Let's Talk Diaries

With vaccine season bearing down on us, maybe we should discuss diaries briefly. No, I'm not talking about the diaries that talk about our day and other things like that, I'm talking about the diaries used in clinical trials that collect a nimiety of information reported by the research participant. As any coordinator who has worked with subject completed diaries knows, they can and often are a huge headache. This comes from several things, including: poor layout/format for the subject to follow on their own, inadequate education on the part of the site, failure to capture necessary information in the trial, and lack of proper review of completed diaries returned to site after completion. Two of these items fall back on the sponsor, and two of the them fall upon the site staff.

Since it's virtually impossible to have all sponsors agree on one format for a clinical trial diary, the layout of the diary is unlikely to change across the board for some time. I think eventually paper diaries will be obsolete and electronic diaries that provide data in real time will take over. This also could include diaries done over the phone, but I personally find that they last too long and trial volunteers get tired of "for yes, press 1; for no, press 2. If you are satisfied with your answer, press 1; if you want to change your answer, press 2" for 15 questions. Electronic diaries are faster, allow for a variety of variable inputs such as medication names for adverse events or unsolicited information that you cannot capture on a phone diary, do not waste paper (yea!), and they allow for direct data transfer into the study database.

That being said, until we reach that day when e-diaries are ubiquitous, remember these tips about paper diaries:

-Make sure that your trial participants understand not only how to properly complete the diary, but why it must be completed that way. For example, if a post-vaccination diary asks for a rating of injection site pain, then it asks for muscle aches; the subject should know that those are actually two different types of reactions. A positive response on both can mean different things to the sponsor than a positive response on one or the other.

-Fill out all diary headers (subject number, dates, contact info) ahead of time. You'll thank me later.

- Educate the volunteers thoroughly, but do not lead them. Many diaries will ask for information that may include or exclude a participant from a study, so be sure that the subject knows what each question is actually asking. Do not leave anything open to interpretation, that can lead to subjects being inappropriately lost from participating.

-When a diary is returned review it in detail with the subject. At that time the participant should make corrections and clarify any missing/questionable information recorded in the diary. It is much, much more difficult to clean up any discrepancies after the fact. As a side note, you can try educating volunteers on the "right way" to correct errors (line through, initial, and date. repeat as necessary), but if it takes new coordinators months to burn that into their brains you may not see promising results with study participants trained for less than an hour. I would just say you are better off with reviewing/correcting diaries as they are returned.

I hope you enjoy the tips, and let's all be ready to celebrate when everything goes electronic!

Goldmine for Recruitment?

A recent article in FierceBiotech IT detailed the emergence of a new patient recruiting service co-developed by Pfizer and Private Access. Allegedly this service will allow potential trial participants to log on to a website (presumably) and upload their health information to be made available to participating clinical investigators and physicians. Participants could also then restrict the amount of information available as well as to whom the information is provided (good for the patient in terms of privacy). Where this behooves the researcher is that an investigator would be able to only see the information of those who either seem to meet a study's criteria initially or are interested in a particular trial. But let's dig a little deeper, shall we?

The article goes on to say that most people don't participate in clinical trials because they either aren't aware the studies are going on or they can't find one for which they may be eligible. I agree with this statement. There are clinical trial search engines available today like centerwatch.com, clinicaltrials.gov, and clinicalconnection.com; but not enough people know about them to make them wildly successful. THAT is what Pfizer and Private Access need to take into account when developing this program.

I think the first step is to get the word out to clinicians who are not participating in trials to encourage their patients to sign up for this service or using Pfizer's vast financial resources to run marketing campaigns to get the ball rolling. Part B of the first step is to ensure patients that this information is indeed confidential because the whole idea of being able to access your protected health information from any computer and your PHI sitting out there in the clouds is a bit intimidating to most. The second step would be to prove that this service actually does what is supposed to do. It would have to be successful in hitting underrepresented populations of study participants, which is another problem with clinical trial recruitment as it stands today. The final step is showing that even though the service is developed by Pfizer, the service would not benefit Pfizer any more than it would benefit any other pharmaceutical developer. People generally have a bad taste in their mouths about profiteering in the world of Big Pharma and although Pfizer could potentially benefit greatly from this service, they would be better suited to open it up to any investigator participating in any trial.

Generally I think this product is a step in the right direction if implemented properly, and the ball lies in the court of the developer to make it happen.

