Risk Based Monitoring: A Primer for Small to Mid-size Sponsors

SPOTLIGHT EVENTRisk-Based Monitoring – Part Two In Depth ReviewMarch 13, 2014
Cambridge, MassachusettsDownload BrochureRegister

RELATED
- Increasing Intensity of On-Site Monitoring a Troubling Trend
- Industry Metric Indicates Low ROI with Full Source Document Verification
- Has FDA Guidance on Risk-based Monitoring Impacted SDV Coverage Yet?More in Risk-Based Monitoring

I’m on board with the FDA’s August 2011 document, “Guidance for Industry Oversight of Clinical Investigations — A Risk-Based Approach to Monitoring.” But it needs to provide more specifics, especially for small to mid-size sponsors who don’t have the resources or knowledge to develop their own risk-based plans.

Sponsors, CROs, regulators, sites and opinion leaders have discussed the pros and cons of a risk-based approach to field monitoring for years. The pharmaceutical industry always sought guidance from the FDA, and different approaches for reviewing data collected by sites. Many sponsors and CROs tried different methods but were nervous implementing a full scale risk approach to field monitoring.

I attended a conference recently where an FDA executive said sponsors have the technology to do risk-based monitoring right now but they don’t utilize electronic data capture (EDC) correctly. Sponsors created the separate source and CRF and the four- to six-week monitoring schedule – not the FDA.  Sponsors also created the practice of transcribing from the source to the CRF. That FDA executive got it right. The industry created the problem and should be the ones to fix it.

We need less paperwork, more use of technology and more efficient and better trained study sites. In an industry so focused on time we sure do waste a lot of it in our rush to get drugs to market.

Here is a proposed plan that has been proven to work in line with a risk-based approach as an alternative to the traditional monitoring plan of a visit to the site every four to six weeks.
 
Many companies already do this type of monitoring in Phase IV studies, but hesitate to attempt a risk-based approach in Phase II or III for fear they’ll miss something, or not meet the efficacy of the study designed to win FDA approval. Some including my own company have taken the risk-based approach with Phase III studies, but need to be sure we implement it correctly. (Recently I visited a Phase I center on the east coast with an amazing proprietary system that is a source and CRF combined. I wish they would share it with the world.)

A few steps we should consider before implementing a risk-based plan:

Step 1: For studies that don’t have a third party hospital involved in a study, stop the separate source document process immediately. It will only get worse. Create the new breed CRF that will be utilized as a source and CRF. Your CRF must include an adequate subject case history for this plan to work properly.

Step 2: Develop a CRF that collects “adequate case histories” and instruct your study sites to record all information directly onto the new breed CRF. The only separate paper source a site should have is the consent form (unless they use the system in Step 4), lab records if you are not using a central lab, and any other medical information from the past that is not collected in the CRF. Note: sites with electronic medical records should not use their electronic system for a study subject (this is equivalent to a paper medical chart). Put a note in the electronic chart, “research subject see research chart.” There is no reason to record the identical information in your electronic chart. We are pretty close to having the systems between a site and EDC talk. When this happens you can go back to putting the information in your system and connect it to the EDC. I hope this happens soon. We need the study sites to support this process for it to work.

Step 3: There is nothing wrong or illegal about collecting additional subject data that is not used for the study analysis. It’s part of the subject’s adequate case history to meet regulations. Just remember not to collect patient identifier information. Collect all data, code the data you don’t need for the analysis, and leave it alone. If we can assign a code for concomitant medications and adverse events we certainly can assign a code for “unusable data.” The FDA doesn’t request you analyze or report data you collected on the CRF that is not part of the protocol analysis.

Step 4: If you have an electronic patient reported outcome (ePRO) system put the consent form/process in this system so you don’t need printed consent forms. Provide the subjects with a printed consent form after it is signed from the ePRO system. The ePRO system can set up a document to print a page. There are electronic consent form companies out there ready to help. Recently I signed a few when I picked up prescriptions at the pharmacy and leased a car. ePRO vendors, please set this up so we have options.   

Step 5: Train your sites and certify them with your system across all centers and certify them on a yearly basis. It is often the sponsors and the research sites that remain behind on the technology because of costs and the time and resources it takes to manage the systems. One suggestion is to train the sites while you’re still in startup. Leverage the three months it takes to review and write the protocol and negotiate budgets to train the study sites on the electronic system. The EDC and other system vendors should offer a basic system for training while your system is customized. Imagine the time and money you’ll save cleaning data and monitoring down the road.

