Executive Summary
When research participants who committed to a study withdraw or simply stop participating without communicating with the researchers, the study’s validity may be harmed.
-
Noncompliance doesn’t necessarily compromise the study’s findings since this mirrors real world experience.
-
Studies may have to be discontinued if the number of participants withdrawing makes it scientifically untenable.
-
Emphasize the commitment participants are making during the informed consent process, one expert recommends.
Occasionally, individuals who agreed to participate in a research study withdraw for various reasons; some simply stop participating without communicating with investigators.
“We have a real problem in clinical trials when people decide they just don’t want to participate after they’ve enrolled,” says Jonathan L. Halperin, MD, professor of medicine at the Icahn School of Medicine at Mount Sinai and associate director of the Zena and Michael A. Wiener Cardiovascular Institute at The Mount Sinai Medical Center in New York City.
If some research participants discontinue treatment due to side effects they are experiencing or for other medical reasons, this doesn’t necessarily compromise the study’s findings. The same would presumably be true of actual patients, who are sometimes non-compliant. “In the real world, people interrupt therapy for all kinds of reasons,” says Halperin. “The problem comes when people disappear from view and don’t allow the investigators to know how they are doing.”
While study participants have the right to withdraw consent, “it’s very important for them to realize how damaging that is to the results,” says Halperin. Investigators are unable to determine the participant’s outcome. “In that case, we must consider all of the possible outcomes. Did they die, or is it that they just weren’t in the mood to continue follow-up?” he asks.
Due to this uncertainty, the validity of the study is compromised. This leaves investigators unable to conclude whether the randomized treatment being studied is harmful or helpful. “This is costly — not just in terms of money, but in terms of confidence in the trial results — and we don’t spend that lightly,” says Halperin.
If a study requires 500 participants and 10 stop participating without contacting the investigators, that means that the 490 other participants put themselves at risk for a study with an invalid outcome, he says.
“We may no longer be able to draw a conclusion, because we no longer have the statistical power to prove whether the treatment works not,” says Halperin. “So in a way, 10 people have put 490 others in harm’s way.”
There is no ethical justification to interfere with a participant’s decision to opt out of a study, says Tomas Jose Silber, MD, MASS, director of the Pediatric Ethics Program and research subject advocate at Children’s National Health System in Washington, DC. Nor should there be penalties such as refusing to give an agreed-upon payment, he says.
“Since Nuremberg, the voluntariness in research participation has been enshrined as the essential component of any investigation,” Silber says. “It stands to reason that if enrollment is voluntary, so is opting out.” Thus, some studies may have to be discontinued if the number of participants withdrawing makes it scientifically untenable.
Educate prospective participants
Halperin recommends educating potential participants. Researchers can say, for instance, “If you don’t think you can allow us to follow you until the study is done, it’s better you don’t do this. Think carefully, because you are making a commitment. You have the right to withdraw, but please do so for good reason.”
Silber recommends the following practices:
-
Make it very clear in the informed consent, and during the process of obtaining consent, whether the study holds a prospect of direct benefit to the subject. If it does, and there are study medications that need to be continued or gradually withdrawn, place a clear warning in the research protocol.
-
Emphasize this requirement during the process of obtaining consent, or permission if the participant is a minor.
“If, in spite of this warning, the subject wants to end participation, the researcher should ask for permission to inform the participant’s primary doctor about the interruption of the regimen and further recommendations,” says Silber.
-
Jonathan L. Halperin, M.D., Director of Clinical Cardiology Services, The Zena and Michael A. Wiener Cardiovascular Institute at The Mount Sinai Medical Center, New York City. Phone: (212) 241-7243. Fax: (212) 831-2195. Email: [email protected].
-
Tomas Jose Silber, MD, MASS, Director, Pediatric Ethics Program, Children’s National Health System, Washington, DC. Phone: (202) 476-3066. Fax: (202) 476-3630. Email: [email protected].
