovascular (CCV) events associated with LABA/LAMA compared with a combination of LABA and inhaled corticosteroid (ICS). METHODS: This was a retrospective, observational cohort study using health insurance claims data to identify COPD patients initiating LABA/LAMA or LABA/ICS. CCV outcomes included hospitalizations with a primary diagnosis for acute coronary syndrome, heart failure, cardiac dysrhythmia, stroke, or transient ischemic attack. Patients were followed until they experienced an event, discontinued treatment, initiated medication from the opposite cohort, or lost enrollment. Patients were matched 1:4 on propensity scores, and time to event was compared using Cox proportional hazards models. RESULTS: After matching, there were 3842 patients in the LABA/LAMA cohort and 15 225 in the LABA/ICS cohort. Cardiovascular events in the LABA/LAMA cohort were lower than in the LABA/ICS: hazard ratio (HR) = 0.794; 95% CI = 0.623-0.997. No significant difference in the risk of cerebrovascular events (HR = 1.166; 95% CI = 0.653-1.959) was observed. CONCLUSIONS: Despite concerns about the CCV effects of LAMA and LABA monotherapy, the LABA/LAMA combination had similar or lower risk of these events in comparison to LABA/ICS. Further studies are recommended to confirm these findings.

Comparative Effectiveness of Long-Acting Beta2 -Agonist Combined with a Long-Acting Muscarinic Antagonist or Inhaled Corticosteroid in Chronic Obstructive Pulmonary Disease.

INTRODUCTION: Several dual bronchodilator fixed-dose inhaler medications were recently approved for the treatment of chronic obstructive pulmonary disease (COPD). These products combine a long-acting ß2 -agonist (LABA) and long-acting muscarinic antagonist (LAMA). In clinical trials, the separate mechanisms of the bronchodilators resulted in improved lung function. COPD treatment guidelines currently recommend combination LABA/LAMA as alternative therapy to combination LABA/inhaled corticosteroid (ICS). Evidence is limited on the comparative effectiveness of LABA/LAMA and LABA/ICS in COPD. The objective of this study was to compare real-world COPD exacerbation rates among patients treated with LABA/LAMA with those treated with LABA/ICS. METHODS: This was a retrospective observational study of COPD patients in the United States treated with LABA/LAMA or LABA/ICS combination. Insurance claims from January 1, 2004, through December 31, 2014, were used as the data source. Patients were required to have greater than one prescription filled for the combination medications, and they were followed from 30 days after drug initiation. Individuals were censored if they discontinued a study medication, initiated medication from the opposite cohort (LAMA or ICS), lost enrollment eligibility, or at the study period end. Exacerbation rates were compared using Poisson regression. RESULTS: There were 5384 patients in the LABA/LAMA cohort and 473,388 patients in the LABA/ICS cohort. The LABA/LAMA cohort was older, had more comorbidities, and more severe COPD. Unadjusted annual exacerbation rates were 2.87 events per person-year (standard deviation [SD] 5.14) in the LABA/LAMA cohort and 1.68 (SD 9.82) in the LABA/ICS cohort. The adjusted incidence rate ratio was 0.98 (95% confidence interval 0.95-1.01) for LABA/LAMA compared with LABA/ICS. CONCLUSIONS: The LABA/LAMA combination had similar effectiveness to LABA/ICS as measured by exacerbation rates in COPD patients. As a result, characteristics other than effectiveness, such as symptom control, cost, patient preferences, and adverse events, may be important in selecting between the two regimens.

Cost-effectiveness of direct-acting antiviral regimen ombitasvir/paritaprevir/ritonavir in treatment-naïve and treatment-experienced patients infected with chronic hepatitis C virus genotype 1b in Japan.