Step 6: Set up and document your data collection and monitoring plan with details on how you’ll manage this part of the study. It doesn’t pay to be vague in these documents.  Sites and monitors need clear instructions. Here are the key elements for developing a risk-based monitoring and data collection plan:

• Onsite Monitoring Triggers – Early identification of site issues related to patient safety, data integrity, and site non-compliance is critical to maintain an effective risk based monitoring plan.
   
  

• EDC System for Centralized Monitoring – This system can be very effective for evaluating an investigational site for ongoing compliance with data entry, protocol adherence, and data quality.  Note: Put a data entry requirement in your protocol.  Data must be entered within 48 to 72 hours of the actual subject visit.  Queries must be answered within three days.
  
  

• Critical Study Data Points – As you develop a risk-based monitoring plan you need to identify critical study data points beyond the primary and secondary endpoints. This helps the study manager determine the time and scope needed for both the onsite and centralized monitoring visits.
  
  

• Subject Safety Risk Assessments – Subject safety should always be the central focus of study investigators and sponsors. Site-specific safety issues that crop up during the study always should trigger an onsite monitoring visit; for example, a high number of protocol violations and reported serious adverse events. If trends in violations are noted a Corrective Action and Preventative Action (CAPA) Plan is essential. Don’t ignore the trends in hopes they will get resolved on their own. Most violations are identified during the query process and/or through onsite monitoring.  The sites don’t search for violations to report. They’re not interested in documenting their mistakes.
  
  

• Ongoing Site Performance Evaluation – Risk-based monitoring plans should include the process for evaluating ongoing site performance. They cover how significant issues are escalated and documented during the study, and how poor site performance triggers onsite monitoring visits. The best way to evaluate site compliance is to monitor the data entry process (are the sites entering data within a two- to three-day window of the subject visit, or only when they are called numerous times to do so?). Delayed responses regarding study questions from the support team or monitor also are possible signs of site non-compliance.  If a site doesn’t have the internal resources to complete the paperwork for your study consider not using them or providing them the money to obtain that resource.
  
  

• Documentation – Once you’ve determined the critical considerations for the plan the risk-based monitoring program is incorporated into the Project Plan and, if needed, into the Data Management Plan and CRA Monitoring Guidelines.

A Sample Risk-Based Monitoring Plan

The value of an in-person visit is enormous for all studies. We monitor sites to protect subjects and ensure sites follow the protocol and GCP and don’t commit fraud.  As a field monitor you are policing the investigator and the industry. We all believe the IRB plays a role in subject protection but if they are not monitoring a study site on a regular basis I’m not sure how this is possible. Therefore, the onus is on the sponsor, who often delegates it to the CRO.   

My company has implemented the following plan many times on behalf of pharmaceutical and device companies. It has proven to be a good starting point for sponsors who are not quite ready for a full risk-based plan. It is less risky in terms of time, resource and budget usage.

*From first subject screened to last subject completed and/or database lock

First Onsite Visit

This visit should occur as close to screening start as possible – but no sooner than two weeks from screening. You also can do this visit after one to three subjects are screened if you don’t want to hold up the site screening process. The site staff will retain the information better and be fresh when the study starts, so they’ll make fewer calls to the sponsor or CRO with questions.

If you have an investigator meeting do still conduct an onsite visit whether it’s an initiation or interim visit. Build a relationship with the site and coordinator. Understand their needs for the project and how they like to communicate (not everyone prefers email). Every site is different. The staff will share responsibilities at screening and it might not be just the assigned coordinator who conducts the screening visits. Know all the players and what they will do, and train them with the coordinator. The visit should include the following:

• A review of Protocol Highlights – Spend less than one hour on the protocol and more time explaining your inclusion/exclusion criteria and any special visit procedures. An elaborate slide presentation with 50 slides will only put them to sleep. Your audience is busy and usually doesn’t have time for a two hour slide presentation. They’re generally not interested in how you’ll evaluate your data or write your study report.

• A Quiz – Give coordinators and investigators a quick, four-to-six question quiz on the protocol. Ask a predetermined set of questions to the Primary Investigator and coordinator and record their responses. They don’t know the protocol well at this point but will be better prepared if they know you’ll quiz them.

