Risk of Anaphylaxis Among New Users of GLP-1 Receptor Agonists: A Cohort Study

OBJECTIVE

To assess risk of anaphylaxis among patients with type 2 diabetes mellitus who are initiating therapy with a glucagon-like peptide 1 receptor agonist (GLP-1 RA), with a focus on those starting lixisenatide therapy.

RESEARCH DESIGN AND METHODS

A cohort study was conducted in three large, U.S. claims databases (2017-2021). Adult (aged ≥18 years) new users of a GLP-1 RA who had type 2 diabetes mellitus and ≥6 months enrollment in the database before GLP-1 RA initiation (start of follow-up) were included. GLP-1 RAs evaluated were lixisenatide, an insulin glargine/lixisenatide fixed-ratio combination (FRC), exenatide, liraglutide or insulin degludec/liraglutide FRC, dulaglutide, and semaglutide (injectable and oral). The first anaphylaxis event during follow-up was identified using a validated algorithm. Incidence rates (IRs) and 95% CIs were calculated within each medication cohort. The unadjusted IR ratio (IRR) comparing anaphylaxis rates in the lixisenatide cohort with all other GLP-1 RAs combined was analyzed post hoc.

RESULTS

There were 696,089 new users with 456,612 person-years of exposure to GLP-1 RAs. Baseline demographics, comorbidities, and use of other prescription medications in the 6 months before the index date were similar across medication cohorts. IRs (95% CIs) per 10,000 person-years were 1.0 (0.0-5.6) for lixisenatide, 6.0 (3.6-9.4) for exenatide, 5.1 (3.7-7.0) for liraglutide, 3.9 (3.1-4.8) for dulaglutide, and 3.6 (2.6-4.9) for semaglutide. The IRR (95% CI) for the anaphylaxis rate for the lixisenatide cohort compared with the pooled other GLP-1 RA cohort was 0.24 (0.01-1.35).

CONCLUSIONS

Anaphylaxis is rare with GLP-1 RAs. Lixisenatide is unlikely to confer higher risk of anaphylaxis than other GLP-1 RAs.

Suitability of Preference Methods Across the Medical Product Lifecycle: A Multicriteria Decision Analysis

OBJECTIVES

This study aimed to understand the importance of criteria describing methods (eg, duration, costs, validity, and outcomes) according to decision makers for each decision point in the medical product lifecycle (MPLC) and to determine the suitability of a discrete choice experiment, swing weighting, probabilistic threshold technique, and best-worst scale cases 1 and 2 at each decision point in the MPLC.

METHODS

Applying multicriteria decision analysis, an online survey was sent to MPLC decision makers (ie, industry, regulatory, and health technology assessment representatives). They ranked and weighted 19 methods criteria from an existing performance matrix about their respective decisions across the MPLC. All criteria were given a relative weight based on the ranking and rating in the survey after which an overall suitability score was calculated for each preference elicitation method per decision point. Sensitivity analyses were conducted to reflect uncertainty in the performance matrix.

RESULTS

Fifty-nine industry, 29 regulatory, and 5 health technology assessment representatives completed the surveys. Overall, "estimating trade-offs between treatment characteristics" and "estimating weights for treatment characteristics" were highly important criteria throughout all MPLC decision points, whereas other criteria were most important only for specific MPLC stages. Swing weighting and probabilistic threshold technique received significantly higher suitability scores across decision points than other methods. Sensitivity analyses showed substantial impact of uncertainty in the performance matrix.

CONCLUSION

Although discrete choice experiment is the most applied preference elicitation method, other methods should also be considered to address the needs of decision makers. Development of evidence-based guidance documents for designing, conducting, and analyzing such methods could enhance their use.

Incidence of diabetic ketoacidosis and its trends in patients with type 1 diabetes mellitus identified using a U.S. claims database, 2007-2019

Diabetic ketoacidosis (DKA) is a common complication of type 1 diabetes mellitus (T1DM). We found that the incidence of DKA was 55.5 per 1000 person-years in US commercially insured patients with T1DM; age-sex-standardized incidence decreased at an average annual rate of 6.1% in 2018-2019 after a steady increase since 2011.

Development and validation of a predictive model algorithm to identify anaphylaxis in adults with type 2 diabetes in U.S. administrative claims data

PURPOSE

To use medical record adjudication and predictive modeling methods to develop and validate an algorithm to identify anaphylaxis among adults with type 2 diabetes (T2D) in administrative claims.

METHODS

A conventional screening algorithm that prioritized sensitivity to identify potential anaphylaxis cases was developed and consisted of diagnosis codes for anaphylaxis or relevant signs and symptoms. This algorithm was applied to adults with T2D in the HealthCore Integrated Research Database (HIRD) from 2016 to 2018. Clinical experts adjudicated anaphylaxis case status from redacted medical records. We used confirmed case status as an outcome for predictive models developed using lasso regression with 10-fold cross-validation to identify predictors and estimate the probability of confirmed anaphylaxis.

RESULTS

Clinical adjudicators reviewed medical records with sufficient information from 272 adults identified by the anaphylaxis screening algorithm, which had an estimated Positive Predictive Value (PPV) of 65% (95% confidence interval [CI]: 60%-71%). The predictive model algorithm had a c-statistic of 0.95. The model's probability threshold of 0.60 excluded 89% (84/94) of false positives identified by the screening algorithm, with a PPV of 94% (95% CI: 91%-98%). The model excluded very few true positives (15 of 178), and identified 92% (95% CI: 87%-96%) of the cases selected by the screening algorithm.

CONCLUSIONS

Predictive modeling techniques yielded an accurate algorithm with high PPV and sensitivity for identifying anaphylaxis in administrative claims. This algorithm could be considered in future safety studies using similar claims data to reduce potential outcome misclassification.

Risk of interstitial lung disease in patients treated for atrial fibrillation with dronedarone versus other antiarrhythmics

Purpose

To compare risks of interstitial lung disease (ILD) between patients treated with dronedarone versus other antiarrhythmics.

Methods

Parallel retrospective cohort studies were conducted in the United States Department of Defense Military Health System database (DoD) and the HealthCore Integrated Research Database (HIRD). Study patients were treated for atrial fibrillation (AF) with dronedarone, amiodarone, sotalol, or flecainide. Propensity score matching was employed to create analysis cohorts balanced on baseline variables considered potential confounders of treatment decisions. The study period of July 20, 2008 through September 30, 2014 included a 1‐year baseline and minimum 6 months of follow‐up, for patients with drugs dispensed between July 20, 2009 and March 31, 2014. Suspect ILD outcomes were reviewed by independent adjudicators. Cox proportional hazards regression compared risk of confirmed ILD between dronedarone and each comparator cohort. A sensitivity analysis examined the effect of broadening the outcome definition.

Results

A total 72 ILD cases (52 DoD; 20 HIRD) were confirmed among 27 892 patients. ILD risk was significantly higher among amiodarone than dronedarone initiators in DoD (HR = 2.5; 95% CI = 1.1–5.3, p = 0.02). No difference was detected in HIRD (HR = 1.0; 95% CI = 0.4–2.4). Corresponding risks in sotalol and flecainide exposure groups did not differ significantly from dronedarone in either database.