OBJECTIVE: This study compared the cost-effectiveness of chronic hepatitis C virus (HCV) genotype 1b (GT1b) therapy ombitasvir/paritaprevir/ritonavir (OBV/PTV/r) vs daclatasvir + asunaprevir (DCV/ASV) and no treatment in patients without cirrhosis. Cost-effectiveness analyses (CEAs) that compared OBV/PTV/r against DCV/ASV and sofosbuvir/ledipasvir (SOF/LDV) in Y93H mutation-negative, GT1b patients with and without cirrhosis were also included. METHODS: A health state transition model was developed to capture the natural history of HCV. A CEA over a lifetime horizon was performed from the perspective of the public healthcare payer in Japan. Costs, health utilities, and rates of disease progression were derived from published studies. Sustained virologic response (SVR) rates of OBV/PTV/r and DCV/ASV were extracted from Japanese clinical trials. Analyses were performed for treatment-naïve and -experienced patients. Alternative scenarios and input parameter uncertainty on the results were tested. RESULTS: OBV/PTV/r exhibited superior clinical outcomes vs comparators. For OBV/PTV/r, DCV/ASV, and no treatment, the lifetime risk of decompensated cirrhosis in treatment-naïve patients without cirrhosis was 0.4%, 1.4%, and 9.2%, and hepatocellular carcinoma was 6.5%, 11.4%, and 49.9%, respectively. Quality-adjusted life years (QALYs) were higher in treatment-naïve and -experienced patients without cirrhosis treated with OBV/PTV/r (16.41 and 16.22) vs DCV/ASV (15.83 and 15.66) or no treatment (11.34 and 11.23). In treatment-naïve and -experienced patients without cirrhosis, the incremental cost-effectiveness ratios (ICERs) of OBV/PTV/r vs DCV/ASV were JPY 1,684,751/QALY and JPY 1,836,596/QALY, respectively; OBV/PTV/r was dominant compared with no treatment. In scenario analysis, including GT1b patients with and without cirrhosis who were Y93H mutation-negative, the ICER of OBV/PTV/r vs DCV/ASV was below the Japanese willingness-to-pay threshold of JPY 5 million/QALY, while the ICER of SOF/LDV vs OBV/PTV/r was above this threshold; thus, OBV/PTV/r was cost-effective. CONCLUSION: OBV/PTV/r appears to be a cost-effective treatment for chronic HCV GT1b infection against DCV/ASV. OBV/PTV/r dominates no treatment in patients without cirrhosis.

Economic evaluation of ombitasvir/paritaprevir/ritonavir and dasabuvir for the treatment of chronic genotype 1 hepatitis c virus infection.

OBJECTIVES: To estimate clinical outcomes and cost-effectiveness of ombitasvir/paritaprevir/ritonavir and dasabuvir ± ribavirin (OMB/PTV/r + DSV ± RBV) compared with treatment regimens including pegylated interferon (PegIFN) for patients with chronic genotype 1 hepatitis C virus (HCV) infection. METHODS: An Excel spreadsheet Markov model tracking progression through stages of liver disease was developed. Costs and patient utilities for liver disease stages were taken from published studies. Rates of disease progression were based on studies of untreated HCV infection and long-term follow-up of those achieving sustained virologic response (SVR) after drug treatment. Impact of OMB/PTV/r + DSV ± RBV and other drug regimens on progression was estimated through SVR rates from clinical trials. Analyses were performed for treatment-naive and treatment-experienced patients. Impact of alternative scenarios and input parameter uncertainty on the results were tested. RESULTS: For genotype 1 treatment-naive HCV patients, for OMB/PTV/r + DSV ± RBV, PegIFN + ribavirin (PegIFN/RBV), sofosbuvir + PegIFN/RBV, telaprevir + PegIFN/RBV, boceprevir + PegIFN/RBV, lifetime risk of decompensated liver disease was 5.6%, 18.9%, 7.4%, 11.7%, and 14.9%; hepatocellular carcinoma was 5.4%, 9.2%, 5.7%, 7.0%, and 7.4%; and death from liver disease was 8.7%, 22.2%, 10.4%, 14.8%, and 17.6%, respectively. Estimates of the cost-effectiveness of OMB/PTV/r + DSV ± RBV for treatment-naive and treatment-experienced patients indicated that it dominated all other regimens except PegIFN/RBV. Compared with PegIFN/RBV, the incremental cost-effectiveness ratios were £13,864 and £10,258 per quality-adjusted life-year (QALY) for treatment-naive and treatment-experienced patients, respectively. The results were similar for alternative scenarios and uncertainty analyses. LIMITATIONS: A mixed-treatment comparison for SVR rates for the different treatment regimens was not feasible, because many regimens did not have comparator arms; instead SVR rates were based on those from recent trials. CONCLUSIONS: OMB/PTV/r + DSV ± RBV is a cost-effective oral treatment regimen for chronic genotype 1 HCV infection compared with standard treatment regimens and is estimated to reduce the lifetime risks of advanced liver disease.

