|Products and Services:||240|
|Articles and publications:||101|
|Tenders & Vacancies:||1|
Importance of Data Integrity
The scientific community has been a witness to some of the worst tragedies in the history of clinical trials data integrity. From the year 2015 to date, The Journal of the American Medical Association (JAMA) and the JAMA network journals have published at least 18 notices citing concern over data error and/or falsification of data.1 For instance, the trials conducted by a Japanese anesthesiologist and researcher, to treat post-operative nausea and vomiting, were reviewed by the Japanese Society of Anaesthesiologists (JSA) in the year 2012 to find startling revelations. The data obtained from the trials were either totally fabricated or fraudulent and approximately 210 papers published by the anesthesiologist had falsified data.2 Lapses in data integrity causes significant loss of revenue with the direct costs estimated to be close to 525,000 US dollars while indirect costs amounting to approximately 1.3 million US dollars.3
Such scientific misconduct served as a wake-up call to tighten regulations and laws to monitor drug development and drug use. Scientists acknowledged the need for data integrity at every stage to safeguard human subjects, starting from pre-clinical development to pharmacovigilance.
What is Data integrity?
Data integrity is defined as paper-based or electronic data that is complete, accurate, consistent, and reliable through its lifecycle from the time of data creation, archival, scanning, retention, and destruction.4 the updated International Council for Harmonization Guideline for Good Clinical Practice (ICH GCP E6 [R2]) reiterates the need for data integrity as well as the importance of monitoring clinical data throughout the study.
The United States Food and Drug Administration (FDA) uses the ALCOA acronym to define expectations with respect to data integrity.
Data Compliance Issues
The FDA issued Good manufacturing practices (GMP) warning letters to various countries outside the United States (US) citing compliance issues over data integrity.
Data Integrity Check Posts
Data integrity can be monitored by keeping a check on the following areas:
• Source Data Verification (SDV)
• Data access and control
• Training of personnel involved in data collection such as investigators, data processors, analysts, site staff, and report writers
• Data monitoring: On-site, centralized, and risk-based monitoring
• Clinical trial quality assurance units (QAU): Some sponsors set up internal QAUs or external QAUs with a contract research organization (CRO) to ascertain trial compliance with standard operating procedures (SOPs) and FDA regulations. QAUs also eliminates the risk of internal bias.
• Clinical trial audits
Strict adherence to good documentation practices (GDP) in clinical trial records is a way to ensure data integrity. GDP should be followed for paper records as well as electronic records and signatures. Equally important is the need to retain and organize essential documents required before the start of a clinical trial, during the trial, and after completion or termination of a trial. The collection of essential documents that are kept at the sponsor site and investigator site is called the clinical trial master file (TMF). TMF plays a major role in facilitating trial conduct and management thereby allowing for data integrity and GCP compliance at all stages of the clinical trial. The TMF is the document that is reviewed during an audit or inspection.
Many pharmaceutical companies are now moving towards electronic TMF (e-TMF) for easier management of large and complex clinical trials that involve numerous departments or CROs.
Data Access and Control
It is necessary to exercise caution while handling data from clinical trials. Confidentiality of data should be maintained during all the phases of a clinical trial including interim data results.9 the ability to tamper with data such as changing, deleting, or falsifying data should be restricted by clearly demarcating roles. This also prevents potential conflict of interest between similar roles that may hamper data integrity.
The National Institute of Health (NIH) states that only voting members of the Data and Safety Monitoring Board (DSMB) should be permitted to look at the interim analyses results unless circumstances make it necessary to share data, such as in the case of serious adverse events.9 In addition, the DMC members should not have any conflict of interest that would influence the outcome data. The FDA has also recommended the use of an “independent statistician” model to analyze interim data who is independent of the principal investigator and trial sponsor and reports unbiased results to the DMC.
It is necessary to set up an independent data monitoring committee (DMC) that prioritizes the safety and interests of enrolled subjects and scrutinizes the authenticity of data as well as the clinical trial conduct.
On-site monitoring: is carried out to trace any discrepancy between the source data and data entered. It is also particularly useful to see if the site staff is familiar with the study document and if the staff has demonstrated accountability to carry out the trial ethically and responsibly.
Centralized risk-based approach: ICH GCP E6 (R2) emphasizes the need for centralized monitoring to reduce the number of trial visits by the clinical monitor and to allow for remote spotting of reliable and unreliable data by statisticians or other data management staff.
Risk-based monitoring: The sponsor company is required to develop a robust risk management plan to prevent or mitigate any risk to human subjects by overseeing trial conduct and monitoring data quality across trial sites.
Data Integrity Audits 12
• Specific audits lookout for any data or metadata that previously went unnoticed such as deleted or unchecked, misused, orphaned, or reprocessed data.
• The entire data lifecycle should be subjected to scrutiny by all departments involved in the trial such as, but not limited to, data management, safety, quality risk management, and statisticians for compliance issues in areas of data management and data access control. 4,8
• Unnecessary incentivization for speedy results or data from high-risk phase II trials should be closely monitored for unscrupulous activities.
• Weightage should be given to raw data and not summary reports and results should be backtracked for any compliance issues.
To avoid huge financial repercussions and loss of business, sponsor companies and CROs should lay sufficient emphasis on maintaining data integrity at every step of the clinical study for its completeness, accuracy, and consistency.