Conclusions

ILD risk among AF patients initiated on dronedarone therapy was comparable to or lower than that of amiodarone initiators, and similar to that of new sotalol or flecainide users. This finding suggests that elevated ILD risk associated with amiodarone does not necessarily extend to dronedarone or other antiarrhythmic drugs.

Evaluating the Risk of Digitalis Intoxication Associated With Concomitant Use of Dronedarone and Digoxin Using Real-World Data

PURPOSE

Dronedarone may increase digoxin plasma levels through inhibition of P-glycoprotein. Using real-world data, we evaluated the risk of digitalis intoxication in concomitant users of dronedarone and digoxin compared digoxin-alone users.

METHODS

We used the Clinformatics DataMart, a US claims database, to identify adult patients with atrial fibrillation (AF) or atrial flutter (AFL) who concomitantly used dronedarone and digoxin and those who used digoxin alone between July 2009 and March 2016. Digitalis intoxication during follow-up until March 2016 was ascertained using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) and International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM). Adjusted hazard ratios (HR) for digitalis intoxication in concomitant users versus users of digoxin alone were estimated, controlling for age, sex, cohort entry year, number of medical encounters for AF or AFL, history of congestive heart failure, diabetes, hypertension, stroke, myocardial infarction, renal failure, use of drugs interacting with digoxin, and digoxin dose.

FINDINGS

Overall, 524 concomitant users and 32,459 users of digoxin alone were identified, among which 3 and 301 events of digitalis intoxication occurred during follow-up, respectively. Incidence rates were 17.25 and 9.17 cases per 1000 person-years, respectively. The adjusted HR for digitalis intoxication in concomitant users versus users of digoxin alone was 1.56 (95% CI, 0.50-4.88; P = 0.45). When digitalis intoxication was defined by ICD-9-CM and ICD-10-CM codes accompanied by laboratory testing for digoxin/digitoxin or hospitalization within 30 days, no events occurred in the concomitant users and 40 events occurred in the users of digoxin alone (incidence rate of 1.22 cases per 1000 person-years).

IMPLICATIONS

Concomitant use of dronedarone and digoxin was uncommon in this study, and no significant increase in the risk of digitalis intoxication with concomitant use was found.

Performance assessment of different machine learning approaches in predicting diabetic ketoacidosis in adults with type 1 diabetes using electronic health records data

Purpose

To assess the performance of different machine learning (ML) approaches in identifying risk factors for diabetic ketoacidosis (DKA) and predicting DKA.

Methods

This study applied flexible ML (XGBoost, distributed random forest [DRF] and feedforward network) and conventional ML approaches (logistic regression and least absolute shrinkage and selection operator [LASSO]) to 3400 DKA cases and 11 780 controls nested in adults with type 1 diabetes identified from Optum® de‐identified Electronic Health Record dataset (2007–2018). Area under the curve (AUC), accuracy, sensitivity and specificity were computed using fivefold cross validation, and their 95% confidence intervals (CI) were established using 1000 bootstrap samples. The importance of predictors was compared across these models.

Results

In the training set, XGBoost and feedforward network yielded higher AUC values (0.89 and 0.86, respectively) than logistic regression (0.83), LASSO (0.83) and DRF (0.81). However, the AUC values were similar (0.82) among these approaches in the test set (95% CI range, 0.80–0.84). While the accuracy values >0.8 and the specificity values >0.9 for all models, the sensitivity values were only 0.4. The differences in these metrics across these models were minimal in the test set. All approaches selected some known risk factors for DKA as the top 10 features. XGBoost and DRF included more laboratory measurements or vital signs compared with conventional ML approaches, while feedforward network included more social demographics.

Conclusions

In our empirical study, all ML approaches demonstrated similar performance, and identified overlapping, but different, top 10 predictors. The difference in selected top predictors needs further research.

Structured benefit-risk evaluation for medicinal products: review of quantitative benefit-risk assessment findings in the literature

A favorable benefit–risk profile remains an essential requirement for marketing authorization of medicinal drugs and devices. Furthermore, prior subjective, implicit and inconsistent ad hoc benefit–risk assessment methods have rightly evolved towards more systematic, explicit or “structured” approaches. Contemporary structured benefit–risk evaluation aims at providing an objective assessment of the benefit–risk profile of medicinal products and a higher transparency for decision making purposes. The use of a descriptive framework should be the preferred starting point for a structured benefit–risk assessment. In support of more precise assessments, quantitative and semi-quantitative methodologies have been developed and utilized to complement descriptive or qualitative frameworks in order to facilitate the structured evaluation of the benefit–risk profile of medicinal products. In addition, quantitative structured benefit–risk analysis allows integration of patient preference data. Collecting patient perspectives throughout the medical product development process has become increasingly important and key to the regulatory decision-making process. Both industry and regulatory authorities increasingly rely on descriptive structured benefit–risk evaluation and frameworks in drug, vaccine and device evaluation and comparison. Although varied qualitative methods are more commonplace, quantitative approaches have recently been emphasized. However, it is unclear how frequently these quantitative frameworks have been used by pharmaceutical companies to support submission dossiers for drug approvals or to respond to the health authorities’ requests. The objective of this study has been to identify and review, for the first time, currently available, published, structured, quantitative benefit–risk evaluations which may have informed health care professionals and/or payor as well as contributed to decision making purposes in the regulatory setting for drug, vaccine and/or device approval.

Age- and sex-specific incidence of non-traumatic lower limb amputation in patients with type 2 diabetes mellitus in a U.S. claims database

AIM

To estimate age- and sex-specific incidence rates (IRs) of non-traumatic lower limb amputations (LLA) in patients with type 2 diabetes mellitus (T2DM) using a claims database from the United States (US).

METHODS

Patients with T2DM 18 years and older were identified using the Truven Health MarketScan database from January 1, 2007 to September 30, 2018. The overall and age- and sex-specific IRs of all non-traumatic LLA, minor LLA (amputation at or below the ankle), and major LLA (amputation above ankle) were calculated.

RESULTS

Among the 6,117,981 patients with T2DM, 14,627 LLA events occurred (minor LLA; 72.8%; major LLA: 27.2%). The IRs (95% CI) of all LLA, minor LLA, and major LLA per 1000 person-years or PY were 0.86 (0.85, 0.88), 0.63 (0.62, 0.64), and 0.23 (0.23, 0.24), respectively. The IR (95% CI) of all LLA per 1000 PY in males was higher compared to females [1.24 (1.22, 1.26) vs. 0.46 (0.45, 0.48)]. The incidence of all LLA increased with an increasing age (highest IR in age-group of ≥80 years).

CONCLUSIONS

This study identified males and older patients with T2DM at higher risk of developing LLA in the US, warranting further exploration of risk factors of LLA in these subgroups.

