Prescriber perceptions of boxed warnings: A qualitative study.

PURPOSE: To explore how boxed warning (BW) information fits within the context of prescribers' overall treatment decision-making and communication with patients. METHODS: In-depth interviews (N = 52) were conducted with primary care providers and specialists. Participants were presented with one of two prescribing scenarios: (1) estrogen vaginal inserts to treat vulvovaginal atrophy (VVA) associated with menopause; or (2) direct-acting antivirals (DAA) to treat chronic hepatitis C virus infection (HCV). The semi-structured interviews explored participants' treatment decision-making within the scenario, reactions to current prescribing information for a product within the FDA-approved drug class, as well as their perceptions of BWs generally. RESULTS: Across scenarios, providers described that the BW is only one of several factors that influence treatment decision-making. In the VVA scenario, symptom severity, family history, and experience with nonprescription drugs were raised as common factors that influence prescribing considerations; compared to comorbid infections, viral load, and HCV genotype in the HCV scenario. Perceptions of the DAA BW were generally positive or neutral, as many participants found the information important and appropriate. The VVA BW was viewed less favorably, with many participants stating the BW overstates the risk for this drug. CONCLUSIONS: Findings suggest that BWs are one of several factors that influence providers' treatment decisions, and BW influence largely depends on context. Providers across scenarios expressed notable differences in their perceptions of the risk information provided in the presented BWs; however, across scenarios participants expressed consideration of how patients may perceive the BW.

Trend changes of national zolpidem users and exposure cases after FDA drug safety communications.

PURPOSE: To evaluate the impact of FDA's 2013 zolpidem Drug Safety Communications (DSCs), which recommended lowering the initial dose to mitigate drowsiness, on national estimates of zolpidem users and zolpidem exposure cases. METHODS: We analyzed trend changes of national zolpidem users from the IQVIA Total Patient Tracker (TPT) and zolpidem exposure cases reported to the National Poison Data System (NPDS), 2009-2018. To control for time varying confounding, the adjusted trends were analyzed using simple and controlled interrupted time series (ITS). We also adjusted for seasonal changes. Three sedating antidepressants were used together as a control. RESULTS: The national estimates of high-dose zolpidem users in TPT decreased significantly in the month immediately post-DSC; the absolute level decrease was -12.51 (95% CI: -14.12, -10.89) per 10 000 U.S. population relative to sedating antidepressants. The trend continuously decreased post-DSC, resulting in a 59% overall decrease by the end of the study period. There was a larger decrease in high-dose zolpidem use in females than in males. There was a level decrease of zolpidem exposure cases in the NPDS immediately post-DSC, -0.37 absolute decline (95% CI, -0.53, -0.20) per 10 000 national zolpidem users; or -1.33 absolute decline (95% CI, -1.54, -1.13) per 1000 total NPDS exposure cases relative to sedating antidepressants. Similar patterns were observed for cases reporting drowsiness. The results from the single ITS and controlled ITS were similar. CONCLUSIONS: Zolpidem users and exposure cases decreased significantly post-DSC, suggesting practitioners and patients became aware of and responded to the zolpidem DSCs.

Validity of ICD-10-CM diagnoses to identify hospitalizations for serious infections among patients treated with biologic therapies.

PURPOSE: Identifying hospitalizations for serious infections among patients dispensed biologic therapies within healthcare databases is important for post-marketing surveillance of these drugs. We determined the positive predictive value (PPV) of an ICD-10-CM-based diagnostic coding algorithm to identify hospitalization for serious infection among patients dispensed biologic therapy within the FDA's Sentinel Distributed Database. METHODS: We identified health plan members who met the following algorithm criteria: (1) hospital ICD-10-CM discharge diagnosis of serious infection between July 1, 2016 and August 31, 2018; (2) either outpatient/emergency department infection diagnosis or outpatient antimicrobial treatment within 7 days prior to hospitalization; (3) inflammatory bowel disease, psoriasis, or rheumatological diagnosis within 1 year prior to hospitalization, and (4) were dispensed outpatient biologic therapy within 90 days prior to admission. Medical records were reviewed by infectious disease clinicians to adjudicate hospitalizations for serious infection. The PPV (95% confidence interval [CI]) for confirmed events was determined after further weighting by the prevalence of the type of serious infection in the database. RESULTS: Among 223 selected health plan members who met the algorithm, 209 (93.7% [95% CI, 90.1%-96.9%]) were confirmed to have a hospitalization for serious infection. After weighting by the prevalence of the type of serious infection, the PPV of the ICD-10-CM algorithm identifying a hospitalization for serious infection was 80.2% (95% CI, 75.3%-84.7%). CONCLUSIONS: The ICD-10-CM-based algorithm for hospitalization for serious infection among patients dispensed biologic therapies within the Sentinel Distributed Database had 80% PPV for confirmed events and could be considered for use within pharmacoepidemiologic studies.

