Impact of thiazolidinedione safety warnings on medication use patterns and glycemic control among veterans with diabetes mellitus

Deriving Weight from Big Data: A Comparison of Body Weight Measurement Cleaning Algorithms (Preprint)

Background

Patient body weight is a frequently used measure in biomedical studies, yet there are no standard methods for processing and cleaning weight data. Conflicting documentation on constructing body weight measurements presents challenges for research and program evaluation.

Objective

In this study, we aim to describe and compare methods for extracting and cleaning weight data from electronic health record databases to develop guidelines for standardized approaches that promote reproducibility.

Methods

We conducted a systematic review of studies published from 2008 to 2018 that used Veterans Health Administration electronic health record weight data and documented the algorithms for constructing patient weight. We applied these algorithms to a cohort of veterans with at least one primary care visit in 2016. The resulting weight measures were compared at the patient and site levels.

Results

We identified 496 studies and included 62 (12.5%) that used weight as an outcome. Approximately 48% (27/62) included a replicable algorithm. Algorithms varied from cutoffs of implausible weights to complex models using measures within patients over time. We found differences in the number of weight values after applying the algorithms (71,961/1,175,995, 6.12% to 1,175,177/1,175,995, 99.93% of raw data) but little difference in average weights across methods (93.3, SD 21.0 kg to 94.8, SD 21.8 kg). The percentage of patients with at least 5% weight loss over 1 year ranged from 9.37% (4933/52,642) to 13.99% (3355/23,987).

Conclusions

Contrasting algorithms provide similar results and, in some cases, the results are not different from using raw, unprocessed data despite algorithm complexity. Studies using point estimates of weight may benefit from a simple cleaning rule based on cutoffs of implausible values; however, research questions involving weight trajectories and other, more complex scenarios may benefit from a more nuanced algorithm that considers all available weight data.

Construction and Use of Body Weight Measures from Administrative Data in a Large National Health System: A Systematic Review

OBJECTIVE

Administrative data are increasingly used in research and evaluation yet lack standardized guidelines for constructing measures using these data. Body weight measures from administrative data serve critical functions of monitoring patient health, evaluating interventions, and informing research. This study aimed to describe the algorithms used by researchers to construct and use weight measures.

METHODS

A structured, systematic literature review of studies that constructed body weight measures from the Veterans Health Administration was conducted. Key information regarding time frames and time windows of data collection, measure calculations, data cleaning, treatment of missing and outlier weight values, and validation processes was collected.

RESULTS

We identified 39 studies out of 492 nonduplicated records for inclusion. Studies parameterized weight outcomes as change in weight from baseline to follow-up (62%), weight trajectory over time (21%), proportion of participants meeting weight threshold (46%), or multiple methods (28%). Most (90%) reported total time in follow-up and number of time points. Fewer reported time windows (54%), outlier values (51%), missing values (34%), or validation strategies (15%).

CONCLUSIONS

A high variability in the operationalization of weight measures was found. Improving methods to construct clinical measures will support transparency and replicability in approaches, guide interpretation of findings, and facilitate comparisons across studies.

Unintended Effects of Communicating About Drug Safety Issues: A Critical Review of the Literature

Communications about the safety and effectiveness of human drugs can influence patients' and prescribers' perceptions and behaviors, which in turn can affect the public's health more broadly. We conducted a critical review of the literature on the unintended effects from communicating information to the public about safety issues with prescription and over-the-counter drugs. We searched PubMed for peer-reviewed studies published from 1990 to 2017 where study authors reported probable unintended effects of communicating drug safety. The types of communications included in these studies were news reports, direct-to-consumer advertisements, and those released by government agencies. Among the 26 studies identified, the most commonly reported unintended effects were decreased drug use or discontinuation. Other unintended effects included spillover to populations not targeted by the communications (e.g., discontinuation of antidepressants among adults following communications concerning use among youth), shifts in clinical diagnoses (e.g., fewer diagnoses of depression), increased use of alternative therapies, and other undesirable behaviors (e.g., possible increased suicide attempts because antidepressants were discontinued). Limitations to the literature include the inability to establish causation or to isolate the effects of multiple communication sources and messages. Further, because the intended effect of many communications was not known, our study was limited by challenges in defining some effects as unintended. Most studies used health insurer claims data to identify unintended effects of communications, which provide an incomplete picture; few used self-reported or other methodologies that could help illuminate the reasons underlying the effects observed in the claims data. Best practices for communicating about the potential benefits and harms of drugs in a manner that minimizes negative unintended effects are needed to protect and improve public health.

