Developing a Risk Assessment for a Policy 0070 submission must follow many regulations and standards. There are two main types of risk assessments, a qualitative risk assessment and a quantitative risk assessment.
Qualitative Risk Assessment
A qualitative risk assessment seeks to use ‘relative’ levels of risk based on factors such as prevalence of the indicated disease or symptom and number of patients participating in the trial. Using this approach is currently simpler to perform; however, because this approach only evaluates ‘relative’ level of risk, it is entirely possible that the risk of reidentification for certain patients can still be very high, based on the level of information provided by their quasi-identifiers such as age, height, weight, location, etc.
Quantitative Risk Assessment
A quantitative risk assessment, in short, seeks to assign a ‘risk-of-reidentification’ probability for each patient, and for an overall study or submission. This requires more work and mathematical modeling, but is the best way to ensure that the risk-of-reidentification for individual patients is minimized.
Three risk models are often associated with risk: Prosecutor risk, Marketer Risk, Journalist Risk. We will discuss these various models in other articles.
For technical information on quasi identifiers and risk models, please see this link for more information on some standards in the data anonymization community: CDISC Standards Workshop
Also, if you’re interested in software automation tools that help with risk assessment, anonymization, and redaction, please check out the latest ARARA Technology by Real Life Sciences: Clinical Trial Disclosure Deidentification, Redaction and Anonymization Automation