Wednesday, August 7, 2013

Assessment of FDA’s Final Guidance on Risk-based Monitoring

Yesterday, FDA finalized its 2011 draft guidance on risk-based monitoring (RBM). Without preamble, here’s what has changed.

Compared with the draft version of the guidance, the final version:

  • Reinforces, even strengthens, its support of RBM
  • Suggests “more intensive monitoring” does not necessarily indicate on-site monitoring
  • Provides specific examples of critical/non-critical data and tasks that could be accomplished remotely
  • Emphasizes the importance of protocol and CRF design
  • Notes that, under RBM, delegating monitoring obligations to a CRO requires more clarity, communication, and evaluation than traditional monitoring

Strong(er) Support for RBM
It is widely accepted that the draft guidance on RBM did not simply permit RBM, it encouraged it.  The final version may actually go farther.  Consider this phrasing from the draft document: "The overarching goal of this guidance is to enhance human subject protection and the quality of clinical trial data. This guidance is intended to make clear that sponsors can use a variety of approaches…related to monitoring.”  The final version sounds similar, but has a different emphasis: “The overarching goal of this guidance is to enhance human subject protection and the quality of clinical trial data by focusing sponsor oversight on the most important aspects of study conduct and reporting.”  (Emphasis added.)  The word “by” is critical.  It’s not just that sponsors can use RBM to enhance subject safety and data integrity, but that RBM is the way to achieve it.

Both documents note “there is a growing consensus that risk-based approaches to monitoring…are more likely to ensure subject protection and overall study quality,” citing several industry papers.  The final guidance adds, “FDA believes that risk-based monitoring could improve sponsor oversight of clinical investigations.”  That’s a subtle, yet important distinction; there’s a big difference between “it is believed by many,” and “I believe.”

More Intensive Monitoring Might Still Be Centralized
I’ve read in many papers and heard in many webinars that “an analysis of risk indicators can be used to trigger an on-site monitoring visit.”  Again, in the final guidance, FDA takes things further.  The draft states that certain risk factors, such as investigator inexperience or complexity of study population, “may require more intensive on-site monitoring.”  The final document says that those same risk factors “may require more intensive monitoring and consideration of on-site monitoring.  Again, the change in wording is subtle, but anyone who has ever arduously worked to turn a draft into a final publication – by committee, no less – knows that changes are seldom incidental; they’ve been painstakingly negotiated.  This rewording says to me that on-site monitoring should be considered for higher risk sites or protocols, but is still not required; remote monitoring, albeit more intensive in these cases, is acceptable. 

Additional Specifics
The final guidance expands the draft’s definition of Centralized Monitoring by adding a new section called “Examples of Alternative Monitoring Techniques.”  The section includes the following activities as examples of those that can be done remotely:

  • verifying continuous institutional review board (IRB) approval by reviewing electronic IRB correspondence
  • performing portions of investigational product (IP) accountability, such as comparison of randomization and CRF data, to preliminarily assess whether the subject was given the assigned product and to evaluate consistency between IP receipt, use, and records
  • verifying whether previously requested CRF corrections were made
This final guidance added some examples of technologies that enable centralized analysis (eICFs and eCRFs) and facilitate remote site communication (videoconferencing).

Some of the data that the protocol requires be collected may not be considered critical “because a small error rate in those variables would not affect the outcome of the trial.”  The final guidance offers some study-dependent examples:
  • concomitant medications
  • body temperature
  • body weight
  • demoographic characteristics
  • some routine laboratory tests performed as part of subject monitoring

Emphasis on RBM as Component of Study Quality
The final guidance emphasizes that RBM is a part of a high-quality study, but that quality has to be built from the beginning.  FDA very strongly states that “the most important tool for ensuring human subject protection and high-quality data is a well-designed and articulated protocol.”  FDA also makes clear that RBM can’t fix everything and that “some risks are better managed through activities other than monitoring.”  Perhaps the only way to remove the source of the  risk is to modify the protocol.  The final guidance includes several new sections devoted to this message.

Considerations for Delegating Monitoring Responsibilities
In the traditional monitoring approach, sponsors and CROs had a shared expectation about SDV percentage and on-site monitoring frequency.  The gold standard of 100% SDV and regularly scheduled monitoring visits came “pre-negotiated.”  With RBM, that’s no longer so.  It’s for this reason that FDA added a section to its guidance that discusses a sponsor’s delegation of monitoring responsibilities to a CRO.  It’s important that sponsors share information, such as risk assessment results, that may inform CRO monitoring practices for a trial.  FDA also suggests that sponsors and CROs have processes in place for timely exchange of relevant information (e.g., significant monitoring findings, significant changes in risk for a trial.)

This is, literally, a footnote from the final guidance (#26, to be specific):  “Two studies are on-going as of June 2013 that compare the effectiveness of on-site to alternative (e.g., centralized) monitoring methods (OPTIMON study (https://ssl2.isped.u and ADAMON study (”  I probably should have, but I didn’t, know about this.  After this enjoyable morning of comparing the draft RBM guidance with its final version, I know what’s next on my reading list.

I’d love to know your assessment of the finalized guidance.  Feel free to comment and share.

By Laurie Meehan

We at Polaris hope you find this information helpful.  Contact us at with a question or post a comment.



  1. This is a fantastic summary of the changes. Thanks so much for sharing it. I am pleased to see the additional examples that were added, those in the "Additional Specifics" section above.

  2. I would like to see more clarification on the data collection processes. In particular, whether local regulations and licensures are going to be enforced.

  3. thank you for sharing your summar! In Sweden, the authorities have already encourage RBM and hopefully, with correct implementation of RBM, more clinical studies would be performed by academic researchers as well as pharma industries.

  4. Wondering what others think about the sites scanning in and uploading signed consents to a web portal, as suggested by FDA. Seems to just transfer burden from the Sponsor's monitor to the study coordinator.

    1. Hi - That's certainly not the first time I heard that concern. I've spoken with a few CRCs for whom, even this early on, the extra work load has been an issue. I don't think it's unresolvable, but the transition to RBM has to be deliberate, well though out, and take site concerns into account.

  5. I am mildly annoyed when I read the prophecy of RBM saving sponsors money.

    I appreciated your review of the guidance and how you frankly said that the sponsor will need to work on the communication plan with their CRO and provide additional oversight and exchange; there is a cost associated with that activity. Yes, perhaps there could be less on-site time and less SDV but there is an oversight cost to implementing and doing RBM properly.

    Really liked your perspective and summary!

    1. Hi Nadia -

      Thanks for your comments. As I responded to the commenter above, I've definitely heard from overburdened sites. I'd like to be proven wrong, but I think the tranisition is going to be rocky.

  6. I head that RBM wouldn't be successful. God knows, if they have done proper risk assessment then may be they are already ready with plan B. Thank you Polaris for the information.

    Arnold Brame