Adverse Event Data

In the FDA s Guidance for Industry E6 Good Clinical Practice Consolidated Guidance, an adverse event is defined as follows  

Any untoward medical occurrence in a patient or clinical investigation subject administered a pharmaceutical product and that does not necessarily have a causal relationship with this treatment. An AE can therefore be any unfavorable and unintended sign (including an abnormal laboratory finding), symptom, or disease temporally associated with the use of a medicinal (investigational) product, whether or not related to the medicinal (investigational) product. 

The adverse event form is fairly standard across clinical trials. The form consists of a list of events for which data are entered as free text and are later coded with a dictionary such as MedDRA and some associated event attribute variables. In just about any clinical trial, an adverse event form very similar to the following sample will be found. 

The adverse event form is a cornerstone of patient safety monitoring, and as such it contains very important data. There are several data issues for the statistical programmer to be concerned about here. 

In guidance document ICH E3, Structure and Content of Clinical Study Reports, the FDA defines treatment-emergent signs and symptoms (TESS) as events not seen at baseline and events that worsened even if present at baseline. As simple as that may sound, it can sometimes be quite difficult to program. The important data variables that come into play are dosing record dates and times, adverse event start and stop times, and adverse event severity. All of these data variables need to be completed accurately for TESS to be calculated properly. 

---

Reproducible capabilities. The data and analyses must be electronically accessible in real-time, easily reanalyzable, and easily reproduced, even years after the adverse event data were collected. 

Additional examples of variability in data collection (which, in turn, affects data interpretation) include questionnaires and physical exam forms. Questionnaires often utilize open-ended questions that allow great variability in the type and extent of adverse event information gathered. Physical exam forms—even when designed in a checklist format—may elicit variable collection of adverse event data what is a serious event to one clinician may not be serious to another. 

Traditional analytical methods make extensive use of computers, but typically these methods still require constant restructuring of the data and multiple analytical tools. This endless restructuring wastes time and productivity and also makes the analytical processes difficult to document, audit, and reproduce in real time. This situation also makes it difficult to reconstruct and update analyses in real time when new adverse event data become available or when new questions need to be asked. The application of comprehensive data standards allows the use of integrated, reusable software for analyzing adverse event data. This integration facilitates the reproducibility of the results. 

Demographics and Trial-Specific Baseline Data 27 Concomitant or Prior Medication Data 27 Medical History Data 29 Investigational Therapy Drug Log 30 Laboratory Data 31 Adverse Event Data 32 Endpoint/Event Assessment Data 35 Clinical Endpoint Committee (CEC) Data 36 Study Termination Data 37 Treatment Randomization Data 38 Quality-of-Life Data 40... 

Program 2.2 Summarizing Free-Text Adverse Event Data... 

In the end, because of the importance of the data, it is imperative that the entire adverse event form data are cleaned. Reconciling the adverse event data with other clinical data in the clinical data management system can be very difficult if the data management system does not provide variable keys for linking such data. Adverse event data fall into the safety area of statistical analyses and are considered an event from a CDISC perspective. 

Imagine you have a data set of adverse event data and a data set of concomitant medications, and you want to know if a concomitant medication was given to a patient during the time of the adverse event. The following program defines the two data sets and joins them with PROC SQL so that you get all medications taken during any specific... 

INPUT SAMPLE ADVERSE EVENT DATA WHERE SUBJECT = PATIENT ID AND ADVERSE EVENT = ADVERSE EVENT TEXT. data aes ... 

There are myriad ways in which adverse event data can be summarized. Adverse events are summarized by overall occurrence, by maximum severity, and by maximum... 

INPUT SAMPLE ADVERSE EVENT DATA. data ae ... 

KEEP ONLY LAST RECORD PER SUBJECT AT HIGHEST SEVERITY AS WE ONLY WANT TO COUNT A PATIENT ONCE AT MAX SEVERITY IF THEY HAD ANY ADVERSE EVENTS. data bysev set bysev ... 

Phase III Extended large-scale trials to obtain additional evidence of efficacy and safety, and definition of adverse effects. Humans exposed several hundred to several thousand Phase IV Post-marketing surveillance occurs after the chnical trials programme is complete. It is used to collect adverse event data from a large patient population. Humans exposed 10 000+... 

A number of variables influence the likelihood of an adverse event being reported. These include the length of time that a product has been marketed, the market share, experience and sophistication of the population using the product, and publicity about adverse events. Currently there is little incentive for health professional reporting of adverse events, which partially imderlies the problem with underreporting. Lack of exposure data and the issue of underreporting preclude estimation of incidence rates. Causality assessment is difficult or impossible because of the quality of the data received and the lack of a comparator (control) group. Finally, comparisons of product safety cannot be directly obtained from adverse event data. 

Anonymous Update Adverse event data and revised American Thoracic Society/CDC recommendations against the use of... 

Finished product manufacturing Commercial packaging Adverse event data management Validation services Auditing services... 

O Neill, R.T., 1987, Statistical analyses of adverse event data from clinical trials Special emphasis on serious events, Drug Information Journal, 21 9-20. 

Adverse event data from other sources including the world literature should be included. 

Pharmacovigilance systems capture, store, process, maintain, classify, and report adverse event data. Any snch systems generating reports for regnlatory authorities (e.g., expedited reports, periodic safety updates) and the interfaces into them from a variety of sources, should be validated. Specific considerations when validating these systems are ... 

Reconciliation of adverse event data from the clinical trial database, and other sources, with the pharmacovigilance database throngh electronic interfaces... 

A limited number of medicines are monitored at any one time on the IMMP. The list of medicines monitored is published in the Presenter Update and New Ethicals publications and is included on the Medsafe web site. In addition to reports of adverse events, data are collated from all prescriptions issued for these medicines in New Zealand, plus their total sales figures, so that an accurate assessment can be made of the causality and relevance of any events that occur. 

During clinical trials, investigators are instructed to collect all adverse events reported by patients enrolled in the study, which are tabulated. During final study reports or product marketing applications, adverse event data are analyzed and compared among treatment arms. Overall analyses of... 

The momentum of international harmonization efforts will continue to improve the ability with which adverse events are classified, analyzed and communicated. Paper-based system will be replaced by electronic transmission of adverse event information. The WHO monitoring program will extend the Bayesian artificial neural networks for analysis of the large amounts of adverse event data at its disposal.In the United States, efforts are underway at the FDA to improve the content and format of product prescribing information. 

Postmarketing surveillance occurs after the clinical trials programme is complete. It is used to collect adverse event data from a large patient population. 

An adverse event in a clinical study that is unexpected, unlabeled, and associated with the investigational drug must be reported to the FDA within 15 days. If the adverse event is life-threatening, the FDA must be notified within seven days. On a yearly basis adverse event data on the most frequent and serious adverse events are submitted to the IND along with updates on all investigations with the drug. 

Table 5.22 Pharmacokinetics, patient characteristics, and adverse event data for a new oncolytic drug.

---