Design, Conduct, and Use of Patient Preference Studies in the Medical Product Life Cycle: A Multi-Method Study

Objectives: To investigate stakeholder perspectives on how patient preference studies (PPS) should be designed and conducted to allow for inclusion of patient preferences in decision-making along the medical product life cycle (MPLC), and how patient preferences can be used in such decision-making.

Methods: Two literature reviews and semi-structured interviews (n = 143) with healthcare stakeholders in Europe and the US were conducted; results of these informed the design of focus group guides. Eight focus groups were conducted with European patients, industry representatives and regulators, and with US regulators and European/Canadian health technology assessment (HTA) representatives. Focus groups were analyzed thematically using NVivo.

Results: Stakeholder perspectives on how PPS should be designed and conducted were as follows: 1) study design should be informed by the research questions and patient population; 2) preferred treatment attributes and levels, as well as trade-offs among attributes and levels should be investigated; 3) the patient sample and method should match the MPLC phase; 4) different stakeholders should collaborate; and 5) results from PPS should be shared with relevant stakeholders. The value of patient preferences in decision-making was found to increase with the level of patient preference sensitivity of decisions on medical products. Stakeholders mentioned that patient preferences are hardly used in current decision-making. Potential applications for patient preferences across industry, regulatory and HTA processes were identified. Four applications seemed most promising for systematic integration of patient preferences: 1) benefit-risk assessment by industry and regulators at the marketing-authorization phase; 2) assessment of major contribution to patient care by European regulators; 3) cost-effectiveness analysis; and 4) multi criteria decision analysis in HTA.

Conclusions: The value of patient preferences for decision-making depends on the level of collaboration across stakeholders; the match between the research question, MPLC phase, sample, and preference method used in PPS; and the sensitivity of the decision regarding a medical product to patient preferences. Promising applications for patient preferences should be further explored with stakeholders to optimize their inclusion in decision-making.

Opportunities and challenges for the inclusion of patient preferences in the medical product life cycle: a systematic review

BACKGROUND

The inclusion of patient preferences (PP) in the medical product life cycle is a topic of growing interest to stakeholders such as academics, Health Technology Assessment (HTA) bodies, reimbursement agencies, industry, patients, physicians and regulators. This review aimed to understand the potential roles, reasons for using PP and the expectations, concerns and requirements associated with PP in industry processes, regulatory benefit-risk assessment (BRA) and marketing authorization (MA), and HTA and reimbursement decision-making.

METHODS

A systematic review of peer-reviewed and grey literature published between January 2011 and March 2018 was performed. Consulted databases were EconLit, Embase, Guidelines International Network, PsycINFO and PubMed. A two-step strategy was used to select literature. Literature was analyzed using NVivo (QSR international).

RESULTS

From 1015 initially identified documents, 72 were included. Most were written from an academic perspective (61%) and focused on PP in BRA/MA and/or HTA/reimbursement (73%). Using PP to improve understanding of patients' valuations of treatment outcomes, patients' benefit-risk trade-offs and preference heterogeneity were roles identified in all three decision-making contexts. Reasons for using PP relate to the unique insights and position of patients and the positive effect of including PP on the quality of the decision-making process. Concerns shared across decision-making contexts included methodological questions concerning the validity, reliability and cognitive burden of preference methods. In order to use PP, general, operational and quality requirements were identified, including recognition of the importance of PP and ensuring patient understanding in PP studies.

CONCLUSIONS

Despite the array of opportunities and added value of using PP throughout the different steps of the MPLC identified in this review, their inclusion in decision-making is hampered by methodological challenges and lack of specific guidance on how to tackle these challenges when undertaking PP studies. To support the development of such guidance, more best practice PP studies and PP studies investigating the methodological issues identified in this review are critically needed.

Patient Preferences in the Medical Product Life Cycle: What do Stakeholders Think? Semi-Structured Qualitative Interviews in Europe and the USA

BACKGROUND

Patient preferences (PP), which are investigated in PP studies using qualitative or quantitative methods, are a growing area of interest to the following stakeholders involved in the medical product lifecycle: academics, health technology assessment bodies, payers, industry, patients, physicians, and regulators. However, the use of PP in decisions along the medical product lifecycle remains limited. As the adoption of PP heavily relies on these stakeholders, knowledge of their perceptions of PP is critical.

OBJECTIVE

This study aimed to characterize stakeholders' attitudes, needs, and concerns with respect to PP in decision making along the medical product lifecycle.

METHODS

Semi-structured interviews (n = 143) were conducted with academics (n = 24), health technology assessment/payer representatives (n = 24), industry representatives (n = 24), patients, caregivers and patient representatives (n = 24), physicians (n = 24), and regulators (n = 23) from seven European countries and the USA. Interviews were conducted between April and August 2017. The framework method was used to organize the data and identify themes and key findings in each interviewed stakeholder group.

RESULTS

Interviewees reported being unfamiliar (43%), moderately familiar (42%), or very familiar (15%) with preference methods and studies. Interviewees across stakeholder groups generally supported the idea of using PP in the medical product lifecycle but expressed mixed opinions about the feasibility and impact of using PP in decision making. Interviewees from all stakeholder groups stressed the importance of increasing stakeholders' understanding of the concept of PP and preference methods and ensuring patients' understanding of the questions asked in PP studies. Key concerns and needs in each interviewed stakeholder group were as follows: (1) academics: investigating the validity, reliability, reproducibility, and generalizability of preference methods; (2) health technology assessment/payer representatives: developing quality criteria for evaluating PP studies and gaining insights into how to weigh them in reimbursement/payer decision making; (3) industry representatives: obtaining guidance on PP studies and recognition on the importance of PP from decision makers; (4) patients, caregivers, and patient representatives: providing an incentive and adequate information towards patients when participating in PP studies; (5) physicians: avoiding bias as a result of commercial agendas in PP studies and clarifying how to deal with subjective and emotional elements when measuring PP; and (6) regulators: avoiding the misuse of PP study results to overrule the traditional efficacy and safety criteria used for marketing authorization and obtaining robust PP study results.

CONCLUSIONS

Despite the interest all interviewed stakeholder groups reported in PP, the effective use of PP in decision making across the medical product lifecycle is currently hampered by a lack of standardization and consensus on how to both measure and use PP.

Factors and Situations Affecting the Value of Patient Preference Studies: Semi-Structured Interviews in Europe and the US

Objectives: Patient preference information (PPI) is gaining recognition among the pharmaceutical industry, regulatory authorities, and health technology assessment (HTA) bodies/payers for use in assessments and decision-making along the medical product lifecycle (MPLC). This study aimed to identify factors and situations that influence the value of patient preference studies (PPS) in decision-making along the MPLC according to different stakeholders.

Methods: Semi-structured interviews (n = 143) were conducted with six different stakeholder groups (physicians, academics, industry representatives, regulators, HTA/payer representatives, and a combined group of patients, caregivers, and patient representatives) from seven European countries (the United Kingdom, Sweden, Italy, Romania, Germany, France, and the Netherlands) and the United States. Framework analysis was performed using NVivo 11 software.

