Comparative effectiveness without head-to-head trials: a method for matching-adjusted indirect comparisons applied to psoriasis treatment with adalimumab or etanercept.

The absence of head-to-head trials is a common challenge in comparative effectiveness research and health technology assessment. Indirect cross-trial treatment comparisons are possible, but can be biased by cross-trial differences in patient characteristics. Using only published aggregate data, adjustment for such biases may be impossible. Although individual patient data (IPD) would permit adjustment, they are rarely available for all trials. However, many researchers have the opportunity to access IPD for trials of one treatment, a new drug for example, but only aggregate data for trials of comparator treatments. We propose a method that leverages all available data in this setting by adjusting average patient characteristics in trials with IPD to match those reported for trials without IPD. Treatment outcomes, including continuous, categorical and censored time-to-event outcomes, can then be compared across balanced trial populations. The proposed method is illustrated by a comparison of adalimumab and etanercept for the treatment of psoriasis. IPD from trials of adalimumab versus placebo (n = 1025) were re-weighted to match the average baseline characteristics reported for a trial of etanercept versus placebo (n = 330). Re-weighting was based on the estimated propensity of enrolment in the adalimumab versus etanercept trials. Before matching, patients in the adalimumab trials had lower mean age, greater prevalence of psoriatic arthritis, less prior use of systemic treatment or phototherapy, and a smaller mean percentage of body surface area affected than patients in the etanercept trial. After matching, these and all other available baseline characteristics were well balanced across trials. Symptom improvements of ≥75% and ≥90% (as measured by the Psoriasis Area and Severity Index [PASI] score at week 12) were experienced by an additional 17.2% and 14.8% of adalimumab-treated patients compared with the matched etanercept-treated patients (respectively, both p < 0.001). Mean percentage PASI score improvements from baseline were also greater for adalimumab than for etanercept at weeks 4, 8 and 12 (all p < 0.05). Matching adjustment ensured that this indirect comparison was not biased by differences in mean baseline characteristics across trials, supporting the conclusion that adalimumab was associated with significantly greater symptom reduction than etanercept for the treatment of moderate to severe psoriasis.

Detection of spironolactone-associated hyperkalaemia following the Randomized Aldactone Evaluation Study (RALES).

INTRODUCTION: A population-based analysis has suggested that the publication of the RALES (Randomized Aldactone Evaluation Study) in late 1999 was associated with both the wider use of spironolactone to treat heart failure and a corresponding increase in hyperkalaemia-associated morbidity and mortality in patients also being treated with ACE inhibitors. OBJECTIVES: To gain further insight into the reporting of spironolactone-associated hyperkalaemia in an independent dataset by analysing the spontaneous reporting experience in relation to the publication of RALES, and to determine whether the implementation of a commonly used data mining algorithm (DMA) might have directed the attention of safety reviewers to the spironolactone/hyperkalaemia association in advance of epidemiological findings. METHODS: We calculated the reporting rate of spironolactone-associated hyperkalaemia per 1,000 reports per year from 1970 through to the end of 2005 by identifying relevant cases in the US FDA Adverse Event Reporting System. We did this for reports of spironolactone-associated hyperkalaemia (where spironolactone was listed as a suspect drug) and according to whether the reports listed an ACE inhibitor as a co-suspect or concomitant medication. A further statistical analysis of the overall reporting of spironolactone (suspect drug)-associated hyperkalaemia was also performed. We also performed 3-dimensional (3-D; drug-drug-event) disproportionality analyses using a DMA known as the multi-item gamma-Poisson shrinker, which allows the calculation and display of a 3-D disproportionality metric known as the 'interaction signal score' (INTSS). This metric is a measure of the strength of a higher order reporting relationship of a triplet (i.e. drug-drug-event) association above and beyond what would be expected from the largest disproportionalities associated with the individual 2-way associations. RESULTS: Visual inspection of a graph of the reporting frequency of spironolactone (suspect drug)-associated hyperkalaemia per 1,000 reports was highly suggestive of a change point. The t-test on the arcsine-transformed data showed a significant difference in reporting of spironolactone-hyperkalaemia combination through 1999 compared with 2000 onwards (p < 0.001). When examining the reporting time trends according to the presence or absence of an ACE inhibitor, the change point seemed to be mostly attributable to an increase in the number of spironolactone (suspect drug)-associated hyperkalaemia reports with ACE inhibitors listed as a co-suspect drug. No obvious change points in INTSSs for spironolactone-ACE inhibitor-hyperkalaemia reports were observed. DISCUSSION: Although we could not pinpoint the relative contribution of many possible artifacts in the reporting process, as well as increased drug exposure, increased adverse event incidence and/or a change in patient monitoring practices, to our findings, we observed a notable change in reporting frequency of spironolactone-associated hyperkalaemia in temporal proximity to the publication of RALES. Evidence of this was provided by a trend analysis depicted in a simple graph that was supported by statistical analysis. The observed trend was in large part due to increased reporting of spironolactone-associated hyperkalaemia with reported co-medication with ACE inhibitors. CONCLUSION: These findings are consistent with those originally reported in an epidemiological analysis. In this retrospective exercise, a simple graph was more illuminating than more complex data mining analyses.

A comparison of pioglitazone and rosiglitazone for hospitalization for acute myocardial infarction in type 2 diabetes.