Use of Time-Dependent Propensity Scores to Adjust Hazard Ratio Estimates in Cohort Studies with Differential Depletion of Susceptibles.

Estimating hazard ratios (HR) presents challenges for propensity score (PS)-based analyses of cohorts with differential depletion of susceptibles. When the treatment effect is not null, cohorts that were balanced at baseline tend to become unbalanced on baseline characteristics over time as "susceptible" individuals drop out of the population at risk differentially across treatment groups due to having outcome events. This imbalance in baseline covariates causes marginal (population-averaged) HRs to diverge from conditional (covariate-adjusted) HRs over time and systematically move toward the null. Methods that condition on a baseline PS yield HR estimates that fall between the marginal and conditional HRs when these diverge. Unconditional methods that match on the PS or weight by a function of the PS can estimate the marginal HR consistently but are prone to misinterpretation when the marginal HR diverges toward the null. Here, we present results from a series of simulations to help analysts gain insight on these issues. We propose a novel approach that uses time-dependent PSs to consistently estimate conditional HRs, regardless of whether susceptibles have been depleted differentially. Simulations show that adjustment for time-dependent PSs can adjust for covariate imbalances over time that are caused by depletion of susceptibles. Updating the PS is unnecessary when outcome incidence is so low that depletion of susceptibles is negligible. But if incidence is high, and covariates and treatment affect risk, then covariate imbalances arise as susceptibles are depleted, and PS-based methods can consistently estimate the conditional HR only if the PS is periodically updated.

Changes in emergency department visits for zolpidem-attributed adverse drug reactions after FDA Drug Safety Communications.

Purpose: To identify possible changes in U.S. emergency department (ED) visits from zolpidem-attributed adverse drug reactions (ADRs) after 2013 Food and Drug Administration (FDA) Drug Safety Communications (DSCs), which notified the public about FDA's new dosing recommendations for zolpidem. Methods: We estimated the occurrence of ED visits from zolpidem-attributed ADRs using nationally representative, public health surveillance of medication harms (National Electronic Injury Surveillance System-Cooperative Adverse Drug Event Surveillance project, 2010-2017). We estimated the number of zolpidem prescriptions using IQVIA National Prescription Audit, 2010-2017. We calculated rates of ED visits for zolpidem-attributed ADRs per 10 000 dispensed zolpidem prescriptions and identified time trends and potential inflection points using joinpoint regression. For comparison, we repeated these analyses for sedating antidepressants commonly used to treat disordered sleep (trazodone, doxepin, and mirtazapine). Results: The best-fit regression model for rates of ED visits for zolpidem-attributed ADRs by 6-month intervals identified a single inflection point in the second half of 2014 (P = .024) with a 6.7% biannual decrease from 2010 to 2014 ([-13.1%, 0.3%], P = .059) and a 13.9% biannual increase from the second half of 2014 through 2017 ([-1.1%, 31.3%], P = .068). No change or inflection points were identified for rates of ED visits for sedating antidepressant-attributed ADRs. Conclusions: While there was a nominal decline in the rate of ED visits for ADRs in the time period before and for 18 months after FDA's 2013 zolpidem DSCs, the decrease was not sustained, and thus questions remain concerning the long-term impact of the zolpidem DSCs on ADRs.

FDA postmarketing safety labeling changes: What have we learned since 2010 about impacts on prescribing rates, drug utilization, and treatment outcomes.