Results: Fifteen factors affecting the value of PPS in the MPLC were identified. These are related to: study organization (expertise, financial resources, study duration, ethics and good practices, patient centeredness), study design (examining patient and/or other preferences, ensuring representativeness, matching method to research question, matching method to MPLC stage, validity and reliability, cognitive burden, patient education, attribute development), and study conduct (patients’ ability/willingness to participate and preference heterogeneity). Three types of situations affecting the use of PPS results were identified (stakeholder acceptance, market situations, and clinical situations).

Conclusion: The factors and situation types affecting the value of PPS, as identified in this study, need to be considered when designing and conducting PPS in order to promote the integration of PPI into decision-making along the MPLC.

Benefit-risk evaluation: the past, present and future

In the last two decades there has been a shift in the approach to evaluating the benefit-risk (BR) profiles of medicinal products from an unstructured, subjective, and inconsistent, to a more structured and objective, process. This article describes that shift from a historical perspective; the past, the present, and the future, and highlights key events that played critical roles in changing the field.

Comparison of text processing methods in social media-based signal detection

PURPOSE

Adverse event (AE) identification in social media (SM) can be performed using various types of natural language processing (NLP) and machine learning (ML). These methods can be categorized by complexity and precision level. Co-occurrence-based ML methods are rather basic, as they identify simultaneous appearance of drugs and clinical events in a single post. In contrast, statistical learning methods involve more complex NLP and identify drugs, events, and associations between them. We aimed to compare the ability of co-occurrence and NLP to identify AEs and signals of disproportionate reporting (SDR) in patient-generated SM. We also examined the performance of lift in SM-based signal detection (SD).

METHODS

Our examination was performed in a corpus of SM posts crawled from open online patient forums and communities, using the spontaneously reported VigiBase data as reference data set.

RESULTS

We found that co-occurrence and NLP produce AEs, which are 57% and 93% consistent with VigiBase AEs, respectively. Among the SDRs identified both in SM and in VigiBase, up to 55.3% were identified earlier in co-occurrence, and up to 32.1% were identified earlier in NLP-processed SM. Using lift in SM SD provided performance similar to frequentist methods, both in co-occurrence and in NLP-processed AEs.

CONCLUSION

Our results indicate that using SM as a data source complementary to traditional pharmacovigilance sources should be considered further. Various levels of SM processing may be considered, depending on the preferred policies and tolerance for false-positive to false-negative balance in routine pharmacovigilance processes.

Proximal HbA1C Level and First Hypoglycemia Hospitalization in Adults With Incident Type 2 Diabetes

CONTEXT

Hemoglobin A1C (HbA1C) is an important predictor of severe hypoglycemia.

OBJECTIVE

To determine the association of proximal HbA1C level with first hypoglycemia hospitalization (HH) in adults with incident type 2 diabetes (T2D).

DESIGN, SETTING, AND PARTICIPANTS

A nested case-control study was designed using linked data from the Clinical Practice Research Datalink and Hospital Episode Statistics in England in 1997 to 2014. The first hypoglycemia event as primary diagnosis for hospitalization after T2D diagnosis was identified. Proximal HbA1C was measured within 90 days before the first HH.

MAIN OUTCOME MEASURE

OR for developing HH.

RESULTS

The association of proximal HbA1C level with first HH was similar between HbA1C levels of 6.0% (OR, 1.54; 95% CI, 1.12 to 2.11) and 9.0% [1.48 (1.01 to 2.17)] compared with the reference HbA1C level of 7.0%. For proximal HbA1C level of 4.0% to 6.5%, every additional 0.5% increase in HbA1C was associated with lower first HH risk, with ORs (95% CI) ranging between 0.37 (0.20 to 0.67) and 0.86 (0.76 to 0.98). For proximal HbA1C level of 8.0% to 11.5%, every additional 0.5% increase in HbA1C was associated with higher first HH risk, with ORs (95% CI) ranging between 1.16 (1.04 to 1.29) and 1.34 (1.18 to 1.52). The U-shaped association between proximal HbA1C level and first HH did not exist among current sulfonylurea users but persisted among current insulin users (Pinteraction = 0.002). Among current noninsulin nonsulfonylurea users who had a first HH, 78% took insulin or sulfonylureas before the HH.

CONCLUSIONS

Having either poor or near-normal HbA1C was associated with a higher risk of first HH within 3 months in T2D.

Methods for exploring and eliciting patient preferences in the medical product lifecycle: a literature review

Preference studies are becoming increasingly important within the medical product decision-making context. Currently, there is limited understanding of the range of methods to gain insights into patient preferences. We developed a compendium and taxonomy of preference exploration (qualitative) and elicitation (quantitative) methods by conducting a systematic literature review to identify these methods. This review was followed by analyzing prior preference method reviews, to cross-validate our results, and consulting intercontinental experts, to confirm our outcomes. This resulted in the identification of 32 unique preference methods. The developed compendium and taxonomy can serve as an important resource for assessing these methods and helping to determine which are most appropriate for different research questions at varying points in the medical product lifecycle.

Web-Based Signal Detection Using Medical Forums Data in France: Comparative Analysis

BACKGROUND

While traditional signal detection methods in pharmacovigilance are based on spontaneous reports, the use of social media is emerging. The potential strength of Web-based data relies on their volume and real-time availability, allowing early detection of signals of disproportionate reporting (SDRs).

OBJECTIVE

This study aimed (1) to assess the consistency of SDRs detected from patients' medical forums in France compared with those detected from the traditional reporting systems and (2) to assess the ability of SDRs in identifying earlier than the traditional reporting systems.

METHODS

Messages posted on patients' forums between 2005 and 2015 were used. We retained 8 disproportionality definitions. Comparison of SDRs from the forums with SDRs detected in VigiBase was done by describing the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), accuracy, receiver operating characteristics curve, and the area under the curve (AUC). The time difference in months between the detection dates of SDRs from the forums and VigiBase was provided.

RESULTS

The comparison analysis showed that the sensitivity ranged from 29% to 50.6%, the specificity from 86.1% to 95.5%, the PPV from 51.2% to 75.4%, the NPV from 68.5% to 91.6%, and the accuracy from 68% to 87.7%. The AUC reached 0.85 when using the metric empirical Bayes geometric mean. Up to 38% (12/32) of the SDRs were detected earlier in the forums than that in VigiBase.

CONCLUSIONS

The specificity, PPV, and NPV were high. The overall performance was good, showing that data from medical forums may be a valuable source for signal detection. In total, up to 38% (12/32) of the SDRs could have been detected earlier, thus, ensuring the increased safety of patients. Further enhancements are needed to investigate the reliability and validation of patients' medical forums worldwide, the extension of this analysis to all possible drugs or at least to a wider selection of drugs, as well as to further assess performance against established signals.