Liver-specific case fatality due to chronic hepatitis C virus infection: a systematic review.

Despite reports that mortality is increasing, overall case fatality due to hepatitis C virus (HCV) is thought to be low. Given the variability in published rates, we aimed to synthesize estimates of liver-specific case fatality and all-cause mortality in chronic HCV according to follow-up duration, sustained viral response (SVR) to treatment, and liver disease severity. A systematic review was conducted of studies published in English from 2003 to 2013, reporting liver-specific case fatality estimates from HCV-infected samples. Thirty-five eligible articles were identified; 26 also presented estimates of all-cause mortality. Among community-based samples, liver-specific case fatality ranged from 0.3% over 5.7 years to 9.2% over 8.2 years of follow-up; and of all-cause mortality, from 4.0% over 5.7 years, to 23.0% over 8.2 years of follow-up. Estimates were higher among clinic-based samples and those with more severe liver disease. Among treated patients achieving SVR, liver-specific case fatality was low: up to 1.4% over 11.5 years of follow-up among samples with any severity of liver disease. Estimates were higher among those without SVR: up to 14.0% over 10 years of followup among samples with any severity of liver disease, and higher still among samples with more severe liver disease. The proportion of deaths attributable to liver-specific causes ranged from 55 to 85% among those with severe liver disease. Published estimates of fatality are high among certain populations of chronic HCV patients, with liver-specific causes being an important contributor. Understanding current HCV mortality rates is important for quantifying the total burden of HCV disease.

Association Between Cognitive Function and Health Care Costs 3 Months and 6 Months After Initiating Antidepressant Medication for Depressive Disorders.

BACKGROUND: Major depressive disorder is one of the most common and disabling mental health disorders and is associated with substantial costs in terms of direct health care utilization and workplace productivity. Cognitive dysfunction, which alone substantially increases health care costs, is commonly associated with major depressive disorder. However, the health care costs of cognitive dysfunction in the context of depressive disorder are unknown. Recovery from mood symptoms is not always associated with resolution of cognitive dysfunction. Thus, cognitive dysfunction may contribute to health care burden even with successful antidepressant therapy.  OBJECTIVE: To compare health care utilization and costs for patients with a depressive disorder with and without cognitive dysfunction, at 3 and 6 months after initiation of antidepressant medication.  METHODS: This was an observational study, combining a cross-sectional patient survey, administered during a telephone interview, with health care claims data from a large, geographically diverse U.S. health plan. Included patients had at least 1 pharmacy claim for an antidepressant medication between August 1 and September 30, 2012, and no claim for any antidepressant during the 6 months prior to the index date. In addition to other criteria assessed in the claims data, patients confirmed a diagnosis of depression or major depressive disorder and the absence of any exclusionary neurological diagnoses possibly associated with cognitive impairment. Eligible patients were administered validated cognitive function assessments of verbal episodic memory (Hopkins Verbal Learning Test-Revised, Delayed and Total); attention (Digit Span Forward Maximum Sequence Length); working memory (Digit Span Backward Maximum Sequence Length); and executive function (D-KEFS-Letter Fluency Test). Based on comparison of scores with normative data, patients were assigned to cognitive dysfunction or cognitive normal cohorts. All-cause (all diagnoses) and depressive disorder-related health care utilization and costs (all from a payer perspective) were assessed 6 months prior (baseline) to antidepressant initiation and 3 months and 6 months after (follow-up) initiation of antidepressant medication. Health care utilization and costs included ambulatory (office and hospital outpatient), emergency room, inpatient hospital, pharmacy, other medical (e.g., laboratory and diagnostics), and total (all categories combined). All-cause and depressive disorder-related total costs during the 3- and 6-month follow-up periods were modeled with generalized linear modeling with gamma distribution and log link, while adjusting for potential confounders (age, race, gender, education, employment, and comorbidities). RESULTS: Of the 13,537 patients who were mailed an invitation, 824 (6%) were eligible and agreed to participate. Of these, 563 patients provided informed consent, completed the interview, maintained eligibility, and were included in the 3-month calculations. Among these, 255 (45%) were classified as having cognitive dysfunction. Mean patient age was 41.3 (± 12.5) years; 80% were female. Most patients were white and employed. More patients in the cognitive normal cohort were white (P less than 0.001) and employed full time (P = 0.029), had higher education attainment (P less than 0.001), and had fewer comorbidities (P = 0.007) than those in the cognitive dysfunction cohort. Over the first 3 months, patients with cognitive dysfunction had higher adjusted all-cause costs ($3,309 vs. $2,157, P = 0.002) and higher adjusted depressive disorder-related costs ($718 vs. $406, P less than 0.001) than patients without cognitive dysfunction. At 6 months, data from 4 patients were removed from the analysis because of exclusionary diagnoses. Over 6 months, patients with cognitive dysfunction had higher adjusted all-cause costs ($4,793) than patients without cognitive dysfunction ($3,683, P = 0.034). Over 6 months, depressive disorder-related costs did not significantly differ between patients with ($771) and without cognitive dysfunction ($594, P = 0.071). The main drivers of all-cause costs were office visits, outpatient hospital visits, and inpatient costs, and the main driver of depressive disorder-related costs was inpatient costs. CONCLUSIONS: Cognitive dysfunction was associated with higher adjusted all-cause and depressive disorder-related costs 3 months after initiation of an antidepressant medication. This difference persisted for all-cause costs through 6 months. Identification and treatment of cognitive dysfunction in patients with depressive disorder might reduce health care costs.