PURPOSE: Prior literature reviews have identified gaps in understanding of how postmarketing safety labeling changes and related FDA communications impact key clinical and behavioral outcomes. We conducted a review of newly published studies on this topic to determine what new evidence exists and to identify which gaps may still remain. We believe that this information can support FDA as it develops and implements future risk communication approaches. METHODS: We searched PubMed and Embase for studies published between January 1, 2010, and August 7, 2017 that examined the impact of labeling changes or associated FDA safety-related communications. For each study, we extracted information on research design and findings for key clinical outcomes and behaviors. We also conducted a ROBINS-I review to identify potential for bias in the research design of each study. RESULTS: We found that the estimated impacts of FDA labeling changes on several key outcomes-including adverse events-varied. Labeling changes also yielded unintended consequences on drug prescribing in some cases, despite low provider adherence. Finally, some studies we reviewed exhibited potential for bias due to confounding, among other factors. CONCLUSIONS: The new studies we reviewed contain many of the same limitations identified in previously published reviews. While there are several challenges to conducting this research there is substantial room for improvement in the quality of the evidence base. More information, particularly with respect to the types of populations and medications affected by labeling changes, is needed to support the development of more effective and targeted safety communications.

Long-Term Risk of Acute Myocardial Infarction, Stroke, and Death With Outpatient Use of Clarithromycin: A Retrospective Cohort Study.

In a retrospective cohort study of patients enrolled in the UK Clinical Practice Research Datalink during 2000-2013, we evaluated long-term risks of death, stroke, and acute myocardial infarction (AMI) in adults prescribed clarithromycin. Patients were outpatients aged 40-85 years, who were prescribed clarithromycin (n = 287,748), doxycycline (n = 267,729), or erythromycin (n = 442,999), or Helicobacter pylori eradication therapy with a proton pump inhibitor, amoxicillin, and either clarithromycin (n = 27,639) or metronidazole (n = 14,863). We analyzed time to death, stroke, or AMI with Cox proportional hazards regression. The long-term hazard ratio for death following 1 clarithromycin versus 1 doxycycline prescription was 1.29 (95% confidence interval (CI): 1.21, 1.25), increasing to 1.62 (95% CI: 1.43, 1.84) for ≥5 prescriptions of clarithromycin versus ≥5 prescriptions for doxycycline. Erythromycin showed smaller risks in comparison with doxycycline. Stroke and AMI incidences were also increased after clarithromycin but with smaller hazard ratios than for mortality. For H. pylori eradication, the hazard ratio for mortality following clarithromycin versus metronidazole regimens was 1.09 (95% CI: 1.00, 1.18) overall, and it was higher (hazard ratio = 1.65, 95% CI: 0.88, 3.08) following ≥2 prescriptions in subjects not on statins at baseline. Outpatient clarithromycin use was associated with long-term mortality increases, with evidence for a similar, smaller increase with erythromycin.

Data Mining for Adverse Drug Events With a Propensity Score-matched Tree-based Scan Statistic.

The tree-based scan statistic is a statistical data mining tool that has been used for signal detection with a self-controlled design in vaccine safety studies. This disproportionality statistic adjusts for multiple testing in evaluation of thousands of potential adverse events. However, many drug safety questions are not well suited for self-controlled analysis. We propose a method that combines tree-based scan statistics with propensity score-matched analysis of new initiator cohorts, a robust design for investigations of drug safety. We conducted plasmode simulations to evaluate performance. In multiple realistic scenarios, tree-based scan statistics in cohorts that were propensity score matched to adjust for confounding outperformed tree-based scan statistics in unmatched cohorts. In scenarios where confounding moved point estimates away from the null, adjusted analyses recovered the prespecified type 1 error while unadjusted analyses inflated type 1 error. In scenarios where confounding moved point estimates toward the null, adjusted analyses preserved power, whereas unadjusted analyses greatly reduced power. Although complete adjustment of true confounders had the best performance, matching on a moderately mis-specified propensity score substantially improved type 1 error and power compared with no adjustment. When there was true elevation in risk of an adverse event, there were often co-occurring signals for clinically related concepts. TreeScan with propensity score matching shows promise as a method for screening and prioritization of potential adverse events. It should be followed by clinical review and safety studies specifically designed to quantify the magnitude of effect, with confounding control targeted to the outcome of interest.

The impact of FDA regulatory activities on incident dispensing of LABA-containing medication: 2005-2011.