Chemical Crosslinking Mass Spectrometry Reveals the Conformational Landscape of the Activation Helix of PPARγ; a Model for Ligand-Dependent Antagonism

Peroxisome proliferator-activated receptors (PPARs) are pharmacological targets for the treatment of metabolic disorders. Previously, we demonstrated the anti-diabetic effects of SR1664, a PPARγ modulator lacking classical transcriptional agonism, despite its poor pharmacokinetic properties. Here, we report identification of the antagonist SR11023 as a potent insulin sensitizer with significant plasma exposure following oral administration. To determine the structural mechanism of ligand-dependent antagonism of PPARγ, we employed an integrated approach combining solution-phase biophysical techniques to monitor activation helix (helix 12) conformational dynamics. While informative on receptor dynamics, hydrogen/deuterium exchange mass spectrometry and nuclear magnetic resonance data provide limited information regarding the specific orientations of structural elements. In contrast, label-free quantitative crosslinking mass spectrometry revealed that binding of SR11023 to PPARγ enhances interaction with co-repressor motifs by pushing H12 away from the agonist active conformation toward the H2-H3 loop region (i.e., the omega loop), revealing the molecular mechanism for active antagonism of PPARγ.

Geographic Variation in Rosiglitazone Use Surrounding FDA Warnings in the Department of Veterans Affairs

BACKGROUND

Geographic variation in the use of prescription drugs, particularly those deemed harmful by the FDA, may lead to variation in patient exposure to adverse drug events. One such drug is the glucose-lowering drug rosiglitazone, for which the FDA issued a safety alert on May 21, 2007, following the publication of a meta-analysis that suggested a 43% increase in the risk of myocardial infarction with the use of rosiglitazone. This alert was followed by a black box warning on August 14, 2007, that was updated 3 months later. While large declines have been documented in rosiglitazone use in clinical practice, little is known about how the use of rosiglitazone and other glucose-lowering drugs varied within the Department of Veterans Affairs (VA), surrounding the FDA alerts. Understanding this variation within integrated health care systems is essential to formulating policies that enhance patient protection and quality of care.

OBJECTIVE

To document variation in the use of rosiglitazone and other glucose- lowering drugs across 21 Veterans Integrated Service Networks (VISNs).

METHODS

We conducted a retrospective analysis of drug use patterns for all major diabetes drugs in a national cohort of 550,550 veterans with diabetes from 2003 to 2008. This included the time periods when rosiglitazone was added to (November 2003) and removed from (October 2007) the VA national formulary (VANF). We employed multivariable logistic regression models to statistically estimate the association between a patient's location and the patient's odds of using rosiglitazone.

RESULTS

Aggregate rosiglitazone use increased monotonically from 7.7%, in the quarter it was added to the VANF (November 4, 2003), to a peak of 15.3% in the quarter when the FDA issued the safety alert. Rosiglitazone use decreased sharply afterwards, reaching 3.4% by the end of the study period (September 30, 2008). The use of pioglitazone, another glucose-lowering drug in the same class as rosiglitazone, was low when the FDA issued the safety alert (0.4%) but increased sharply afterwards, reaching 3.6% by the end of the study period. Insulin use increased monotonically; metformin use remained relatively flat; and sulfonylurea use exhibited a general declining trend throughout the study period. Statistically significant geographic variation was observed in rosiglitazone use throughout the study period. The prevalence range, defined as the range of minimum to maximum use across VISNs was 3.7%-12.4% in the first quarter (January 1 to March 31, 2003); 1.0%-5.5% in the last quarter of study period (July 1 to September 30, 2008); and reached a peak of 9.6%-25.5% in the quarter when the FDA safety alert was issued (April 1 to March 31, 2007). In 5 VISNs, peak rosiglitazone use occurred before the FDA issued the safety alert. The odds ratio of using rosiglitazone in a given VISN varied from 0.55 (95% CI = 0.52-0.59; VISN 10) to 1.58 (95% CI = 1.50-1.66; VISN 15), with VISN 1 being the reference region. The variation was higher in the periods after the FDA issued the safety alert. Much less variation was observed in the use of pioglitazone, metformin, sulfonylurea, and insulin.

CONCLUSIONS

Our results show statistically significant variation in the way VISNs within the VA responded to the FDA alerts, suggesting a need for mechanisms that disseminate information and guidelines for drug use in a consistent and reliable manner. Further study of regions that adopted ideal practices earlier may provide lessons for regional leadership and practice culture within integrated health care systems.