BACKGROUND: Recent studies have raised concerns about potential increased cardiovascular (CV) risk in type 2 diabetes patients treated with some peroxisome proliferator-activated receptor gamma (PPAR-gamma) agonists. OBJECTIVE: To ascertain the risk of hospitalization for acute myocardial infarction (AMI) in type 2 diabetes patients treated with pioglitazone relative to rosiglitazone. METHODOLOGY: Using data covering 2003-2006 from a large health care insurer in the US, a retrospective cohort study was conducted in patients who initiated treatment with pioglitazone or rosiglitazone. The hazard ratio (HR) of incident hospitalization for AMI after initiation of treatment with these drugs was estimated from multivariate Cox's proportional hazards survival analysis; similarly, the HR was ascertained for hospitalization for the composite endpoint of AMI or coronary revascularization (CR). RESULTS: A total of 29 911 eligible patients were identified in the database; 14 807 in the pioglitazone and 15 104 in the rosiglitazone group. Baseline demographics, medical history, and dispensed medications were generally well balanced between groups. The unadjusted HR for hospitalization for AMI was 0.82, 95%CI: 0.67-1.01. After adjustment for baseline covariates the HR was 0.78, 95%CI: 0.63-0.96. The adjusted HR for the composite of AMI or CR was 0.85, 95%CI: 0.75-0.98. CONCLUSION: This retrospective cohort study showed that pioglitazone, in comparison with rosiglitazone, is associated with a 22% relative risk reduction of hospitalization for AMI in patients with type 2 diabetes.

Illusions of objectivity and a recommendation for reporting data mining results.

OBJECTIVE: Data mining algorithms (DMAs) are being applied to spontaneous reporting system (SRS) databases in the hope of obtaining timely insights into post-licensure safety data. Some DMAs have been characterized as "objective" screening tools. However, there are numerous available modifiable configuration parameters to choose from, including choice of vendor, that may affect results. Our objective is to compare the data mining results on pre-selected drug-event combinations (DECs) between two commonly used software programs using similar protocols. METHODS: Two DMAs, using three thresholds, were retrospectively applied to the USFDA safety database through Q2 2005 to a set of eight pre-selected DECs. RESULTS: Differences between the two vendors were found for the number of cases associated with a signal of disproportionate reporting (SDR), first year of SDRs, and the magnitude of the SDR scores for the selected DECs. These were deemed to be potentially significant for 45.8% (11/24) of the data points. CONCLUSION: The observed differences between vendors could partially be explained by their differing methods of data cleaning and transformation as well as by the specific features of individual algorithms. The choices of vendors and available data mining configurations maximize the exploratory capacity of data mining, but they also raise questions about the claimed objectivity of data mining results and can make data mining exercises susceptible to confirmation bias given the exploratory nature of data mining in pharmacovigilance. When reporting results, the vendor and all data mining configuration details should be specified.

Reports of hyperkalemia after publication of RALES--a pharmacovigilance study.

PURPOSE: A population-based study and anecdotal reports have indicated that the publication of the Randomized Aldactone Evaluation Study (RALES) was associated with not merely a broader use of spironolactone in the treatment of heart failure, but also with a coinciding sharp increase in hyperkalemia-associated morbidity/mortality in patients also being treated with ACE-inhibitors. Data mining algorithms (DMAs) are being applied to spontaneous reporting system (SRS) databases in hopes of obtaining early warnings/additional insights into post-licensure safety data. We applied two DMAs (i.e. multi-item gamma Poisson shrinker [MGPS] and proportional reporting ratios [PRRs]) to spontaneous reporting system (SRS) data to determine if these DMAs could have provided an earlier indication of a possible hyperkalemia safety issue. METHODS: MGPS and PRRs were retrospectively applied to US FDA-AERS, an SRS database. Year-by-year analysis and analysis of increasing cumulative time intervals were performed on cases in which both spironolactone and hyperkalemia and possibly related cardiac events had been reported. RESULTS: Neither of the DMAs initially provided a compelling signal of disproportionate reporting (SDR) for hyperkalemia after publication of RALES. However, using events consistent with clinical sequelae of hyperkalemia (e.g,. sudden death), SDRs were identified with PRRs. CONCLUSIONS: The quality and usefulness of data mining analysis is highly situation dependent and may vary with the knowledge and experience of the drug safety reviewer. Our analysis suggests that contemporary DMAs may have significant limitations in detecting increased frequency of labeled events in real-life prospective pharmacovigilance. There is a paucity of research in this area and we recommend further research for new approaches to detecting increased frequency of labeled events.

The role of data mining in pharmacovigilance.

A principle concern of pharmacovigilance is the timely detection of adverse drug reactions that are novel by virtue of their clinical nature, severity and/or frequency. The cornerstone of this process is the scientific acumen of the pharmacovigilance domain expert. There is understandably an interest in developing database screening tools to assist human reviewers in identifying associations worthy of further investigation (i.e., signals) embedded within a database consisting largely of background 'noise' containing reports of no substantial public health significance. Data mining algorithms are, therefore, being developed, tested and/or used by health authorities, pharmaceutical companies and academic researchers. After a focused review of postapproval drug safety signal detection, the authors explain how the currently used algorithms work and address key questions related to their validation, comparative performance, deployment in naturalistic pharmacovigilance settings, limitations and potential for misuse. Suggestions for further research and development are offered.