OBJECTIVE: Evidence of safety issues associated with long-acting beta2-agonist (LABA) treatment has led to multiple regulatory activities by the U.S. Food and Drug Administration (FDA) on this class of medications. This study describes the impact of the regulatory activities on incident LABA-containing medication dispensing. METHODS: A monthly rolling cohort of asthma patients who were eligible to initiate a LABA-containing product was created in the Mini-Sentinel Distributed Database between January 2005 and June 2011. Cohorts of individuals who initiated LABA were examined for the changes in the proportions of single-ingredient to fixed-dose inhaled corticosteroid (ICS)-LABA initiators, appropriate initiation of LABA-containing products, and use of controller medications. The impact of the 2005 and 2010 FDA regulatory activities associated with LABA-containing products was measured using interrupted time series with segmented regression. RESULTS: LABA-containing product initiation was declining prior to the 2005 regulatory activities and continued to decline over the study period, accompanied by increased initiation of fixed dose ICS-LABA among LABA initiators. While the 2010 regulatory activities had no immediate impact on the proportion of LABA initiation in patients with prior controller medication dispensing and/or poor asthma control, there was an increasing positive trend toward LABA initiation in the appropriate patient population after the regulatory activities. CONCLUSION: The 2005 and 2010 FDA regulatory activities likely had an impact on communicating the safety concerns of LABA products. However, the impact cannot be viewed independent of scientific publications, guidelines for asthma treatment and other regulatory activities.

Changes in prescribing and healthcare resource utilization after FDA Drug Safety Communications involving zolpidem-containing medications.

PURPOSE: Products containing the sedative/hypnotic zolpidem were subject to Drug Safety Communications (DSCs) in January and May 2013 describing the risk of next-morning impairment and recommending lower starting doses particularly for women. This study aimed to assess whether zolpidem DSCs were associated with prescribing-pattern changes between January 2011 and December 2013. METHODS: We assessed overall dispensings of zolpidem-containing products between January 2011 and December 2013 by conducting a time-series analysis. Analyses were stratified by gender because the DSC contained gender-specific information. Participants were patients drawn from the Optum Clinformatics data source of commercially insured people in the USA. We evaluated changes in mean prescribed dose of the two drugs and health care utilization metrics. RESULTS: Each month of the study, more than 80 000 patients received a zolpidem-containing product and approximately one-tenth as many received eszopiclone. The two DSCs did not affect the downward trajectory of new zolpidem prescriptions. However, there was an increase in use of lower-dose forms of zolpidem (30% increase, p < 0.001), coupled with a reduction in higher-dose forms (13% decrease, p = 0.03), so that the average dose decreased after the DSCs (from 9.7 mg to 9.4 mg, p < 0.001), a change that was not seen with eszopiclone (from 2.74 mg to 2.74 mg, p = 0.45). CONCLUSION: The DSCs related to zolpidem-containing products shifted prescribing toward the lower-dose formulations, consistent with the recommendations in the DSCs. Copyright © 2017 John Wiley & Sons, Ltd.

Media Coverage of FDA Drug Safety Communications about Zolpidem: A Quantitative and Qualitative Analysis.

FDA issues Drug Safety Communications (DSCs) to alert health care professionals and the public about emerging safety information affecting prescription and over-the-counter drugs. News media may amplify DSCs, but it is unclear how DSC messaging is transmitted through the media. We conducted a content analysis of the lay media coverage reaching the broadest audience to characterize the amount and content of media coverage of two zolpidem DSCs from 2013. After the first DSC, zolpidem news stories increased from 19 stories/week in the preceding 3 months to 153 following its release. Most (81%) appeared in the lay media, and 64% focused on the DSC content. After the second DSC, news stories increased from 24 stories/week in the preceding 3 months to 39 following. Among the 100 unique lay media news stories, at least half correctly reported three key DSC messages: next-day impairment and drowsiness as common safety hazards, lower doses for some but not all zolpidem products, and women's higher risk for impairment. Other DSC messages were reported in fewer than one-third of stories, such as the warning that impairment can happen even when people feel fully awake. The first-but not the second-zolpidem DSC generated high-profile news coverage. The finding that some messages were widely reported but others were not emphasizes the importance of ensuring translation of key DSC content.

