Thursday, October 23, 2014

Part 2: Responding to FDA483 Observations and Fulfilling Commitments

Part 2: Organizing to Get the Work Done

Responding to FDA483 observations was my focus in Part 1 of this series. Responses lead to commitments, and commitments lead to changes that are intended to prevent recurrence of the underlying problem that led to the observation.

In Part 2, I share my views on getting the work done to fulfill the FDA commitments. The following have become benchmarks from those who are the best at getting the job done in a sustainable way and getting the problems behind them:
  • Form teams. This might seem obvious, but forming cross functional teams representing stakeholders is often overlooked. The solution will very likely involve creating different ways of working that cross functional lines. It is a rookie mistake to think that the fix is something for the Quality Department to do on its own, or just an editorial change to a procedure. Depending on the problem, the team may need to be dedicated to allow focus, particularly if accelerated timelines are involved. Include subject-matter-experts that have the scientific, technical, and cGMP backgrounds to provide appropriate guidance.
  • Use consultants judiciously. Consultants can provide valuable additional support. They should bring relevant industry experience that can help scope the project and give practical advice that keeps the project moving toward the solution. The best consultants are good listeners who can help you design the right solution that works for you. But, if the only team members that show up to meetings are the consultants, something is terribly wrong. Do not outsource your responsibility to consultants. Let them help you, but it's your company and you own the results.
  • Determine the deliverables. Many teams go off the rails because they fail to determine the tangible deliverables that address the problem. You cannot grasps the degree of effort, determine the resource needs, or estimate the timeline until there is a clear vision of what "done" looks like. If you cannot describe "done," then you do not know the problem well enough.
  •  Solve the problem, not just the Commitment. Make a careful assessment of what it will take to prevent recurrence of the observation, not just satisfy the specific commitment. The FDA always states in the FDA483 that the observations are just examples and are not all-inclusive. Companies are expected to determine all that is necessary to permanently resolve the problem that underlies the observation wherever it applies. Perform a root cause analysis and ensure that the solutions are operational and sustainable. That invariably involves more work that the narrow view of a specific example cited in an observation or the response given.
  • Determine how you will know the problem is solved. In true fashion of a CAPA, determine how you will measure effectiveness. What will tell you that the solution worked? What will be your first sign that it did not work? Consider whether this is yet another attempt at solving the same old problem. Remember that recurring observations lead to increased levels of enforcement. Be sure you know the problem is truly solved.
  • Avoid Information Technology (IT) Solutions. This may sound a bit strange, but in my personal experience, nearly every commitment made in a FDA483 response that involved an IT solution missed the commitment date by light-years. I know that it should not have to be that way. I'm just saying that in my 20 years of consulting experience, that has always been the case. Consider applying an IT solution as a later improvement, not within the commitment time line if at all possible.
  • Use Project Management Principles and Tools. Quality System development work is not like a capital project where much of what is needed to estimate the time and cost is known upfront. Quality system and compliance projects are more often like exploratory surgery where a lot is learned once you open it up the patient. You may discover that multiple linking systems are affected after you start looking at the problem more deeply.  Recognize that plans will need to be adjusted along the way. Do not confuse progressive revelation with scope creep. Be adaptable. Establish plans and reports based on the work and target dates, not subjective feel-good dashboards with perpetual dates that slip. Anticipate problems with hitting the date and mitigate them before failure is a foregone conclusion. When you see that the timeline is slipping, ask for help and consider first what it will take to get back on track. Don't just keep moving the target date back as the default solution.
  • Provide Infrastructure Support. Remediating compliance problems often require new ways of thinking, not just new ways of working. Compliance problems frequently go hand-in-hand with company culture problems. Progressive companies that recognize the business value of an effective Quality Management System also recognize the opportunity that regulatory inspections provide to examine its culture and values. Change management coaches can help a company adapt to the changes required to support a quality and regulatory compliance culture. Recurring compliance problems are frequently a culture problem that discounts the value of an effective Quality Management System and regulatory compliance to the business. 
  • Provide Management Support. Ensure that each project team has a sponsor who is the best fit based on the functional owner of the process being remediated. Clearly identify the role of the sponsor and ensure that it includes removing obstacles such as lack of organization alignment, stalled decisions, team dysfunction, and inadequate budget. Sponsors are the champions who are willing to delve into the level of the working team to provide visible support, rather than sit on the management team and undermine their own folks at project update meetings. Being sponsors of quality system remediation projects provides a unique opportunity for function heads to understand dependent processes across functional lines.
  • Establish Governance Oversight. The work associated with fulfilling commitments is tightly connected to managing risk. Thus, it is important to keep site governance councils frequently updated with progress and informed about unacceptable risks. These risks always include the risk of meeting a commitment date made to the FDA. Take commitment dates seriously. Other unacceptable conditions uncovered along the way should also be communicated to the governance body. Establish clear lines of communication and escalation between the team that actually does the work and governance. Learn how to escalate and resolve potential problems quickly. Nothing consumes the timeline as much as reluctant escalation and indecision.
  • Verify that Deliverables Solve the Problem. Take an independent look to ensure the work is completed and that the tangible deliverables are in-place and in-use. This is a good role for an independent consultant. It is far better to find out for yourself that the mark was missed than for the FDA to call it to your attention as a repeated observation. Be a healthy skeptic and ask, "How do we know for sure the problem is permanently solved?"
  • Revisit Commitments. Although you may have verified that the solution is in-place and in-use, revisit the problem at an appropriate time interval to ensure that the organization has not reverted to the old ways of working.
How have you managed regulatory commitments? What are your success stories? What have been epic failures, and why?