Factors and situations influencing the value of patient preference studies along the medical product lifecycle: a literature review

Industry, regulators, health technology assessment (HTA) bodies, and payers are exploring the use of patient preferences in their decision-making processes. In general, experience in conducting and assessing patient preference studies is limited. Here, we performed a systematic literature search and review to identify factors and situations influencing the value of patient preference studies, as well as applications throughout the medical product lifecyle. Factors and situations identified in 113 publications related to the organization, design, and conduct of studies, and to communication and use of results. Although current use of patient preferences is limited, we identified possible applications in discovery, clinical development, marketing authorization, HTA, and postmarketing phases.

Trends in Hospital Admission for Diabetic Ketoacidosis in Adults With Type 1 and Type 2 Diabetes in England, 1998-2013: A Retrospective Cohort Study

OBJECTIVE

This study determined trends in hospital admission for diabetic ketoacidosis (DKA) in adults with type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) from 1998 to 2013 in England.

RESEARCH DESIGN AND METHODS

The study population included 23,246 adults with T1DM and 241,441 adults with T2DM from the Clinical Practice Research Datalink (CPRD) and Hospital Episode Statistics (HES). All hospital admissions for DKA as the primary diagnosis from 1998 to 2013 were identified. Trends in hospital admission for DKA in incidence, length of hospital stay, 30-day all-cause readmission rate, and 30-day and 1-year all-cause mortality rates were determined using joinpoint regression, negative binomial regression, and logistic regression models.

RESULTS

For T1DM, the incidence of hospital admission for DKA increased between 1998 and 2007 and remained static until 2013. The incidence in 2013 was higher than that in 1998 (incidence rate ratio 1.53 [95% CI 1.09-2.16]). For T2DM, the incidence increased 4.24% (2.82-5.69) annually between 1998 and 2013. The length of hospital stay decreased over time for both diabetes types (P ≤ 0.0004). Adults with T1DM were more likely to be discharged within 2 days compared with adults with T2DM (odds ratio [OR] 1.28 [1.07-1.53]). The 30-day readmission rate was higher in T1DM than in T2DM (OR 1.61 [1.04-2.50]) but remained unchanged for both diabetes types over time. Trends in 30-day and 1-year all-cause mortality rates were also stable, with no difference by diabetes type.

CONCLUSIONS

In the previous two decades in England, hospitalization for DKA increased in adults with T1DM and in those with T2DM, and associated health care performance did not improve except decreased length of hospital stay.

HbA1C variability and hypoglycemia hospitalization in adults with type 1 and type 2 diabetes: A nested case-control study

AIMS

To determine association between HbA_1C variability and hypoglycemia requiring hospitalization (HH) in adults with type 1 diabetes (T1D) and type 2 diabetes (T2D).

METHODS

Using nested case-control design in electronic health record data in England, one case with first or recurrent HH was matched to one control who had not experienced HH in incident T1D and T2D adults. HbA_1C variability was determined by standard deviation of ≥3 HbA_1C results. Conditional logistic models were applied to determine association of HbA_1C variability with first and recurrent HH.

RESULTS

In T1D, every 1.0% increase in HbA_1C variability was associated with 90% higher first HH risk (95% CI, 1.25-2.89) and 392% higher recurrent HH risk (95% CI, 1.17-20.61). In T2D, a 1.0% increase in HbA_1C variability was associated with 556% higher first HH risk (95% CI, 3.88-11.08) and 573% higher recurrent HH risk (95% CI,1.59-28.51). In T2D for first HH, the association was the strongest in non-insulin non-sulfonylurea users (P<0.0001); for recurrent HH, the association was stronger in insulin users than sulfonylurea users (P=0.07). The HbA_1C variability-HH association was stronger in more recent years in T2D (P≤0.004).

CONCLUSIONS

HbA_1C variability is a strong predictor for HH in T1D and T2D.

Incidence and Trends in Hypoglycemia Hospitalization in Adults With Type 1 and Type 2 Diabetes in England, 1998-2013: A Retrospective Cohort Study

OBJECTIVE

To determine trends in hospitalization for hypoglycemia in adults with type 1 diabetes mellitus (T1DM) and type 2 diabetes mellitus (T2DM) in England.

RESEARCH DESIGN AND METHODS

Adults with T1DM or T2DM were identified from 398 of the 684 practices within the Clinical Practice Research Datalink, for which linkage to the Hospital Episode Statistics was possible. Hypoglycemia as the primary reason for hospitalization between 1998 and 2013 was extracted. Trends were estimated using joinpoint regression models for adults with T1DM, young and middle-aged adults with T2DM (18-64 years), and elderly adults with T2DM (≥65 years), respectively.

RESULTS

Among 23,246 adults with T1DM, 1,591 hypoglycemia hospitalizations occurred during 121,262 person-years. Among 241,441 adults with T2DM, 3,738 hypoglycemia hospitalizations occurred during 1,344,818 person-years. In adults with T1DM, the incidence increased 3.74% (95% CI 1.70-5.83) annually from 1998 to 2013. In young and middle-aged adults with T2DM, the annual incidence increase was 4.12% (0.61-7.75) from 1998 to 2013. In elderly adults with T2DM, the incidence increased 8.59% (5.76-11.50) annually from 1998 to 2009, and decreased 8.05% (-14.48 to -1.13) annually from 2009 to 2013, but the incidence was still higher in 2013 than 1998 (adjusted rate ratio 3.01 [1.76-5.14]). Trends in HbA_1c level did not parallel trends of hypoglycemia hospitalization for both diabetes types. A possible reason for declined hypoglycemia trend in 2009-2013 in elderly adults with T2DM may be continuously decreased sulfonylurea use after 2009, which was not seen in young and middle-aged adults with T2DM.

CONCLUSIONS

Hypoglycemia requiring hospitalization has been an increasing burden in adults with T1DM and T2DM in England in the previous two decades, with the exception of the decline in elderly adults with T2DM starting in 2009.

Dietary intake and risk of non-severe hypoglycemia in adolescents with type 1 diabetes

AIMS

To determine the association between dietary intake and risk of non-severe hypoglycemia in adolescents with type 1 diabetes.

METHODS

Type 1 adolescents from a randomized trial wore a blinded continuous glucose monitoring (CGM) system at baseline for one week in free-living conditions. Dietary intake was calculated as the average from two 24-h dietary recalls. Non-severe hypoglycemia was defined as having blood glucose <70mg/dL for ≥10min but not requiring external assistance, categorized as daytime and nocturnal (11PM-7AM). Data were analyzed using logistic regression models.

RESULTS

Among 98 participants with 14,277h of CGM data, 70 had daytime hypoglycemia, 66 had nocturnal hypoglycemia, 55 had both, and 17 had neither. Soluble fiber and protein intake were positively associated with both daytime and nocturnal hypoglycemia. Glycemic index, monounsaturated fat, and polyunsaturated fat were negatively associated with daytime hypoglycemia only. Adjusting for total daily insulin dose per kilogram eliminated all associations.