Disparities in Discontinuing Rosiglitazone Following the 2007 FDA Safety Alert

BACKGROUND

Responsiveness to the Food and Drug Administration (FDA) rosiglitazone safety alert, issued on May 21, 2007, has not been examined among vulnerable subpopulations of the elderly.

OBJECTIVE

To compare time to discontinuation of rosiglitazone after the safety alert between black and white elderly persons, and across sociodemographic and economic subgroups.

RESEARCH DESIGN

A cohort study.

SUBJECTS

Medicare fee-for-service enrollees in 2007 who were established users of rosiglitazone identified from a 20% national sample of pharmacy claims.

MEASURES

Outcome of interest was time to discontinuation of rosiglitazone after the May alert. We modeled the number of days following the warning to the end of the days' supply for the last rosiglitazone claim during the study period (May 21, 2007-December 31, 2007) using multivariable proportional hazards models.

RESULTS

More than 67% of enrollees discontinued rosiglitazone within six months of the advisory. In adjusted analysis, white enrollees (hazard ratio=0.90; 95% confidence interval, 0.86-0.94) discontinued rosiglitazone later than the comparison group of black enrollees. Enrollees with a history of low personal income also discontinued later than their comparison group (hazard ratio=0.84; 95% confidence interval, 0.81-0.87). There were no observed differences across quintiles of area-level socioeconomic status.

CONCLUSIONS

White race and a history of low personal income modestly predicted later discontinuation of rosiglitazone after the FDA's safety advisory in 2007. The impact of FDA advisories can vary among sociodemographic groups. Policymakers should continue to monitor whether risk management policies reach their intended populations.

Novel Therapeutics for Diabetes: Uptake, Usage Trends, and Comparative Effectiveness

The number of available therapies for treating type 2 diabetes has grown considerably in recent years. This growth has been fueled by availability of newer medications, whose benefits and risks have not been fully established. In this study, we review and synthesize the existing literature on the uptake, efficacy, safety, and cost-effectiveness of novel antidiabetic agents. Specifically, we focus on three drug classes that were introduced in the market recently: thiazolidinediones (TZDs), dipeptidyl peptidase-4 (DPP-4) inhibitors, and glucagon-like peptide-1 (GLP-1) receptor agonists. Not surprisingly, we find that the usage trends reflect the efficacy and safety profile of these novel drugs. The use of TZDs increased initially but decreased after a black-box warning was issued for rosiglitazone in 2007 that highlighted the cardiovascular risks associated with using the drug. Conversely, DPP-4 inhibitors and GLP-1 receptor agonists gained market shares due to their efficacy in glycemic control as an add-on treatment to metformin. DPP-4 inhibitors were the most commonly prescribed agents among the three novel drug classes, likely because they are relatively less expensive, have better safety profile, are administered orally, and are weight neutral. Sitagliptin was the most preferred DPP-4 inhibitor. The level of evidence on the comparative effectiveness, safety, and cost implications of using novel antidiabetic agents remains low and further studies with long-term follow-ups are needed.

Rosiglitazone use and associated adverse event rates in Canada: an updated analysis

Background

We previously reported on the change in the use of rosiglitazone-containing products (RCP) and adverse event reporting rates in Canadian patients between 2004 and 2010. The present study extends this analysis to include the January 2011 to December 2012 time period.

Methods

RCP utilization rates were obtained from IMS Health Brogan’s longitudinal de-identified patient database, LRx. GlaxoSmithKline’s global adverse events database was used to extract adverse events (AE), serious adverse events (SAE), and cardiac adverse events (CAE) reported in Canadian patients receiving RCP from April 2004 to December 2012. The patient utilization information from the LRx database was used to estimate rates per 100,000 patients.

Results

An estimated 182,841 patients were dispensed RCP prescriptions between April 2004 and December 2012. The total number of patients using RCP decreased by 85 % from 2011 to 2012. From its peak use in 2007, the number of patients filling a prescription decreased 97 %. A total of 1069 AEs were reported during the study period, of which 32 AE’s were reported from Jan 2011 to Dec 2012. The average monthly reporting rates of AE’s, SAE’s and CAE’s over 2011–2012 were 10.8/100,000 patients, 9.1/100,000 patients and 5.0/100,000 patients, respectively.