Costs and absence of HCV-infected employees by disease stage.

OBJECTIVES: Quantify the costs and absenteeism associated with stages of the Hepatitis C virus (HCV). STUDY DESIGN: Retrospective analysis of the HCMS integrated database from multiple geographically diverse, US-based employers with employee information on medical, prescription, and absenteeism claims. METHODS: Employee data were extracted from July 2001-March 2013. Employees with HCV were identified by ICD-9-CM codes and classified into disease severity cohorts using diagnosis/procedure codes assigning the first date of most severe claim as the index date. Non-HCV employees (controls) were assigned random index dates. Inclusion required 6-month pre-/post-index eligibility. Medical, prescription, and absenteeism cost and time were analyzed using two-part regression (logistic/generalized linear) models, controlling for potentially confounding factors. Costs were inflation adjusted to September 2013. RESULTS: All direct costs comparisons were statistically significant (p ≤ 0.05) with mean medical costs of $1813 [SE = $3] for controls (n = 727,588), $4611 [SE = $211] for non-cirrhotic (n = 1007), $4646 [SE = $721] for compensated cirrhosis (CC, n = 87), $12,384 [SE = $1122] for decompensated cirrhosis (DCC, n = 256), $33,494 [SE = $11,753] for hepatocellular carcinoma (HCC, n = 17) and $97,724 [SE = $32,437] for liver transplant (LT, n = 19) cohorts. Mean short-term disability days/costs were significantly greater for the non-cirrhotic (days = 2.03 [SE = 0.36]; $299 [SE = $53]), DCC (days = 6.20 [SE = 1.36]; $763 [SE = $169]), and LT cohorts (days = 21.98 [SE = 8.21]; $2537 [SE = $972]) compared to controls (days = 1.19 [SE = 0.01]; $155 [SE = $1]). Mean sick leave costs were significantly greater for non-cirrhotic ($373 [SE = $22]) and DCC ($460 [SE = $54]) compared to controls ($327 [SE = $1]). CONCLUSIONS: Employees with HCV were shown to have greater direct and indirect costs compared to non-HCV employee controls. Costs progressively increased in the more severe HCV disease categories. Slowing or preventing disease progression may avert the costs of more severe liver disease stages and enable employees with HCV to continue as productive members of the workforce.

Patient health utility, work productivity, and lifestyle impairment in chronic hepatitis C patients in France.