CONCLUSIONS

Dietary intake was differentially associated with daytime and nocturnal hypoglycemia. Over 80% of type 1 adolescents had hypoglycemia in a week, which may be attributed to the mismatch between optimal insulin dose needed for each meal and actually delivered insulin dose without considering quality of carbohydrate and nutrients beyond carbohydrate.

CLINICAL TRIAL REGISTRATION

ClinicalTrials.gov identifier: NCT01286350.

The US Food and Drug Administration's Risk Evaluation and Mitigation Strategy (REMS) Program - Current Status and Future Direction

The US Food and Drug Administration (FDA) Amendments Act of 2007 granted the FDA new authorities to enhance drug safety by requiring application holders to submit a proposed Risk Evaluation and Mitigation Strategy (REMS). A REMS is a required risk management plan that uses tools beyond the package insert. REMS elements may include a medication guide and patient package insert for patients and a communication plan focused on health care professionals. Elements to assure safe use (ETASUs) are put in place to mitigate a specific known serious risk when other less restrictive elements of a REMS are not sufficient to mitigate such risk. An implementation system is required for an REMS that includes the ETASUs. With approximately eight years of experience with REMS programs, many health care settings have created systems to manage REMS and also to integrate REMS into their practice settings. At the same time, there are issues associated with the development and implementation of REMS. In 2011, FDA created the REMS Integration Initiative to develop guidance on how to apply statutory criteria to determine when a REMS is required, to improve standardization and assessment of REMS, and to improve integration of REMS into the existing healthcare system. A key component of the REMS Integration Initiative is stakeholder outreach to better understand how existing REMS programs are working and to identify opportunities for improvement. This review attempts to share our company's experience with the REMS program, and to provide updates on FDA's efforts to improve REMS communication, to standardize REMS process, to reduce REMS program burdens and to build a common REMS platform.

Recommendations for benefit-risk assessment methodologies and visual representations

PURPOSE

The purpose of this study is to draw on the practical experience from the PROTECT BR case studies and make recommendations regarding the application of a number of methodologies and visual representations for benefit-risk assessment.

METHODS

Eight case studies based on the benefit-risk balance of real medicines were used to test various methodologies that had been identified from the literature as having potential applications in benefit-risk assessment. Recommendations were drawn up based on the results of the case studies.

RESULTS

A general pathway through the case studies was evident, with various classes of methodologies having roles to play at different stages. Descriptive and quantitative frameworks were widely used throughout to structure problems, with other methods such as metrics, estimation techniques and elicitation techniques providing ways to incorporate technical or numerical data from various sources. Similarly, tree diagrams and effects tables were universally adopted, with other visualisations available to suit specific methodologies or tasks as required. Every assessment was found to follow five broad stages: (i) Planning, (ii) Evidence gathering and data preparation, (iii) Analysis, (iv) Exploration and (v) Conclusion and dissemination.

CONCLUSIONS

Adopting formal, structured approaches to benefit-risk assessment was feasible in real-world problems and facilitated clear, transparent decision-making. Prior to this work, no extensive practical application and appraisal of methodologies had been conducted using real-world case examples, leaving users with limited knowledge of their usefulness in the real world. The practical guidance provided here takes us one step closer to a harmonised approach to benefit-risk assessment from multiple perspectives.

Risk of Wernicke's encephalopathy and cardiac disorders in patients with myeloproliferative neoplasm

The objectives of the study were to estimate the incidence of Wernicke's encephalopathy (WE) and cardiac disorders among patients with myeloproliferative neoplasms (MPN), and compare it with those without MPN. A total of 39,761 MPN patients were identified from the US Marketscan database. Approximately 27% of them were 65+ years of age, and 51% were male. Patients with MPN had higher rates of WE, compared to those without MPN (MPN vs. non-MPN: 1.09 vs. 0.39/1000 person-year, adjusted HR=2.19, 95%CI 1.43-3.34). Patients with MPN also had higher rates of cardiac events (congestive heart failure: MPN vs. non-MPN: 9.27 vs. 3.70/1000 person-year, adjusted HR=1.64, 95%CI 1.42-1.88; acute myocardial infarction: MPN vs. non-MPN: 10.45 vs. 5.02/1000 person-year, adjusted HR=1.44, 95%CI 1.27, 1.63; cardiac arrhythmia: MPN vs. non-MPN: 106.5 vs. 53.9/1000 person-year, adjusted HR=1.42, 95%CI 1.36-1.48). Physicians who care for patients with MPN should be aware of increased risk of WE and cardiac disorders in this population.

Benefit-risk assessment in a post-market setting: a case study integrating real-life experience into benefit-risk methodology

PURPOSE

Difficulties may be encountered when undertaking a benefit-risk assessment for an older product with well-established use but with a benefit-risk balance that may have changed over time. This case study investigates this specific situation by applying a formal benefit-risk framework to assess the benefit-risk balance of warfarin for primary prevention of patients with atrial fibrillation.

METHODS

We used the qualitative framework BRAT as the starting point of the benefit-risk analysis, bringing together the relevant available evidence. We explored the use of a quantitative method (stochastic multi-criteria acceptability analysis) to demonstrate how uncertainties and preferences on multiple criteria can be integrated into a single measure to reduce cognitive burden and increase transparency in decision making.

RESULTS

Our benefit-risk model found that warfarin is favourable compared with placebo for the primary prevention of stroke in patients with atrial fibrillation. This favourable benefit-risk balance is fairly robust to differences in preferences. The probability of a favourable benefit-risk for warfarin against placebo is high (0.99) in our model despite the high uncertainty of randomised clinical trial data. In this case study, we identified major challenges related to the identification of relevant benefit-risk criteria and taking into account the diversity and quality of evidence available to inform the benefit-risk assessment.

CONCLUSION

The main challenges in applying formal methods for medical benefit-risk assessment for a marketed drug are related to outcome definitions and data availability. Data exist from many different sources (both randomised clinical trials and observational studies), and the variability in the studies is large.

Risk of Cardiovascular Events, Stroke, Congestive Heart Failure, Interstitial Lung Disease, and Acute Liver Injury: Dronedarone versus Amiodarone and Other Antiarrhythmics

No published studies have evaluated the risks of cardiovascular (CV) events, stroke, congestive heart failure (CHF), interstitial lung disease (ILD), and severe acute liver injury (ALI) related to antiarrhythmics treatment in real-world clinical practice setting. We examined the relationship between the above events and the selected antiarrhythmics in the real-world setting in the US. Using a retrospective cohort design, the hazard ratios of the outcome events were analyzed from 10,455 adult patients with a diagnosis of atrial fibrillation/atrial flutter and a new treatment with dronedarone (comparison drug), amiodarone, sotalol, flecainide, or propafenone between 07/20/2009 and 12/31/2010 from the Clinformatics Data MartTM database. The patients were followed until: 1) switch to another antiarrhythmic drug, 2) occurrence of the outcome event, 3) end of enrollment, or 4) end of the study period, whichever occurred first. No significant differences were observed in the hazard ratios of the outcome events between dronedarone, amiodarone, and the other antiarrhythmics, except that amiodarone was associated with a higher risk of CV events (adjusted HR = 1.7, 95%CI: 1.1-2.4) and stroke (adjusted HR = 2.0, 95%CI: 1.33.2), compared to dronedarone, especially amongst patients without a CHF history (adjusted HR = 2.4, 95%CI: 1.4-3.8 and 2.2, 95%CI: 1.23.9). A higher risk of CHF was also associated with amiodarone in patients without history of CHF at baseline (adjusted HR = 2.7, 95%CI: 2.03.6). In this real-world investigation, no difference in risk was observed between dronedarone, sotalol, and propafenone initiators for CV events, stroke, CHF, ILD, and ALI. Amiodarone was associated with higher risks of CV events, stroke, and CHF than dronedarone in patients without a CHF history, indicating dronedarone could be an alternative therapy option with lower risk of CV events than amiodarone for the above patients.