Conclusions

The utilization of RCP in Canada has significantly declined. The significance of the adverse event rate information presented is uncertain and must be evaluated within the context of the well known factors that can influence AE reporting rates, as well as limitations to the methods used to estimate these reporting rates.

Effects of safety warnings and risk management plan for Thiazolidinediones in Taiwan

PURPOSE

To evaluate changes in thiazolidinedione use and quality of prescription following safety warnings for thiazolidinediones and cardiac risk in 2007, Risk Management Plan (RMP) policy for rosiglitazone in 2010, and warning for pioglitazone and bladder cancer risk in 2010 in Taiwan.

METHODS

We obtained 2003-2011 claims data from Taiwan's National Health Insurance Research Database. Using an interrupted time series design and segmented regression, we estimated changes in monthly prescribing rates for thiazolidinediones among all and prevalent diabetes patients with and without cardiovascular disease history (CV history). We also compared time to prescription of thiazolidinediones among new diabetes patients with CV history before and after each regulatory action using survival analysis.

RESULTS

Among prevalent patients with and without CV history, the prescribing rates of rosiglitazone decreased 36.88% and 28.92% after safety warnings in 2007 respectively. Pioglitazone prescriptions increased 13% among patients with CV history, but no changes were detected among patients without CV history. After rosiglitazone's RMP policy in 2010, large reductions in prescriptions were observed in patients with CV history (-101.67%) and those without CV history (-88.04%). Among new diabetes patients with CV history, cardiac safety warnings in 2007 significantly delayed the prescription of rosiglitazone, but no significant change was found for pioglitazone.

CONCLUSIONS

The Taiwan FDA regulatory actions for thiazolidinediones communicated possible risks of cardiac events and bladder cancer. Different safety regulatory actions had differential impacts on the use of rosiglitazone and pioglitazone and the quality use of these drugs among the high-risk patients.

How Did Multiple FDA Actions Affect the Utilization and Reimbursed Costs of Thiazolidinediones in US Medicaid?

PURPOSE

The US Food and Drug Administration (FDA) communicated the potential cardiovascular risk of thiazolidinediones (rosiglitazone and pioglitazone) in 2007 and required a Risk Evaluation and Mitigation Strategy (REMS) for rosiglitazone in 2010. It also communicated in 2010 the potential risk of bladder cancer with pioglitazone use. This study examined the effects of these multiple FDA actions on utilization and reimbursed costs of thiazolidinediones in state Medicaid programs.

METHODS

State Drug Utilization Data from the Centers for Medicare & Medicaid Services were assessed. An interrupted time series design and segmented linear regression models were used to examine changes in market shares according to both prescription volume and reimbursed costs for rosiglitazone and pioglitazone in the Northeast and Midwest regions of the United States after multiple FDA actions.

FINDINGS

Compared with expected rates, there were relative reductions of 65.84% (Northeast region) and 55.09% (Midwest region) in the use of rosiglitazone at 1 year after the 2007 FDA actions for thiazolidinediones and cardiac risk. At the same time, relative increases of 7.30% and 9.28% in the use of pioglitazone were observed in the Northeast and Midwest regions, respectively. Changes in both use and costs of rosiglitazone after the 2010 REMS program could not be estimated because of the already low rates (~1%) before REMS was implemented. One year after the 2010 FDA actions for pioglitazone and its possible association with bladder cancer, relative reductions in pioglitazone use of 21.41% (Northeast region) and 18.12% (Midwest region) were detected.

IMPLICATIONS

The Northeast and Midwest regions reported similar patterns of changes after the FDA actions. Use and costs of rosiglitazone were substantially reduced after the 2007 FDA actions for cardiovascular risk, and this drug was rarely used after the 2010 REMS program. Conversely, use and costs of pioglitazone were substantially reduced after the 2010 FDA actions regarding the drug's possible risk of bladder cancer.

The effect of withdrawal of rosiglitazone on treatment pathways, diabetes control and patient outcomes: a retrospective cohort study

AIMS

To describe the withdrawal of rosiglitazone and the impact upon glycaemic control; intensification of therapy; and progression to major adverse cardiovascular events (MACE), cancer and mortality.

METHODS

Data were from the Clinical Practice Research Datalink (CPRD), a longitudinal U.K. database. Rosiglitazone use was profiled from launch (2000) until withdrawal (2010). Patients discontinuing from July 2010 were included in the analysis to ascertain the impact on glycaemic control; therapy intensification; and progression to MACE, death and cancer. For comparison, patients were matched to those maintained on pioglitazone as a control group.

