Tags

, , ,

By: John C. Strong

john-strong_photoQuality risk management (QRM) is a central concept in quality by design, and the assessment of risk is a specific component of QRM that figures prominently in the development, new drug application submission, and subsequent lifecycle of a marketed drug.

Risk is usually described as the combination of the probability of a hazard occurring (O) and the severity (S) of the resulting harm. Often in pharmaceutical risk assessment, the detectability (D) of the hazard prior to the harm occurring is considered as well. A popular approach to account for these components of risk is the calculation of the so-called risk priority number (RPN), where numerical rankings ascribed to each of these risk components (e.g., 1–5 or 1–10 ) are multiplied together, i.e., RPN = S×O×D. The product S×O, coined the “criticality,” is also commonly employed, but this is lumped under the label of RPN for the purposes of our AAPS Newsmagazine article. RPN is typically used to triage risk mitigation activities or to compare against a threshold value as a trigger for taking mitigation action.

RPN remains a common approach today in the pharmaceutical industry for estimating risk despite its serious shortcomings for that purpose, as this article will describe. Curiously, lack of awareness of the weaknesses of this methodology in the pharmaceutical industry seems to persist despite numerous critical publications in peer -reviewed quality and statistical literature spanning more than two decades.

Our article discusses the pitfalls of RPN, why over -reliance on this method persists in pharmaceutical QRM, and alternative options. Pharmaceutical QRM can take advantage of well-established and more advanced mathematical approaches to reduce the risk inherent in the very methods popularly used to rate risk. As a first step, awareness is the goal.

Ultimately, by reading our article The Risk of Trusting Risk Priority Numbers in the AAPS Newsmagazine March issue, you’ll learn that risk assessment is not accomplished by simple recipes. Neither should it be done to merely “check the box.” Quantitative methods may seem to represent a daunting leap in complexity over the simple qualitative methods, but drug product development itself is a highly complex process—it stands to reason that this should be reflected in the sophistication of our risk assessments.

John Strong received his Ph.D. in biochemical engineering from University of Maryland Baltimore County and has nearly 20 years of experience at AbbVie in applying mathematical and engineering principles to pharmaceutical oral dosage development.