Changing patterns of asthma medication use related to US Food and Drug Administration long-acting ß2-agonist regulation from 2005-2011.

BACKGROUND: Safety concerns associated with long-acting ß2-agonists (LABAs) have led to many US Food and Drug Administration (FDA) regulatory activities for this class of drugs. Little is known about the effect of these regulatory activities on use of LABA-containing agents or other asthma medications. METHODS: We created rolling cohorts of pediatric and adult asthmatic patients in the Mini-Sentinel Distributed Database between January 2005 and June 2011. The proportions of asthmatic patients using LABA-containing products, inhaled corticosteroids (ICSs), leukotriene modifiers, short-acting ß2-agonists, oral corticosteroids, other bronchodilators, and no medications were measured on a monthly basis, and the changes were evaluated by using interrupted time series with segmented regression analysis. RESULTS: When the 2005 regulatory activity was announced, there were statistically significant decreases in the use of fixed-dose ICS-LABA agents in children (-0.98 percentage points) and adults (-1.24 percentage points). Increased use of ICSs and leukotriene modifiers was observed just after the regulatory activities were announced in both children and adults. Although of smaller magnitude, continued favorable changes in the use of LABA agents were observed after the 2010 FDA regulatory activity. CONCLUSION: The 2005 and 2010 FDA regulatory activities might have contributed to reduced use of LABA agents, as intended; however, their effect, independent of other factors, cannot be determined. Use of other classes of asthma medications was similarly affected.

Trends in rates of acetaminophen-related adverse events in the United States.

PURPOSE: The goal of this study is to summarize trends in rates of adverse events attributable to acetaminophen use, including hepatotoxicity and mortality. METHODS: A comprehensive analysis of data from three national surveillance systems estimated rates of acetaminophen-related events identified in different settings, including calls to poison centers (2008-2012), emergency department visits (2004-2012), and inpatient hospitalizations (1998-2011). Rates of acetaminophen-related events were calculated per setting, census population, and distributed drug units. RESULTS: Rates of poison center calls with acetaminophen-related exposures decreased from 49.5/1000 calls in 2009 to 43.5/1000 calls in 2012. Rates of emergency department visits for unintentional acetaminophen-related adverse events decreased from 58.0/1000 emergency department visits for adverse drug events in 2009 to 50.2/1000 emergency department visits in 2012. Rates of hospital inpatient discharges with acetaminophen-related poisoning decreased from 119.8/100 000 hospitalizations in 2009 to 108.6/100 000 hospitalizations in 2011. After 2009, population rates of acetaminophen-related events per 1 million census population decreased for poison center calls and hospitalizations, while emergency department visit rates remained stable. However, when accounting for drug sales, the rate of acetaminophen-related events (per 1 million distributed drug units) increased after 2009. Prior to 2009, the rates of acetaminophen-related hospitalizations had been slowly increasing (p-trend = 0.001). CONCLUSIONS: Acetaminophen-related adverse events continue to be a public health burden. Future studies with additional time points are necessary to confirm trends and determine whether recent risk mitigation efforts had a beneficial impact on acetaminophen-related adverse events. Published 2015. This article is a U.S. Government work and is in the public domain in the USA.

Long-acting beta2-adrenergic agonist in pediatric and adolescent asthma patients, 2003-2011.

OBJECTIVES: To evaluate changes in the dispensing patterns of long-acting beta2-adrenergic agonist (LABA) in pediatric and adolescent asthma patients in relation to multiple Food and Drug Administration (FDA) regulatory activities from 2003 to 2011. METHODS: We estimated LABA dispensing to pediatric asthma patients across three periods: 2003-2004 (after the first labeling change), 2005-2009 (after regulatory activities in 2005 and before 2010 LABA labeling change) and 2010-2011 (after 2010 LABA labeling change), using the IMS Health Plan Claims database. We estimated dispensing patterns over time for single-ingredient (SI) LABA and fixed-dose combination (FDC) of inhaled corticosteroid (ICS) and LABA (FDC-ICS/LABA). We also evaluated prior use of non-LABA asthma-control medication (ACM) before LABA initiation. RESULTS: Of the 147 862 pediatric and adolescent asthma patients who initiated a LABA during the entire study period, the majority (96%) were FDC-ICS/LABA initiators. The proportion of SI-LABA among any LABA initiators was small and declined (9%, 4% and 2%, trend test p < 0.001) for the three periods. Among the patients who initiated, the proportions with prior use of an ACM (1-90 days prior) were 35%, 36% and 39% for the three periods. CONCLUSIONS: The significant decline in the proportion of SI-LABA initiation over these years is consistent with FDA's recommendations. However, the favorable trend cannot be solely attributed to FDA activities as changes to clinical practice guidelines, and media publicity may have played a role. Investigating the reasons for the low ACM use before LABA initiation may inform approaches to further improve appropriate use of LABA in young asthma patients.

