Improve SAE reporting via electronic submission
Reduce unnecessary reports
One of the bigger headaches for IRBs is having to sift through hundreds or thousands of unanticipated problems submissions when most of these should never have been reported.
The Human Studies Committee at Washington University School of Medicine in St. Louis, MO, has developed an electronic submission process that has reduced this problem. Without the system, the IRB reviewed in 2003 more than 11,000 submissions of unanticipated problems, deviations, etc., says Diane Clemens, DC, CIP, manager of research.
"This is a problem in the IRB realm where we've historically gotten inundated with a lot of reports we didn't need to see and which should have just been reported to sponsors," Clemens says.
Thanks to the system's screening and review process, the IRB reviewed 6,664 fewer unanticipated problems than the previous year, a 60 percent drop in volume, Clemens says.
The first person to see the unanticipated problems that have made it through the screening process is one of the four reviewers — all of whom are RNs, Clemens says.
However, if the event is a death, it is automatically routed to the IRB chair, she notes. Previously, unanticipated problems submissions were reviewed by four volunteer sub committees, each consisting of 20 people who met once a month, so the implementation of the electronic system and change to staff reviewers is a more efficient and effective process because these individuals do reviews every day, Clemens says.
"We've seen such a drop in the review volume because the system will screen out external events that do not cause a change in the protocol or consent or increase the risk to participants," Clemens says.
Here's how the system works:
• All initial serious adverse events (SAEs) go through the electronic system.
All initial SAEs go through the electronic system, where they are divided between internal and external SAEs, Clemens says.
"We still have a significant number of SAE follow-up reports that are submitted on paper, some 3,619, but these are dwindling," Clemens notes.
Investigators have access to the system and log in, following on-screen directions for reporting SAEs and other events, she says.
The on-line prompter will ask the investigator whether the risk to participants increased, whether there is a change to the protocol, and whether or not subjects need to be notified of additional risk, Clemens explains.
"Typically, we see if it's a long-term risk that's identified, then the participants need to be contacted," Clemens says." If it's something where after the first dose you are at increased risk for this reaction, but all people taking it at our site have already taken the first dose, then the need to contact them is no longer there because they're no longer at risk."
When investigators log on, they'll see a screen that has a list of all of their studies, both open and closed. They select which study the report pertains to and the next screen asks them to identify the type of submission they will make, Clemens explains.
"The initial page will give the specific item, unanticipated problems, external SAEs, a follow-up report, data monitoring report, a deviation, and we have another problem section," Clemens says. "The most common thing we see is submission of unanticipated problems."
Investigators select the correct path and follow-it through the electronic process.
For instance, for unanticipated problems, the process requests information about the timing of the event, and the system will calculate whether or not the submission is being made within the appropriate time frame, based on reporting guidelines that are pulled from the regulations, Clemens says.
The idea of this lengthy screening process is to assess these reported incidents against a very narrow definition of what needs to be submitted to the IRB, and these are consistent with the recent guidance released by the Office for Human Research Protections, Clemens says.
• System establishes criteria for what should be reported.
The Web page following the prompt about time frame is the screening tool, which helps investigators determine whether or not the event they are considering reporting truly needs to be reported, Clemens says.
The screening form outlines three criteria to determine whether something is an unanticipated problem or serious adverse event that requires submission to the Human Studies Committee and the Office for Human Research Protections, as follows:
- The event is serious;
- The event is unexpected;
- The event is reasonably related to the research.
If all three criteria apply, the investigator is directed to continue with the submission, and if they do not, then the investigator is told that the event does not meet the institutional requirements for reporting and should exit the system.
Since investigators often report events not meeting these three criteria to the IRB per instructions from sponsors, the next prompt also says, "If the Sponsor of the study is asking you to notify the IRB of an event that does not qualify for reporting, please send the following letter to the Sponsor as notification of our review policy."
The investigator could click on the prompt for an SAE review policy screen and print a ready-made letter to send to the sponsor.
The screening page also includes definitions of the criteria for investigators who are uncertain of whether their events qualify. For example, the definition states that an event is serious if it adversely alters the relationship between risks and benefits and includes events that either result in or require intervention to prevent the following:
- Inpatient hospitalization or prolongation of hospitalization;
- Life-threatening reactions;
- Result in persistent or significant disability/incapacity or permanent harm or disability (either physical or psychological);
- Jeopardize the subject;
- Congenital anomaly/birth defect in the offspring of research participant;
- Breach of confidentiality that may have a negative consequence;
- Death.
"Part of our system that's pretty nice is it's set up as a split screen with guidance to investigators on one side," Clemens says. "And on the other side of the screen they're working on questions we're asking them."
If an event meets all three criteria, then the investigator is given a link to a directory of events. They can choose whether it's related to allergy/immunology, general cardiac, etc., she says.
"Then there's a find button and it pulls up all the terms related to cardiac, for example," Clemens adds.
The idea is that the system limits the numbers of terms and helps investigators keep better track of events that have occurred in a study, linking them together as needed, she says.
"It provides some consistency and is especially helpful if you have several different coordinators who may be entering SAEs for one particular study," Clemens explains. "What the system will do is track the number of incidences of a particular event and calculate the percentage of occurrence based on the number of participants."
• Investigator adds pertinent information.
The electronic system's next prompt is for similar SAEs that have occurred at this institution, as well as other institutions, Clemens says.
"There may be events that are not exactly the same, but they have a similar process, or they maybe somewhat linked to this event," she says. "Also, the investigator may have gotten reports from other institutions, which are not something he had to submit to us, but it may be good information to have."
Likewise, the system asks investigators to provide additional comments, including descriptions of related events. For example, there could be a different study using the same drug, resulting in an unanticipated event in a different population, but the metabolic process could be the same, Clemens says.
"Investigators have the flexibility to elaborate on what has happened, which is extremely beneficial for our review, " Clemens says.
Then the system asks for information about when the SAE occurred and when the principal investigator was notified of the event, and that tells the system whether the report is made within the appropriate time frame, Clemens says.
The next thing the system asks is what changes need to be made in light of the event, including possible changes to the informed consent form, changes to the protocol, and if these are changed they are sent to the IRB in attached Microsoft Word documents, she says.
"We have received reports from investigators saying, "I intend to revise the consent form and I will submit it at a later date,'" Clemens notes. "The event is reported on time, and we've reviewed and considered it, but we'll hold it open until we receive the final piece of the revised consent form."
If an investigator chooses not to revise the consent form, he or she is asked to provide a rationale for why the change is not needed, Clemens adds.
One of the bigger headaches for IRBs is having to sift through hundreds or thousands of unanticipated problems submissions when most of these should never have been reported.You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content