• Staff Testing – Test the staff on the EDC and other systems (IVRS, randomization, and diaries). Have the coordinator enter a mock subject into the system so you can test the amount of automatic queries they get and see their skill levels with the system. All sites should be skilled in EDC by this point.

• Standard Procedures – Perform other standard procedures like drug storage conditions and accountability, study supply inventory, regulatory binder review, etc.

Finally, this is the visit that determines how often you need to conduct future visits. You still need this one even if you know the site well.  It’s essential to your study success. Provide a grading system and recommendation for the next visit: Example: Grade B means the site needs limited in-person visits but frequent phone calls or emails.

Site Communication
Send the sites an email communication at least once per week during their first three weeks of screening. Coordinators might not appreciate the frequent emails at first but it will keep them communicating and reiterate their importance to the project. 

I always recommend at least one phone call with the PI and coordinator together to evaluate how they work together and if there is anything you need to know about their relationship. Document the calls in the file. If you have a CTMS or can set up the EDC to document calls, do it. It will allow the study manager to view the same information in real time and will be useful if you change staff - especially monitors - during the study.  Please avoid having multiple systems for documenting study information. If a monitor or study manager can use one system for everything they will be happy and spend less time logging in and out of multiple system.

I don’t recommend a general newsletter this early.  Stick to personalized communication for the first month if possible, and time a brief (no more than two pages) newsletter of study information and trends across all sites to certain points.

Second Onsite Visit

Do this visit within two to four weeks of randomizing the first three to five subjects or sooner if they got a C or D grade at the first visit.

• Conduct a 100% check of the informed consent forms if they are paper consents.

• Review the regulatory binder for any changes to staff (this is a key check and is usually a sign of what’s to come)

• Complete 100% source data on select subjects (I recommend the first, third and fifth subjects). You will have limited source data if you follow the recommendations I noted earlier.

• Confirm drug storage, dispensing, etc.

• Set up your schedule for the remainder of the study.

Remote Monitoring During the Study

Assign a field monitor or in-house person to review data every week. Develop a checklist of items for this person to review.  I recommend the field monitor assigned to the site review the data to keep her/him close to the problems, but an in-house person also can be involved if staffing is limited. Determine how this will be done at the start of the study and stick to your plan. You will need to make changes as you go so expect and plan for them.

For standard checks choose at least one subject per site to review all data points and queries (either automatic or data manager). Review items that will not be picked up by the automatic edit checks. I recommend a few checks to ensure the automatic checks are working correctly. Checking all data points on at least one subject will give you an idea of the data entry, the amount of time it takes to answer and close queries, the type of subjects at this site, etc. Estimate one to three hours per subject based on the type of study (a 30 page CRF with AEs, medical history and concomitant medications will take at least an hour to review). Run reports (all systems should have reporting tools) that you set up at study startup so you don’t have to do any programming later. You can run listings by site or all sites combined. There are advantages for both, though I prefer to look at listings by site and then overall after the data is cleaned.

Third Onsite Visit
If you have a 12- to 24-month study this will be your third onsite visit.

If you have a three- or six-month study this will be your final visit to the site. Conduct the visit within two to four weeks of the last subject completing the final visit. Don’t wait until all subjects complete. You won’t be able to make changes or fix problems if you wait until then.

Complete the items noted in the second visit and add the following:

• Final drug accountability

• Instruction on what to do with the drug and study supplies (this will happen after the last patient completes)

• Close out the regulatory binder (use the typical closeout checklist for closing a site)

• Directions for closing SAEs, publications, etc.

Send a follow-up letter with all items outstanding immediately after your visit and prior to subjects completing the study.
 
Fourth to Sixth Onsite Visits

An eight- to 12-month study should have four to five visits depending on your comfort level with the site and the type of study. A 12- to 24-month study should have six to eight visits.

Closeout Visit

This is strictly based on comfort level. The majority of studies will not require a closeout visit if you have a good site and complete an in-person visit within two to four weeks of a subject’s completing the final visit. If you decide to conduct a closeout visit the standard procedures apply.

About the Author
Darlene Panzitta is President and Founder of DSP Clinical, a Contract Research Organization founded in 1999 that manages Phase I-IV clinical studies for the niche market of small to mid-size pharmaceutical and device companies with a focus on women’s health and reproductive issues. She can be reached at [email protected].