Risk of acute myocardial infarction, stroke, or death in patients initiating olmesartan or other angiotensin receptor blockers - a cohort study using the Clinical Practice Research Datalink.

PURPOSE: Results of two randomized trials (ROADMAP and ORIENT) suggest that high-dose (40 mg/day) olmesartan (Olm) is associated with increased cardiovascular mortality compared to placebo in diabetic patients. We evaluated the risks of acute myocardial infarction (AMI) and death in patients initiating Olm compared with an active comparator group, other angiotensin receptor blockers (ARBs), with a focus on high-dose and diabetic subgroups. METHODS: We conducted a cohort study with patients who initiated Olm or another ARB between 2003 and 2011, using the UK Clinical Practice Research Datalink GOLD. We included patients who had no prior ARB or angiotensin converting enzyme inhibitor exposure during the preceding 6 months. Hazard ratios (HRs) were estimated using Cox regression models with both multivariable adjustment and propensity score matching. RESULTS: There were 3964 Olm and 54 653 other-ARB initiators, respectively. Adjusted HRs comparing Olm and other-ARBs were 1.04 (95% CI: 0.75-1.42) for AMI and 1.16 (0.95-1.42) for death, using multivariable adjustment. Comparing patients initiated with a high-dose Olm and a high-dose other-ARB, HRs were 3.09 (0.94-10.13) for AMI and 2.03 (0.74-5.61) for death, using multivariable adjustment; and 4.38 (0.97-19.66) and 1.99 (0.63-6.32) for AMI and death, using propensity score matching. CONCLUSIONS: Overall, no differences in risk were observed in the main cohort analyses comparing Olm initiators with patients initiating therapy with other ARBs; however, HRs were marginally increased for all study endpoints which compared high-dose subgroups, suggesting potential increased risk may be associated with high-dose Olm. Published 2013. This article is a U.S. Government work and is in the public domain in the USA.

Cardiovascular and mortality risk in elderly Medicare beneficiaries treated with olmesartan versus other angiotensin receptor blockers.

PURPOSE: In the randomized trial, Randomized Olmesartan and Diabetes Microalbuminuria Prevention, acute cardiovascular death was increased nearly fivefold in diabetic patients treated with high-dose olmesartan, an angiotensin receptor blocker (ARB), compared with placebo. METHODS: Medicare beneficiaries were entered into new-user cohorts of olmesartan or other ARBs and followed on therapy for occurrence of acute myocardial infarction, stroke, or death. Analyses focused on specific subgroups defined by diabetes status, ARB dose, and duration of therapy. Hazard ratios (HR) with 95% confidence intervals (CIs) were estimated using Cox proportional hazards regression, with other ARBs as reference. RESULTS: A total of 158,054 olmesartan and 724,673 other ARB users were followed for 54,285 and 260,390 person-years, respectively, during which 9237 endpoint events occurred. Lower-dose olmesartan was not associated with increased risk for any endpoint, regardless of duration of use. High-dose olmesartan for 6 months or longer was associated with increased risk of death in patients with diabetes (HR 2.03, 95%CI 1.09-3.75, p = 0.02) and with reduced risk in nondiabetic patients (HR 0.46, 95%CI 0.24-0.86, p = 0.01). Some, but not all, sensitivity analyses suggested that selective prescribing of olmesartan to healthier patients (channeling bias) may have accounted for the reduced risk in nondiabetic patients. CONCLUSIONS: High-dose olmesartan was associated with an increased risk of death in diabetic patients treated for 6 months or longer and with a reduced risk of death in nondiabetic patients, when compared with use of other ARBs. This latter effect was probably because of selective prescribing of olmesartan to healthier patients, although effect modification cannot be excluded. Published 2013. This article is a U.S. Government work and is in the public domain in the USA.