Here's the article from FierceBiotech IT, FYI: (http://www.fiercebiotech.com/press-releases/pfizer-and-private-access-announce-plans-develop-online-community-accelerate-clinic-0)

Are We Gonna Hurry this Along or What?

There are multiple companies working to develop a H1N1 flu vaccine in time for mass vaccinations this fall and the media is doing a good job of publicizing the necessary clinical trials (well, at least for large academic research centers...the private sites who are probably doing a larger volume of the trials seem to be left out in the wilderness to do their own advertising) needed to complete the development process. However there is one thing worth mentioning that needs to be addressed before we get too far along into the Northern Hemisphere flu season:

What is the FDA doing, exactly? The FDA is trusted with the not so desirable task of regulating the medication/vaccine/device industry in the United States and they must walk a delicate line between a rapid development process and ensuring the safety of the millions of people who may receive a product once approved. The issue is that like many other government agencies, there is significant red tape and there seems to be no real timeline that the FDA follows when working with drug developers. Ordinarily the only people a regulatory delay significantly affects is the developer, but this an entirely different situation. The media makes a big deal out of the implications of a deadly, virulent pandemic on not only the health care system, but also the economy, so one would think that there would be a real sense of urgency from the FDA to get this vaccine (as well as any new treatment for the illness, more efficient diagnostic tools, etc). Unfortunately, I don't get the impression that this is the case. There are situations where a study is ready to start and just before the anticipated start date...the FDA says "stop, we need to talk about changing a few things." What is wrong with this picture? The FDA technically has up to 30 days upon receipt of a prospective protocol to advise for any changes, and they often use as much of that time as possible. According to the regulations the agency can take their time, but should they in this situation?

What I think would make sense is for the FDA to have a 1 business day turnaround time for any H1N1 related protocol. Even if they do request changes, it at least would give the sponsor enough time to implement them without disrupting the enrollment or time lines of the study (hopefully). To the sponsor, although we may not want to admit it, time is money and while it is a great thing to develop the next pandemic vaccine, there are millions (if not billions) of dollars to be had from a successfully developed vaccine. For the rest of us, the longer that we have to wait for clinical trials to start, the more likely it is that we will have a significant pandemic problem starting any day now.

You could argue that the H1N1 crisis is not the only thing on the FDA's plate, but there are few other situations with the potential explosiveness of this one. FDA needs to step up to the plate and quicken their processes before this swine flu gets out of hand.

Delegate or Die Part II

So let's get back to the delegation log discussion. Here are a few more tips on avoiding delegation log issues:

-Every investigator on the 1572 should sign the delegation log, even if an investigator does not perform any trial related activities. Once again, make sure that investigators are appropriately trained and have corresponding training documentation...an untrained investigator evaluating study participants is almost as bad as having an investigator not on the 1572 seeing them.

-Make every effort to have investigators print their own name, sign their own name, and sign their own initials on the delegation log. One purpose of the document is to prove the authenticity of signatures/initials on study documents, this cannot be accomplished by someone else signing the name of an individual on the delegation log.

-If your site uses dedicated recruiters, don't forget to have them sign the log as well. Recruitment is a part of the study enrollment process, so anyone assisting with it should be included on the delegation log

-Keep the log up to date with end dates of study staff. It's difficult to remember when someone started/stopped a study long after the fact, keeping up with it as it occurs is best practice. In addition no one should want to look forward to having to explain why someone had a start date on the delegation log listed after he or she had already performed study assessments. Likewise for ending participation a trial before the last assessment was completed.

Finally, don't be alarmed when the sponsor keeps the original delegation log at the end of the trial....it's supposed to be that way! There are other nuances with this fine document, but if you stick to these rules you'll have fewer items to correct and fewer notes to file to write explaining discrepancies. Enjoy!

Delegate or die! (okay, maybe not)

In a perfect world the principal investigator would perform all his own research, enter his own data, and record all adverse events that occur in a trial. News Flash! That doesn't really happen!

The FDA does realize that the PI cannot do everything so it allows for delegation of certain tasks to individuals who have been trained to do that task and are qualified to do so. Unfortunately there is not too much out there to guide investigators as to what extent this delegation can occur. Consequently, what could be seen as improper delegation in eyes of one individual could be seen as completely acceptable to another, and neither could matter if an FDA inspector feels differently! There are certain guidelines that the FDA mentions such as who can perform physical exams (unless stated otherwise by a sponsor, it would go by what qualifications are acceptable from state to state), but generally delegation of tasks is left to up to the PI.