The QA Pharm

Friday, October 17, 2014

Part 1: Responding to FDA483 Observations and Fulfilling Commitments

Part 1: The FDA483 Response

In this two-part series will discuss responding to the FDA483 and getting the work done to fulfill the commitments in a sustainable way.

Unlike a Warning Letter, a response is not required--but highly advisable. It is standard industry practice to respond to FDA483 observations. Responsible companies that want to have a good relationship with the FDA respond to each observation by saying exactly how problems will be addressed.

There is plenty advice available, even from the FDA, on what constitutes a good response. For certain, the difference between a good and a poor response may differentiate how you are viewed by the FDA.

From my experience, the following are points to consider that have worked well when forming a response:
  • Do you agree with the response? Stating up front that you either agree or disagree, in part or on the whole, stages the rest of the response. Think twice before you say you completely disagree especially with everything-- because the FDA will always have the regulations, industry good practice, and the facts on their side. But it is quite possible that the FDA investigator got the facts wrong or drew the wrong conclusion. This is why a close-out meeting with investigators at the conclusion of the inspection is important. It is better to straighten out the record at the close-out meeting than in the FDA483 response.
  • Defend what you are doing right. The examples of violations cited in observation are just a snapshot in time. Although you may not be proud of what was found, the finding may not be representative of the current state of control. The example may pre-date significant improvements that you have made on your own initiative, although they may be in progress and not fully implemented. Be sure to give yourself credit for such improvements. The last thing you want to do is give the impression that the FDA was the first to call a problem to your attention.
  • Take product lot numbers seriously. When the FDA mentions specific lot numbers associated with a problem, it immediately puts you in the position of having to defend why those lots are still on the market, and why you have not shut down manufacturing associated with that product, process, or facility. Put your best minds and writers to work to lay out the defense of your release decision and the state of control. Be objective and use data. If you come to the conclusion, albeit late in the game, that the product should be recalled, seriously consider the right thing to do. Also take into account causing a critical drug shortage, because the FDA will work with you to avoid a shortage while you resolve inspection-related problems.
  • Defend State of Control. Some types of observations paint a picture of being out of control, particularly when it is associated with validation, sterility assurance, potency, content uniformity--you know--the SISPQ issues. Also, taken collectively, observations that run the gamut of the quality system can indicate a precarious state of control. Take your best, realistic, honest shot at how you know you are manufacturing and distributing a safe, quality product. And, for goodness sake, do not--I repeat--do not anchor your defense that there have been no complaints or dead bodies in the street. Meeting all end product specifications may be the best defense you have if statistically treated, but be aware that a "final product spec" defense will always peg you as a 1970s time traveler.
  • Commit to Action. Address the specific issues that were called to your attention, but do not stop there. Carefully determine the root causes of the issues cited in the observations. There will not always be a one-to-one relationship of root cause to observation. Nonetheless, run the problem to ground and understand what is behind the problem. Determine the action required to prevent the reoccurrence of the problem cited. Undoubtedly, the solution will require changes in multiple areas at the system level. Be sure to convey that you are taking system-level action, not just addressing the superficial, specific examples. Remember, observations are just examples. Indicate the action you promise to take and target completion date. Be clear in advance what "done" looks like and the specific deliverables that will provide the evidence of successful completion.
  • Commit to Developing Appropriate Action. If you do not know exactly what all is necessary to fix the problem, it is completely acceptable to indicate that the scope of activities and target dates will be established in a plan to be provided later. But state in the response when the plan will be completed and provided. Major milestones and target month are usually acceptable in a follow-up communication.
  • Recognize the Minimalism. Very likely you have taken a minimum essential approach to the commitment. By the time all the internal reviewers have edited the response, you have pared down the work to something manageable for normal business. However, the work associated with FDA inspection observations is not normal business. The scope of the actual work required to prevent recurrence of the observation may be more than expected when you actually delve into the root cause and prevent recurrence through systemic changes. 
  • Provide Realistic Target Dates. There is always a balance between showing responsiveness with aggressive target dates and over committing to unrealistic dates. When dealing with operational and quality system problems, there may be more work involved than meets the eye. Some problems may require engineering studies or process development in order to understand the right approach to take when solving a problem. Take time to really think through all that is involved in order to provide realistic target dates. You don't want to go back to the FDA--at least not too often--to revise target dates. Do inform the FDA of changes in completion dates. Don't let a broken promise be a surprise when they return to reinspect. It's far better to inform them in advance.
  • Engage Your Internal Associates. I cannot tell you how often I discover the situation where commitments and target dates were given to the FDA in very well written responses, but those who carry out the work were not involved--even totally clueless. It's best to involve those who have to fulfill the commitments in the actual writing of the response. Time is of the essence when preparing a response, but cutting out those who actually do the work will always uncover significant execution issues at the worse possible time.
  • Take FDA Suggestions. When the FDA makes a suggestion about retaining a GMP consultant to assess your overall quality system and help you to improve--take the suggestion. This recommendation usually comes in Warning Letters where the FDA has come to the conclusion from the breadth of problems and recurring problems that there is no confidence in your Quality Unit or Management Oversight. Consider such a recommendation as a direct hit and taking on water, not just a shot across the bow. So be responsive and act accordingly.
What is your experience with preparing regulatory observation responses? What has worked well? What has not worked well?

In Part 2 I will talk about organizing to get the work done to fulfill commitments.


The QA Pharm