Description and Validation of a Novel AI Tool, LabelComp, for the Identification of Adverse Event Changes in FDA Labeling.

INTRODUCTION: The accurate identification and timely updating of adverse reactions in drug labeling are crucial for patient safety and effective drug use. Postmarketing surveillance plays a pivotal role in identifying previously undetected adverse events (AEs) that emerge when a drug is used in broader and more diverse patient populations. However, traditional methods of updating drug labeling with new AE information have been manual, time consuming, and error prone. This paper introduces the LabelComp tool, an innovative artificial intelligence (AI) tool designed to enhance the efficiency and accuracy of postmarketing drug safety surveillance. Utilizing a combination of text analytics and a trained Bidirectional Encoder Representations from Transformers (BERT) model, the LabelComp tool automatically identifies changes in AE terms from updated drug labeling documents. OBJECTIVE: Our objective was to create and validate an AI tool with high accuracy that could enable researchers and FDA reviewers to efficiently identify safety-related drug labeling changes. RESULTS: Our validation study of 87 drug labeling PDF pairs demonstrates the tool's high accuracy, with F1 scores of overall performance ranging from 0.795 to 0.936 across different evaluation tiers and a recall of at least 0.997 with only one missed AE out of 483 total AEs detected, indicating the tool's efficacy in identifying new AEs. CONCLUSION: The LabelComp tool can support drug safety surveillance and inform regulatory decision-making. The publication of this tool also aims to encourage further community-driven enhancements, aligning with broader interests in applying AI to advance regulatory science and public health.

Trends in Use and Evidence of Adherence to Risk Evaluation and Mitigation Strategy Pregnancy Testing Requirements for Thalidomide, Lenalidomide, and Pomalidomide in the USA, 2000-2020.

INTRODUCTION: Lenalidomide, pomalidomide, and thalidomide are effective treatments for multiple myeloma but are teratogenic. To mitigate this risk, the US Food and Drug Administration (FDA) required risk evaluation and mitigation strategy (REMS) programs for these drugs, which include pregnancy testing among women of childbearing potential-twice before initiation, weekly in the first month on treatment, and every 2-4 weeks thereafter. OBJECTIVE: We evaluated dispensing trends of lenalidomide, pomalidomide, and thalidomide and assessed adherence to REMS pregnancy testing requirements among at-risk patients taking these drugs. METHODS: Using three US health insurance claims databases (Optum Clinformatics® [2004-2020], Merative Marketscan [2003-2019], and Medicaid [2000-2018]), we assessed monthly use of the drugs, patient characteristics and treatment persistence among drug initiators, and claims-based evidence for adherence to pregnancy testing requirements among initiators with child-bearing potential. RESULTS: Lenalidomide was the most prescribed agent following its approval in 2006 and through the end of the study period. A total of 48,311 lenalidomide (mean age = 59 years [standard deviation (SD) = 16]), 17,550 thalidomide (mean age = 65 years [SD = 12]), and 6560 pomalidomide initiators (mean age = 65 years [SD = 11]) were identified; 45% of initiators of each drug were women. Among initiators under follow-up on day 90, 70% were still on therapy. Initiators of childbearing potential comprised 3% (N = 1,920) of all initiators; among this cohort, 12% had evidence in claims data of two pregnancy tests before initiation, and 9% with at least 33 days of follow-up of four tests during the first month of treatment. By contrast, 52% who received a refill had claims-based evidence of a pregnancy test within 7 days of dispensing. CONCLUSION: Although most patients who initiated lenalidomide, pomalidomide, and thalidomide were not of child-bearing potential, further investigation into actual non-adherence to pregnancy testing is needed.

How Do Risk Evaluation and Mitigation Strategies Impact Clinical Practice? A National Survey of Physicians.