Association between body mass index and suicide, and suicide attempt among British adults: the health improvement network database

OBJECTIVE

To examine the associations between body mass index (BMI) and incidence rate (IR) of suicide attempt and suicide.

DESIGN AND METHODS

849,434 British adults were identified from The Health Improvement Network (THIN) database between January 2000 and October 2007. BMI was categorized into six levels: <18.5 (underweight), 18.5-24.9 (normal weight), 25.0-29.9 (overweight), 30.0-34.9, 35.0-39.9, and ≥40 (obese levels I-III).

RESULTS

We identified 3,111 suicide attempts by Read codes and 75 suicides with medical records. The overall IR of suicide attempt was 82.2 cases per 100,000 person-years. The IR decreased with BMI in men with depression (471.3-166.0 cases per 100,000 person-years, P for trend = 0.02) and in men without depression (241.5-58.0 cases per 100,000 person-years, P for trend < 0.0001). In women with depression, an L-shaped relationship was observed, that is, a higher rate in underweight group when compared with reference group (503.2 vs. 282.7 per 100,000 person-years) and no significant differences in others (231.8-195.5 cases per 100,000 person-years). In women without depression, the IR was U-shaped with BMI (125.2 in underweight, 68.6 in reference, and 48.5-79.9 cases in overweight and obese I-III groups per 100,000 person-years, P for trend < 0.0001). The above trends remained after adjustment for the covariates. Regarding suicide, the overall IR was 2.0 cases per 100,000 person-years, which tended to decrease with BMI (P = 0.14).

CONCLUSIONS

We concluded an inverse linear association between BMI and suicide attempt among men, an L-shaped association in nondepressive women, and a U-shaped association in depressive women were observed. The study also suggested an inverse linear tendency between BMI and suicide.

Evaluation of dronedarone use in the US patient population between 2009 and 2010: a descriptive study using a claims database

BACKGROUND

The utilization pattern of dronedarone was unknown, especially regarding prescribers' compliance with the product's prescribing information (PI) following its availability and the implementation of the Food and Drug Administration-approved risk evaluation and mitigation strategy for the drug in the United States.

OBJECTIVE

This study was designed to evaluate the dronedarone prescribers' adherence to PI regarding the following contraindications: (1) patients with heart failure (HF) with a recent decompensation requiring hospitalization or referral to a specialist, (2) concomitant use of potent CYP3A4 inhibitors, and (3) concomitant use of QT-prolonging drugs.

METHODS

Patients prescribed dronedarone between July 2009 and August 2010 were identified through LabRx. The following rates surrounding dronedarone use were examined: (1) atrial fibrillation or atrial flutter in the past year, (2) worsening or hospitalization for HF within the month before prescription, and (3) concomitant prescription of potent CYP3A4 inhibitors and concomitant prescription of QT-prolonging drugs within the following month.

RESULTS

A total of 4595 dronedarone prescriptions were filled by 1820 patients. More than 94% of the participants had ≥1 diagnosis of atrial fibrillation or atrial flutter in the previous year. Worsening of or hospitalization for HF was found in 61 (3.4%) patients within the month before receiving dronedarone, including 18 patients with HF as the primary cause for hospitalization. Potent CYP3A4 inhibitors were prescribed to 10 (0.6%) patients within a month following dronedarone initiation, 6 of whom received them for topical use only. QT-prolonging drugs were prescribed to 67 (3.7%) patients within a month following dronedarone initiation, among which >90% were other antiarrhythmics.

CONCLUSIONS

Dronedarone was used mostly in compliance with PI and risk evaluation and mitigation strategy in the studied population. In the LabRx database, dronedarone was commonly dispensed to patients with cardiovascular risk factors and rarely dispensed to patients with contraindications such as worsening HF or hospitalization for HF or with concomitant prescriptions of potent CYP3A4 inhibitors, QT-prolonging drugs, or both.

Predicting the risk of end-stage renal disease in the population-based setting: a retrospective case-control study

Background

Previous studies of predictors of end-stage renal disease (ESRD) have limitations: (1) some focused on patients with clinically recognized chronic kidney disease (CKD); (2) others identified population-based patients who developed ESRD, but lacked earlier baseline clinical measures to predict ESRD. Our study was designed to address these limitations and to identify the strength and precision of characteristics that might predict ESRD pragmatically for decision-makers--as measured by the onset of renal replacement therapy (RRT).

Methods

We conducted a population-based, retrospective case-control study of patients who developed ESRD and started RRT. We conducted the study in a health maintenance organization, Kaiser Permanente Northwest (KPNW). The case-control study was nested within the adult population of KPNW members who were enrolled during 1999, the baseline period. Cases and their matched controls were identified from January 2000 through December 2004. We evaluated baseline clinical characteristics measured during routine care by calculating the adjusted odds ratios and their 95% confidence intervals after controlling for matching characteristics: age, sex, and year.

Results

The rate of RRT in the cohort from which we sampled was 58 per 100,000 person-years (95% CI, 53 to 64). After excluding patients with missing data, we analyzed 350 cases and 2,114 controls. We identified the following characteristics that predicted ESRD with odds ratios ≥ 2.0: eGFR<60 mL/min/1.73 m² (OR = 20.5; 95% CI, 11.2 to 37.3), positive test for proteinuria (OR = 5.0; 95% CI, 3.5 to 7.1), hypertension (OR = 4.5; 95% CI, 2.5 to 8.0), gout/positive test for uric acid (OR = 2.5; 95% CI, 1.8 to 3.5), peripheral vascular disease (OR = 2.2; 95% CI, 1.4 to 3.6), congestive heart failure (OR = 2.1; 95% CI, 1.4 to 3.3), and diabetes (OR = 2.1; 95% CI, 1.5 to 2.9).

Conclusions

The clinical characteristics needed to predict ESRD--for example, to develop a population-based, prognostic risk score--were often documented during routine care years before patients developed ESRD and required RRT.

Incidence rates of heart failure, stroke, and acute myocardial infarction among Type 2 diabetic patients using insulin glargine and other insulin

PURPOSE

The aim of this study was to compare incidence rates of heart failure, stroke, and acute myocardial infarction (AMI) in Type 2 diabetic patients using different types of insulin.

