Becoming a Clinical Investigator – Subject Retention

 shutterstock_117915070

In this series, we have reviewed why one may want to become a clinical investigator, what one’s credentials should be, what one’s site requirements should entail, how to find the trials, and, critically, how to recruit qualified patients for the programs.  This installment will survey the rationale and strategies for keeping the randomized patients in the program for the required duration.

Several key points are very important to keep in mind.  By getting this far, a clinical investigator already has done the hardest part of being a clinical investigator.  But in no way should this be now seen as coasting to the finish line.  Indeed, from the sponsor and Contract Research Organization/Academic Research Organization (CRO or ARO) point of view, this phase of randomized subject retention is perhaps the most important phase of clinical research process because this is where the endpoints of the clinical trial are, which is the raison d’être for your participation in the first place.

All of the cost of trial has been expended to get the appropriate subject randomized.  Once randomized, this individual is extremely valuable, because the subject now provides the sponsor the evidence of benefit, neutrality, or, gag, harm for the investigational product or device.  This is why the CRO/ARO will send its clinical research associates to regularly dog your site incessantly to provide status reports, seek all pre-specified data points, and monitor you for any evidence of the dreaded protocol violation.

Another key point is that your research site is a reflection of YOU.  YOUR personality and style are on the line for the monitors to evaluate.  If you accept sloppiness in record keeping and good enough is adequate, you should just stop here because under the scrutiny of the professional CRO/ARO, your site will look bad and repeat business is not on the horizon.  Although in reality, the clinical research coordinators who work for you do most of the work, you signed the FDA1572 form, which makes you the one ultimately responsible for the work product.  I have the good fortune to have excellent CRCs, a research director and research administrator to buffer me as a principal investigator, but until you have enough work to justify administrative personnel, you are it.

A cardinal concept when pondering the topic of investigative subject retention is the idea of “intention to treat”.  This basically means that once an individual has been randomized to a treatment group, that subject is considered to be in that group until the trial is officially closed, or what is known as the final database lock.  An example of the complications of “intention to treat” may illustrate its significance.  If death is one of the endpoints of the study, and a subject has been formally randomized into the trial, and on his way home without his having taken even one pill of his assigned medication, and he is run over by a bus, his death is attributed statistically to the treatment he received as if he completed the study years later.  Clearly, there are statistical ways to mitigate such events, but you get the idea.  The same applies if the subject simply disappears or moves away and cannot be found.  The intention to treat mantra assumes the worse and he will be analyzed as if he has died, again attributable to the randomized treatment to which he was originally assigned.  Again, there are statistical strategies to deal with these circumstances, but they may critically weaken the power and statistical significance of the clinical trial, which obviously affects the trial’s results.

A research subject may decide that participation in the clinical trial is a hardship and he no longer wishes to comply with the sometimes rigorous follow-up requirements.  This decision is followed by cajoling by your staff and, if possible, modifications that do not impact the endpoints being studied, and your getting involved in trying to convince the subject to remain in the program.  Sponsors and CROs/AROs will go to great lengths NOT to lose the subject completely (see “intention to treat”).  The concept of “withdrawal of consent” is very strict for trialists.  Even if the subject does not want to be seen at the site, have his blood drawn, or otherwise participate in any study related activities, as long as he agrees to have his vital status checked (alive or dead, for example), he’s technically still in the study.  Sponsors have investigative services to track down randomized patients who have gone off the grid just to make sure they are at least alive (and thus not an adverse endpoint).

Many sites, including mine, provide stipends for visits and travel expenses to keep the subjects happy and minimize any potential hardship.  Having a pleasant environment and staff helps tremendously.  Covering transportation expenses, e.g., bus or taxi/Uber can make it easier for patients to participate and lower potential barriers to follow up.

You know you have done well when a research subject completes a trial and asks the staff if you have another research project in which they can participate.  At that time, I congratulate you for having become a clinical investigator!

credit: Irving Kent Loh

Bio:
Dr. Irving Kent Loh
MD, FACC, FAHA (Epidemiology & Prevention), FCCP, FACP is a board certified internist and sub-specialty board certified cardiac specialist with an emphasis on preventive cardiology. He founded and directs the Ventura Heart Institute, which conducts education, research and preventive cardiovascular programs. Dr. Loh is a former Assistant Professor of Medicine at UCLA School of Medicine. He is Chief Medical Officer and Co-founder of Infermedica, an artificial intelligence company for enhancing clinical decision support for patients and healthcare providers.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>