The US Food and Drug Administration can require risk evaluation and mitigation strategy (REMS) programs for prescription drugs to ensure the benefits of use outweigh the risks. We conducted a national survey of physicians' experiences prescribing eight REMS-covered drugs: (1) ambrisentan; (2) bosentan; (3) clozapine; (4) isotretinoin; (5-7) the multiple myeloma (MM) drugs lenalidomide, pomalidomide, thalidomide; and (8) sodium oxybate. Between May 2022 and January 2023, we surveyed 5,331 physician prescribers of these drugs, and 1,295 (24%) returned surveys (range: 149 for bosentan to 226 for MM drugs). Although 765 (68%) respondents thought the certification process provided useful drug information, 757 (67%) wanted materials to include benefit data and 944 (84%) non-REMS-related risk data. A majority (704, 63%) thought the safe use requirements facilitated discussion with patients, but a similar number (637, 57%) attributed delayed medication access to these requirements. In multivariable modeling, MM drug and isotretinoin respondents were less likely than sodium oxybate respondents to agree that the certification process provided useful drug information (MM drug: odds ratio (OR) = 0.37, 95% confidence interval (CI) = 0.25-0.55; isotretinoin: OR = 0.39, 95% CI = 0.25-0.61), and isotretinoin, clozapine, and bosetan respondents were more likely than sodium oxybate respondents to agree that the safe use requirements often delayed medication access (isotretinoin: OR = 5.83, 95% CI = 3.70-9.19; clozapine: OR = 1.65, 95% CI = 1.08-2.54; bosentan: OR = 1.78, 95% CI = 1.12-2.85). Most physicians believe REMS programs convey useful drug safety information and facilitate discussion with patients but also seek information on benefits and non-REMS-related risks and better integration of REMS processes into clinical workflows.

Removing the FDA's Boxed Hepatotoxicity Warning and Liver Function Testing Requirement for Ambrisentan.

Importance: Endothelin receptor antagonists are first-line therapy for pulmonary arterial hypertension (PAH). The first 2 agents approved in the class, bosentan and ambrisentan, initially carried boxed warnings for hepatotoxicity and required monthly liver function tests (LFTs) as part of a risk evaluation and mitigation strategy (REMS); however, in 2011, as further safety data emerged on ambrisentan, the boxed hepatotoxicity warning and LFT requirements were removed. Objective: To analyze changes in the use of and LFT monitoring for ambrisentan and bosentan after changes to the ambrisentan labeling and REMS. Design, Setting, and Participants: This serial cross-sectional study used data from 3 longitudinal health care insurance claims databases-Medicaid, Optum's deidentified Clinformatics Data Mart, and Merative Marketscan-to perform an interrupted time series analysis of prescription fills and LFTs for patients taking ambrisentan and bosentan. Participants were patients filling prescriptions for ambrisentan and bosentan from July 1, 2007, to December 31, 2018. Data analysis was performed from April 2021 to August 2023. Exposure: Removal of the boxed warning for hepatotoxicity and the REMS LFT monitoring requirements on ambrisentan in March 2011. Main Outcomes and Measures: The primary outcomes were use of ambrisentan (ie, individuals with at least 1 dispensing per 1 000 000 individuals enrolled in the 3 datasets) vs bosentan and LFT monitoring (ie, proportion of initiators with at least 1 ordered test) before initiation and before the first refill. Results: A total of 10 261 patients received a prescription for ambrisentan during the study period (7442 women [72.5%]; mean [SD] age, 52.6 [17.6] years), and 11 159 patients received a prescription for bosentan (7931 women [71.1%]; mean [SD] age, 47.7 [23.7] years). Removal of the ambrisentan boxed hepatotoxicity warning and LFT monitoring requirement was associated with an immediate increase in the use of ambrisentan (1.50 patients per million enrollees; 95% CI, 1.08 to 1.92 patients per million enrollees) but no significant change in the use of bosentan. There were reductions in recorded LFTs before drug initiation (13.1% absolute decrease; 95% CI, -18.2% to -8.0%) and before the first refill (26.4% absolute decrease; 95% CI, -34.4% to -18.5%) of ambrisentan but not bosentan. Conclusions and Relevance: In this serial cross-sectional study of ambrisentan, labeling changes and removal of the REMS-related LFT requirement were associated with shifts in prescribing and testing behavior for ambrisentan but not bosentan. Further clinician education may be needed to maximize the benefits of REMS programs and labeling warnings designed to ensure the safe administration of high-risk medications.