In many situations, the coordinator will complete the delegation log and the PI will sign off on each individual participating in the trial as proof that he is indeed passing on/sharing responsibility with him or her. That being said, here's a few reminders to help avoid problems with the delegation log (as well as a potential 483) in the event of an audit:

-Make sure you have clarification from the sponsor, in writing as always, as to who exactly can perform the following: physical exams, ECG assessments, Lab assessments, AE assessments, special study evaluations like neurological exams. Improper delegation of these tasks can be a huge headache if it has to be resolved after participants have already entered a trial (I'd be happy to elaborate...just let me know if you want more of this!)

-The PI really should be the only one who signs off on the Case Report Forms (CRFs) as accurate representations of data collected during a trial. Sub-investigators do not sign the back of the 1572 and should not be delegated the responsibility of signing CRFs.

-Specialized tasks require people with specialized training. Is there a certain type of diagnostic assessment like an echocardiogram or a sleep study? If so, the person running those assessments should have documentation of appropriate training and should definitely sign the delegation log.

-Make sure that everyone who has not received sponsor provided training on the trial (ie: didn't attend the site initiation visit) has documentation of training in place before the delegation log is completed. This training should note the training involved, who was providing it, who received it, and most importantly, that the PI verified it. Just like improper the improper delegation described above, delegation of tasks without training is another huge headache that is very easily avoided with due diligence.

In the interest of suspense and writer's cramp, that is part I of delegation log tips....more to follow soon!

Report this?

Even though most IRBs only require deviations that affect subject safety or data integrity to be reported, many sponsors/CROs/CRAs still ask that sites report all deviations to the IRB. Why is that? I've heard some sponsors say that they would prefer all deviations to be reported to err on the side of caution and I've also heard some sponsors claim that they want to take the decision making process away from the PI. Those reasons are fair enough because the sponsor really is the one with the financial/corporate assets at stake in a trial and limiting the choices of investigators with regard to deviation reporting (investigators are not typically jumping up and down to send issues to the IRB) can have positive effects for study conduct.

On the other hand IRBs are there to protect the integrity of the study and the safety of the research participants. Many IRBs have numerous protocols to review and if they are overburdened with deviations or problems that don't affect subject safety or data integrity, they may not be able to properly do their job. I can't tell you the number of instances where the site has only submitted a deviation to the IRB because of a sponsor request only to receive an acknowledge back from the IRB indicating that it was received, but not reviewed because it did not meet their reporting requirements. The time and effort needed to send those acknowledgements could be better spent dealing with actual issues that meet those IRB reporting requirements. In addition IRBs function as separate entities from the sponsor, so they technically have their own agenda when reviewing study related issues, documents, etc. By asking sites to report every issue/deviation that emerges, the sponsor can disrupt the essential functions of the IRB.

Also it's worth nothing that along these same lines, the FDA issued guidance earlier this year regarding reporting of adverse events to IRBs as well (http://www.fda.gov/downloads/RegulatoryInformation/Guidances/UCM126572.pdf) Adverse event reporting is not something that occurs as frequently as unexpected problem/deviation reporting at the average research site. However, I would suggest that the FDA should issue guidance (guidances are not set in stone, and they are not actual "regulations"...remember that) on deviation reporting to IRBs in addition to the adverse event reporting AND that sponsors should work to help investigators report adverse events to the IRB instead of just concentrating on the problems in the trial. I think that such action would allow IRBs to do their job more effectively as well as cut down on the amount of unnecessary documentation on the site's end over the course of a trial.

A Coordinator's Quick Guide to Avoiding Queries

Query resolution is a pain. How much time and energy is spent on the sponsor as well as the site's side taking care of queries throughout the study? I couldn't even take a guess, but I would wager it is a significant account. Certain types of queries will always happen, but a coordinator can and should be more proactive to limit the number of avoidable queries.

For the coordinator:

Take a detailed medical history, and remember the following:

- Each surgical procedure MUST have a corresponding diagnosis/reason. For example, do not leave a cholecystectomy without a corresponding indication for the surgery. It may seem insignificant, but as it stands right now data management could easily query the unattached surgical procedure.

- Make sure that medical history entries supersede or match medication start dates. It would not make much sense for a medication for high cholesterol to be reported as starting 5 years before the reported start date of the high cholesterol (unless of course the medication was given prophylactically, but that is a somewhat rare occurrence)

- For female subjects: their child bearing potential status should be documented in the medical history. This includes hysterectomies, a start date for menopause, bilateral tubal ligation, etc.