METHODS

Included were patients with a diagnosis of Type 2 diabetes and at least one insulin prescription from May 2001 to July 2007. Incidence rate ratios (RRs) of heart failure, stroke, and AMI were estimated using Poisson regression with adjustment for age, gender, history of hypertension, dyslipidemia history, days supply, and duration of diabetes.

RESULTS

Incidence rates of heart failure, stroke, and AMI in the insulin glargine group were 306.9 (95%CI: [278.9, 334.8]), 174.8 (95%CI: [153.7, 195.8]), and 105.2 (95%CI: [88.9, 121.5]) cases per 10,000 person-years, respectively. After adjustment for covariates, the incidence rates of CVD events in the insulin glargine were comparable to those in the other long/intermediate acting insulin group (reference), except for AMI, which tended to be lower in the insulin glargine group (RR = 0.81, 95%CI: [0.65, 1.02]). Using the same reference, the incidence rate of stroke was higher in patients taking rapid/short acting insulin, premixed insulin, or mixed use of insulin except insulin glargine (RR = 1.20, 95%CI: [1.04, 1.40]).

CONCLUSION

This study suggested that insulin glargine use might be associated with a lower risk of AMI, compared to the other long/intermediate acting insulin use, and that insulin regimen of rapid/short acting insulin, premixed insulin, or mixed use of insulin except insulin glargine was associated with a higher risk of stroke using the same reference.

Impact of overweight and obesity on hospitalization: race and gender differences

OBJECTIVE

To examine associations between weight status and number of all-cause and cause-specific hospitalizations overall, and by race and gender.

DESIGN

Longitudinal cohort study.

SUBJECTS

White and black adults (n=15 355) from the Atherosclerosis Risk in Communities Study who were normal weight (body mass index: >or=18.5 to <25.0 kg m(-2); n=4997), overweight (>or=25.0 to <30.0 kg m(-2); n=6100), or obese (>or=30.0 kg m(-2); n=4258) at baseline.

MEASUREMENTS

Information on hospitalizations was collected using community and cohort surveillance methods. Negative binomial models adjusted for race, gender, field center, age, physical activity, education level, smoking status, alcoholic beverage consumption and health insurance at baseline. Adjusted numbers of hospitalizations were calculated after setting covariates to the mean value (for continuous variables) or to the average distribution (for categorical variables) observed in the entire cohort and are expressed as the number of hospitalizations per 1000 adults followed over a period of 13 years.

RESULTS

The covariate-adjusted average number of all-cause hospitalizations was 1316 per 1000 normal weight, 1543 per 1000 overweight and 2025 per 1000 obese. Normal weight women had significantly fewer hospitalizations than normal weight men (1173 versus 1515 per 1000), but the increase associated with being obese on the number of all-cause hospitalizations was larger in women than men (791 versus 589 per 1000). There was no significant difference detected between the number of hospitalizations in normal weight whites and blacks, and the increase in hospitalizations with overweight or obesity was also not different. Effects of weight status on several primary causes of hospitalization differed by gender and race group.

CONCLUSION

Our work suggests that obesity prevention may reduce hospitalizations, a major component of rising healthcare costs. The impact of successful obesity prevention is likely to be larger in women than in men, and similar in blacks and whites.

The incidence rate of seizures in relation to BMI in UK adults

A retrospective cohort study using the data from The Health Improvement Network (THIN) database in the United Kingdom was conducted to examine the incidence rates of seizures across different BMI levels in the adult population aged > or = 18 years. Poisson regression was used to examine the relationship between BMI and seizures. The overall incidence rate of seizures was found to be 31.2 cases per 100,000 person-years. The incidence rate of seizures (cases per 100,000 person-years) in obese patients (BMI > or = 30 kg/m2) was 34.8 (95% confidence interval (CI), 23.1, 46.4), comparable to that in patients with normal weight (BMI between 18.5 and 24.9 kg/m2) (35.8, 95% CI (26.6, 44.9)). In contrast, underweight patients (< 18.5 kg/m2) or extremely obese (> or = 40 kg/m2) patients tended to have higher incidence rates than those with normal weight. After adjustment for age, gender, and smoking status, compared to patients with normal weight, those who were underweight or extremely obese had a rate ratio (RR) for seizures of 1.6 (95% CI (0.7, 3.8)) and 1.7 (95% CI (0.7, 3.9)), respectively. To date, we have not found any study that examines the associations between BMI or obesity and seizures. In this study, the incidence rates of seizures in the extremely obese and underweight patients tended to be higher than that in the normal-weight patients.

Adverse events with disease modifying antirheumatic drugs (DMARD): a cohort study of leflunomide compared with other DMARD

OBJECTIVE

To determine and compare the incidence of serious adverse events (AE) during treatment of rheumatoid arthritis (RA) with disease modifying antirheumatic drugs (DMARD), focusing on leflunomide (LEF).

METHODS

A retrospective cohort study of a large US insurance claims database was performed. Study groups were patients with RA classified by DMARD exposure as either no-DMARD therapy, single-agent DMARD (monotherapy), or combination-DMARD therapy. Specific DMARD examined were leflunomide (LEF) and methotrexate (MTX), compared to other DMARD (penicillamine, hydroxychloroquine, sulfasalazine, gold, etanercept, infliximab) and no DMARD (nonsteroidal antiinflammatory drugs, COX-2 inhibitors). All AE reported were considered endpoints; primary endpoints included hepatic, dermatologic, hematologic, infectious, respiratory, hypertension, and pancreatitis AE.

RESULTS

The 40,594 RA patients of the study period (September 1998 to December 2000) accumulated 83,143 person-years (PY) of followup. Followup for each of the groups was: DMARD-monotherapy, 46,054 PY (55% of total); combination-DMARD, 25,830 PY (14%); and no-DMARD, 11,259 PY (14%). The incidence rate of all AE combined was significantly lower for LEF monotherapy (94 events/1000 PY) than MTX (145 events/1000 PY), other DMARD (143 events/1000 PY), or no DMARD (383 events/1000 PY) (p < 0.001 for all comparisons). The "all-AE" rates during combination therapy with LEF + MTX (43/1000 PY) and LEF + other DMARD (59/1000 PY) were lower than the "all-AE" rate for DMARD + MTX (70/1000 PY; p = 0.002). LEF monotherapy had the lowest rate of hepatic events in the DMARD monotherapy groups.

CONCLUSION

The rates of AE in the LEF group, alone and combined with MTX, were generally lower than or comparable to the AE rates seen with MTX and other agents.

Benefit-risk analysis: a proposal using quantitative methods

This is the first part of a two-article series which will introduce the theory and practice of a proposed set of quantitative methods for benefit-risk analysis. Adjustments to number-needed-to-treat (NNT) analysis and a new method, minimum clinical efficacy (MCE) analysis are presented and critically discussed. The goal of these methods is to condense into a summary metric the benefit-risk profile of a product so that manufacturers, regulators, clinicians and patients can better understand and participate in risk management. A second article will present examples of these methods.