BACKGROUND: France has a high prevalence of patients with chronic hepatitis C virus (HCV). Clinical consequences of HCV are well-recognized, while health-related quality of life (HRQoL) and productivity impacts remain less understood. This study evaluates how HCV disease severity and HCV treatment outcomes impact HRQoL and productivity among patients in France. METHODS: From October 2012 to January 2013, physicians treating HCV patients in France completed Patient Record Forms, which included information on patient demographics, disease stage, and treatment status. Subsequently, these HCV patients completed the EQ-5D-3L health-state instrument and the HCV-specific Work Productivity and Activity Impairment (WPAI:HepC) Questionnaire. Results are reported in descriptive and stratified analyses by disease stage and treatment status. Linear regression analyses were performed to determine independent associations between disease severity and treatment status with EQ-5D and WPAI:HepC. RESULTS: There were 297 matched physician and patient response forms completed. Mean EQ-5D Index score was 0.764, and mean EQ-VAS score was 65.85. Regression analyses showed that older age and worse disease severity were significantly associated with lower EQ-5D Index and EQ-VAS scores. Stratification of EQ-5D Index and EQ-VAS scores showed significantly better scores for HCV treatment responders compared to non-responders. Stratification of WPAI:HepC questions by disease stage revealed greater productivity impact on HCV patients with more severe disease. CONCLUSIONS: In a cross-sectional sample of HCV patients in France, worsening HRQoL and productivity/activity impairment was significantly associated with disease progression and increasing age. This information provides insight into the benefits of treating HCV patients and preventing disease progression.

Economic evaluation of the impact of medication errors reported by U.S. clinical pharmacists.

OBJECTIVE: Medication errors defined as "any preventable event that may cause or lead to inappropriate medication use or patient harm" have been highlighted as a top national priority in a report issued by the Institute of Medicine. However, little information is available on precise costs of medication errors. This study estimated the cost of medication errors reported by clinical pharmacists using a modified societal perspective. METHODS: Information on 779 medication errors was collected in the Medication Error Detection, Amelioration and Prevention (MEDAP) study that documented medication errors observed by clinical pharmacists during a consecutive 14-day period. The rate of medication errors, outcomes (number of errors resulting in temporary/permanent patient harm, prolonged hospitalization, or life-sustaining therapy), and interventions (communication, medication changes, patient monitoring, and treatment referrals) were collected. A decision model was developed to estimate the economic impact of medication errors reported by clinical pharmacists. Event probabilities were derived from MEDAP data. Direct costs were obtained through reviews of the literature, hospital charge data, and Medicare and Medicaid reimbursement. One-way and Monte Carlo sensitivity analyses were used to explore uncertainty in the values. RESULTS: In the base case, the mean expected cost of a medication error was $88.57. In the Monte Carlo simulation, the mean cost was $89.35 (± $30.17 SD). One-way sensitivity analysis revealed that changes in the probability of medication errors causing hospitalization and the cost of hospitalization had the greatest variability on the outcome ($50.44-$155.81 [probability of hospitalization], $32.59-$136.40 [cost of hospitalization]). CONCLUSIONS: Medication errors are costly to the health care system. A better understanding of medication error costs may be used to justify initiatives to reduce the risk and inefficiency associated with these errors.

Retracted publications in the drug literature.

Recent studies have suggested an increase in the number of retracted scientific publications. It is unclear how broadly the issue of misleading and fraudulent publications pertains to retractions of drug therapy studies. Therefore, we sought to determine the trends and factors associated with retracted publications in drug therapy literature. A PubMed search was conducted to identify retracted drug therapy articles published from 2000-2011. Articles were grouped according to reason for retraction, which was classified as scientific misconduct or error. Scientific misconduct was further divided into data fabrication, data falsification, questions of data veracity, unethical author conduct, and plagiarism. Error was defined as duplicate publication, scientific mistake, journal error, or unstated reasons. Additional data were extracted from the retracted articles, including type of article, funding source, author information, therapeutic area, and retraction issue. A total of 742 retractions were identified from 2000-2011 in the general biomedical literature, and 102 drug studies met our inclusion criteria. Of these, 73 articles (72%) were retracted for a reason classified as scientific misconduct, whereas 29 articles (28%) were retracted for error. Among the 73 articles classified as scientific misconduct, those classified as unethical author conduct (32 articles [44%]) and data fabrication (24 articles [33%]) constituted the majority. The median time from publication of the original article to retraction was 31 months (range 1-130). Fifty percent of retracted articles did not state a funding source, whereas pharmaceutical manufacturer funding accounted for only 13 articles (13%) analyzed. Many retractions were due to repeat offenses by a small number of authors, with nearly 40% of the retracted studies associated with two individuals. We found that a greater proportion of drug therapy articles were retracted for reasons of misconduct and fraud compared with other biomedical studies. It is important for health care practitioners to monitor the literature for retractions so that recommendations for drug therapy and patient management may be modified accordingly.