- Adverse events should list a diagnosis, not symptoms. Try to make life easier on yourself by grouping symptoms into one diagnosis, if possible. The classic example is grouping cough, sinus congestion, and sore throat as an upper respiratory infection.

- Don't be passive. If you have a question about a particular data entry field, ask your monitor. If you disagree with a query, let data management know. This takes time, and it may be fruitless, but it just may save you the time and effort of redoing many data points because of a misunderstanding of the completion guidelines.

-Read the CRF completion guidelines. They are written and sent to the sites for a reason. That reason is not take up space in the regulatory binder, so have a gander at them...you just might learn something!

This needs an explanation

I understand the need for precision in clinical trials, but some of the requests are a little ridiculous. Let's examine a few examples:

Lab processing-When processing ambient samples, what is the significance of a specific wait time before processing, but after sample collection. The specific situation of which I am speaking is for immunity samples (safety labs do need to be collected according to lab guidelines because specimen stability and accuracy is at stake). What I want to know is how much of a difference 1.5 hours of clot time vs 2.5 hours of clot time makes? All the information that sites are provided is that the samples must be processed between x minutes and x minutes and if they are not done in this time frame, a deviation must be recorded. Why is this? I think sites deserve a medical/scientific explanation for these guidelines because there is variability between lab processing guides and it might actually reduce the number of deviations recorded if that information was provided to sites prior to the start of each trial.

Medical History-What difference does it make if someone had an appendectomy in March of 1976 vs May of 1980 if there is nothing with that procedure that would affect the conduct of the trial? I think that sponsors should limit their requirements for medical history reporting or at the very least, limit the number of queries that are sent to sites regarding medical history. Again, I would like a explanation from a data management point of view as to why such specific start dates are needed for conditions that have occurred many years in the past. The only disclaimer I will put on this statement is that sites should still express diligence in collecting the most accurate information as possible.

Concomitant Medications- The same argument as above could be made for conmeds. What difference does it make in the grand scheme of things (unless, again, that is something that would affect study conduct) if a subject is taking levothyroxine since 1980 vs 1985?

There are many other examples of such over reporting that is mandated by many sponsors, and I think that could easily be reduced to allow for sites to concentrated on more accurate and complete data reporting. I can tell you from experience that coordinators really lose interest in completing queries thoroughly if they are overburdened with seemingly meaningless queries. I would quickly rest my case if more a detailed explanation was provided for why all these data points need to be collected. So, data management gurus, if you're listening....that's an official request for you!

Let's be realistic here

You want to know what I really don't like? It's that period of time between the last subject's last study visit and database lock (I'm sure you thought I was going to say the media overhyping certain basketball-centric schools located in the triangle area, but that's another story...).

But in all seriousness, research coordinators tend to put studies on the back burner after the last subject has completed the last visit and the only thing remaining before close out is database lock. The problem here is that data management often issues queries maybe once or twice a week, during this period of time. This period of time could be a week, it could be several weeks, or it could be six months after the last study visit. In any case most coordinators have better things to do with their time then to sit logged onto electronic data capture systems and wait for queries to emerge. What actually happens is that the queries are generated and sit for several days or weeks before someone at data management or the sponsor starts to wonder why they aren't being addressed in a timely manner. This results in phone calls or emails to the project manager and monitor, who then promptly passes along the demand (often in a not so pleasant manner) to the site coordinator. One would think by reading these emails that the coordinator is a slacker and not upholding his/her responsibility to resolve data issues in a timely manner, but that is not the situation at all. The coordinator has many responsibilities to uphold and one of those is not to hold the hand of the study database as it is preened by data management before the final database lock.

The sponsor could do much to facilitate a smoother data cleaning period by either having automatic emails sent by the data capture system to alert coordinators of new queries or having monitors play more active roles in data management. I recognize that some studies have already implemented that role for monitors, but I have yet to see a data entry system that will alert coordinators when new queries are present. The funny thing is that such an idea is not that far-fetched...if our banks can email us when suspicious activity is posted to our account or our favorite baseball teams can text us when a good deal on tickets is announced, then why can't the same be done for queries in clinical research. It wouldn't have to be anything complicated, just a quick note stating that at X:XX time, X number of queries have been posted. The sponsor could even request a coordinator to log in and acknowledge receipt of the notice, just to confirm any suspicion of laziness (but we all know that no coordinator worth his weight in case report forms would ever have any thought of sloth). I would like to see such a system in place within a year because the technology already exists, it's just a question of data management firms/sponsor having the impetus to implement it on a large scale basis.