A lot of time, effort, and resources are applied across the industry in establishing and running Quality Management Systems (QMSs) for pharmaceutical products, whether in development stages or for commercially-licensed products. The past 15 years saw the most explosive growth in offerings for QMSs, both in terms of competitive vendors’ products, bolt-on modules, and various formats.

While the right QMS can make tracking and trending, collecting, executing, and maintaining of records, and more much easier, paper-based and hybrid systems still exist across the industry (usually in smaller companies, if budget is a major factor). Non-electronic Quality Management Systems (eQMS) are still perfectly viable alternatives, though throughout a developed drug’s lifecycle, the amount of paper-based records can be tremendous and unwieldy, hence the overall industry shift to eQMSs. A few important points and developments in the industry have been ongoing in the QMS space.

Ease of use via better UI

Some of the top providers of QMS platforms have been spending a lot of research time and money on improving the user interfaces (UI), to help deliver a better user experience (UX). A lot of the thinking in this regard comes from application development, where high levels of usability are critical to keep consumers engaged with a particular app product. Since the iPhone was launched in 2007, app developers have radically pioneered and changed the landscape of software development, with an eye to making interaction more graphical, more engaging, and more intuitive.

This design thinking has trickled-down into eQMS providers’ platforms, to offer more seamless experiences, across different software platforms and products, to make user engagement a major factor, allow for cross-talk between applications, easier document upload and download, and document sharing features. For example, drug development documentation in the form of certain clinical records can much more easily be imported, viewed, and shared than ever before, with easier-to-use navigation, and better records repositories that allow for easier use flexibility and tracking and trending of relevant metrics on the current status of the records. Similarly, essential quality records, such as change controls, deviation records, CAPAs, and more can be more easily created, manipulated, edited, shared, and tracked.

The benefit to this particular trend has been with a mind toward having greater visibility into real-time status of documents, and thus (hopefully) fewer and fewer critical records languishing in a stalled state, potentially leading to their aging out of compliance (exceeding due dates, and further penalizing existing backlogs of documents).

Regulatory implications of better QMS

Of course, with easier tracking and trending, comes much more visible status of the state of health of the QMS in question. Typically, with a few mouse clicks, a total export of certain record types can be executed for regulatory authority inspection scrutiny (or for quality auditing). So more sophisticated systems have created a co-evolution of the inspectors (or auditors) being more easily able to detect more issues, which leads to companies relying on QMSs more than in their paper-based past to direct their internal compliance activities in terms of audit and inspection preparation. So, the challenges of the past still exist—backlogs, incomplete records, missing threads—but the usability of the systems themselves to interrogate current compliance status has changed dramatically from a couple decades ago.

What remains true, however, is that a system is only as good as the antecedent activities that lead to the outputs; Making documents easier to share, review, and complete or approve doesn’t necessarily make the human participants better at their jobs of doing exactly those activities. But a good system goes a long way to reducing the effort and friction of accomplishing those tasks. That’s been a major thrust of QMS vendors in recent years.

The future of QMS is AI

What good would a future-looking trend be without mentioning the emerging presence of artificial intelligence (AI) and machine learning (ML) in the equation? Current advancements in the near future of QMS offerings is the application of AI/ML to real-time tracking and trending to alert users and management of slippage of compliance.

It works like this: The system learns about the current status by analyzing current documents, their status, and human activities of participation within the system. From there, departures from the baseline state can be easily flagged by the AI software to alert users and management that, for example, a greater proportion of documents are exceeding their due date(s), that the backlog is growing out-of-trend, and other important considerations of the health of the QMS.

This system-based intervention is designed to increase human participation in correcting and maintaining compliance with the regulations in regard to that company’s particular situation. It’s too early to tell what the longer-term industry wide implications will be of this, but initial trials show greater human involvement given AI-based “nudges” on behavior.

Further future-forward

Further enhancements are in development, such as greater connectivity with other platforms (e.g., DMS, LMS, data historians, etc.) which would allow AI/ML involvement in automatically generating new documents or executing activities without human intervention that would lead to the overall QMS being more compliant with regulatory requirements. As the saying goes, it’s a brave new world, and there have been some responses by industry participants when surveyed on the matter, that having records generate, advance, or complete themselves in the QMS is “defeating the purpose” of having quality management; the implication being that it should require a human eye on the situation.

Of course, there are also adherents to the school of Quality who believe that it’s impossible to “inspect-in” Quality into a process. What this means is that inspection occurs after the execution of upstream tasks, so if the work piece is found to be deficient, it must be scrapped, corrected, or some combination of the two activities. This philosophy, by the way, is exactly why QbD (Quality by Design) was elaborated in the first place: By designing steps of a process to make them easier to do correctly than incorrectly, and error-proofing (or reducing) each step, leads to a higher probability of a successful outcome. So, if Quality Management vis-à-vis the Quality Management System is being executed with greater compliance in each step of the process, then the overall (“rolled throughput yield”) outputs will lead to better outcomes than what we’ve historically seen.

If the goal is greater compliance to the Code of Federal Regulations, and higher Quality of the steps in the process that lead to the industry’s outputs, then advancements of QMSs are on a trajectory to improve the future state. The caveat is that the designers of the platforms need to be cognizant of the overall goal of enhancing Quality in the processes that are amalgamated within the QMS, instead of designing pieces in app silos that aren’t directionally conceived to move the needle in a better direction. 

---

Ben Locwin is a pharmaceutical industry executive who has worked as a global head of quality and regulatory functions, as well as working with QMS developers to provide input into the future of system designs. He has been featured in The Wall Street Journal, USA Today, Forbes, The Associated Press, NPR, Axios, and other top-tier media.