RESULTS

Rosiglitazone use peaked in May 2007. Of patients prescribed rosiglitazone at discontinuation 54.1% patients used a dual-therapy regimen; most commonly with metformin (46.7%). 65.1% patients remained at the same stage of the diabetes pathway following discontinuation. 51.7% of patients replaced rosiglitazone with pioglitazone. Patients discontinuing were more likely (HR=2.29), to subsequently intensify therapy than controls. After discontinuation of rosiglitazone there was a significant increase in HbA1c, from a median of 6.9% to 7.3%. In matched analysis, there was a significantly greater increase in HbA1c for rosiglitazone patients (0.33% versus 0.10%). Following discontinuation, crude rates for MACE, cancer and mortality were 8.4, 17.9 and 15.8 pkpy, respectively. None was significantly different in the matched analysis.

CONCLUSION

Withdrawal of rosiglitazone was associated with worsening glucose control and subsequent intensification of treatment regimen.

Use of antidiabetic drugs in the U.S., 2003-2012

OBJECTIVE

To describe market trends for antidiabetic drugs, focusing on newly approved drugs, concomitant use of antidiabetic drugs, and effects of safety concerns and access restrictions on thiazolidinedione use.

RESEARCH DESIGN AND METHODS

Nationally projected data on antidiabetic prescriptions for adults dispensed from U.S. retail pharmacies were extracted from IMS Health Vector One National and Total Patient Tracker for 2003-2012 and from Encuity Research Treatment Answers and Symphony Health Solutions PHAST Prescription Monthly for 2012.

RESULTS

Since 2003, the number of adult antidiabetic drug users increased by 42.9% to 18.8 million in 2012. Metformin use increased by 97.0% to 60.4 million prescriptions dispensed in retail pharmacies in 2012. Among antidiabetic drugs newly approved for marketing between 2003 and 2012, the dipeptidyl peptidase-4 (DPP-4) inhibitor sitagliptin had the largest share with 10.5 million prescriptions in 2012. Rosiglitazone use plummeted to <13,000 prescriptions dispensed in retail or mail-order pharmacies in 2012. Concomitancy analyses showed that 44.9% of metformin use was for monotherapy. Between 33.4 and 48.1% of sulfonylurea, DPP-4 inhibitor, thiazolidinedione, and glucagon-like peptide 1 analog use was not accompanied by metformin.

CONCLUSIONS

The antidiabetic drug market is characterized by steady increases in volume, and newly approved drugs experienced substantial uptake, especially DPP-4 inhibitors. The use of rosiglitazone has been negligible since restrictions were put in place in 2011. Further study is needed to understand why one-third to one-half of other noninsulin antidiabetic drug use was not concomitant with metformin use despite guidelines recommending that metformin be continued when other agents are added to treatment.

Prescribing pattern of glucose lowering drugs in the United Kingdom in the last decade: a focus on the effects of safety warnings about rosiglitazone

AIM

In the last decade, new glucose lowering drugs (GLDs) have been launched, and also several warnings regarding their safety. The cardiovascular safety of thiazolidinediones (TZD) has been questioned. We analyzed the prescription pattern of GLDs from 2000 to November 2009 in the United Kingdom (UK) using the THIN database with special focus on the effects of the safety warnings about rosiglitazone issued in May 2007 and January 2008.

METHODS

Annual prevalence and incidence of GLD prescriptions were measured. For TZD, the monthly prevalence and incidence of prescription were calculated from May 2006 to January 2009. The switching pattern around the FDA alert and the characteristics of subjects starting treatment with TZD before and after the alerts were observed.

RESULTS

The prevalence of prescriptions of GLDs increased during the 10 year period, metformin increasing more than three times. Rosiglitazone prevalence showed an increased trend until May 2007, (2.3/1000 person-years) and decreased thereafter (January 2009: 1.1/1000 person-years). The use of pioglitazone increased surpassing rosiglitazone from April 2008 onwards. The incidence of rosiglitazone use decreased sharply after May 2007 (0.8/1000 person-years). The prevalence of use of other therapies remained rather stable from 2000 to 2007 but increased afterwards. After May 2007, rosiglitazone users were increasingly switched to pioglitazone. There was an increased proportion of new users of pioglitazone with cardiovascular risk after the alerts.

CONCLUSIONS

The prescription of GLDs in the UK has increased in the last decade. For TZDs, it changed after May 2007 as well as the characteristics of the subjects treated with them.