Signing on and staying in

Illustration: Norm Bendell

The only thing harder than enrolling research patients is retaining them.

Clinical trials rely on regulations, investigators, coordinators, project managers, pharmaceutical companies, medical monitors, trial monitors, auditors, protocols, case books, review boards, and, oh yes . . . patients.

The question of why anyone would volunteer to participate in a clinical trial was addressed previously (see “Participating in clinical research”, Jan 2001), but the question of how researchers find patients and retain them in a research program is a completely separate issue (see also "Sex, drugs, and risk's role". The number of patients required for a particular study varies, but for a typical Phase III, long-term safety and efficacy study, 15–20 patients are necessary at each of 30–35 investigative centers. That doesn’t sound like very many, especially if the investigators manage active, large, well-known, or aggressive medical practices. However, the truth of all trials is that patients are quite hard to come by, even though they are either paid for their services or given a chance to use an experimental medication that may improve their health. In reality, a typical investigator will take up to 8 months to find and enroll 15 patients in a routine Phase III trial.

Patient retention is often more complicated than patient identification and is most significant in long-term trials. There are numerous reasons why patients choose to drop out of studies. For example, they might feel the medication is not helpful or the trial demands more of their time than initially expected. Patients sometimes just stop coming to their visits and ignore attempts by investigative centers to contact them. Of course, patients may also be discontinued from trials for medical reasons and study compliance issues, only some of which are preventable. Thus, keeping a patient over the course of a long-term study is difficult at best.

Investigators typically find their research patients in one of three ways: referrals from other physicians, through advertising (primarily in newspapers and on the radio), or from within their own practices. The best participants come from a doctor’s own practice because their medical histories and relationships with the researcher are already established.Knowing a patient’s history sharply reduces the time needed to determine study eligibility. The patient’s established trust of the investigator also improves the odds of protocol compliance and patient retention.(Both of these points are also somewhat true of referral patients, but the trust factor is generally diminished.)

Advertising, however, constitutes the primary means of attracting patients. Because of this, the U.S. Code of Federal Regulations restricts the content of ads to be certain that they are not misleading. Specifically, advertisements tell potential patients in general terms what the study involves, who is conducting it, how to contact them, and what potential benefits exist. Further, the regulations are explicit that no promises of a positive result may be stated or even implied. As an additional safeguard, the ads must be reviewed and approved by the Investigational Review Boards (IRBs) before use.

Combining known patient lists, referral sources, and advertising is key to finding suitable participants. To be effective, though, sites often must have some marketing savvy embedded in their medical approaches. It is critical not to mislead participants—it’s illegal, and once patients understand that they’ve been misled, retention is almost impossible. Understanding the patient population, though, is fundamental. As an extreme example that makes the point, deaf patients aren’t going to respond to radio ads. A subtler example is patients with anxiety disorders who may bypass a print ad if it is printed in red ink. Studies have shown that red tends to agitate already anxious people.

Eligibility
Determining patient eligibility is like walking a fine line, because no procedures associated with a trial may be conducted before a patient signs informed consent documents and enters the trial.In point of fact, this means that sites cannot, without consent, conduct psychological assessments, collect a blood sample, or even take a patient’s temperature. As a result, whether a patient meets the criteria for study entry is largely unknown at the outset. (This is why, in every trial, there are a significant number of “screen failures”—patients who give consent but do not meet the eligibility requirements and must therefore be immediately discontinued.) Thus, to reach the desired goal of, say, 15 enrolled patients, a site might have to recruit and screen 25. Sometimes, eligibility requirements prove too restrictive. On paper, they may seem reasonable, but when sites begin actively searching for patients, the criteria actually eliminate all but a few rare individuals. When these situations occur, and all possible recruitment options are exhausted, the pharmaceutical companies often amend their protocols. This process is never desired, however, because it takes time and may affect the statistical analyses.

Keeping the patients
The fact that patients are discontinued from trials because of adverse events, drug efficacy, or other medical reasons is a given in human research. But the real problem with patient retention comes from participants who are unwilling to meet the protocol requirements and drop out of the study by choice. Study noncompliance is manifested in many ways: patients skip medication doses, forget to attend visits, or refuse required procedures. In almost every study, protocol requirements state that patients cannot continue if they miss a certain number of treatment doses or if visits are not scheduled and attended within a certain small window. As a result, the relationship a site has with its patients can influence their behaviors and thus affect retention and compliance.

One other issue related to retention has to do, oddly enough, with the way in which some ill patients recover when drugs prove effective. It seems to make sense that any sick patient who becomes increasingly healthy over the course of a trial would want to continue in that trial. Unfortunately, this is not always the case; consider the following, real-life situation:

During a trial for a novel antipsychotic medication, a particular site enrolled a patient suffering from acute paranoid schizophrenia that had first been diagnosed almost 15 years earlier. The disease was so debilitating that the patient, who lived with his parents, was essentially isolated from the outside world. After approximately seven months on the trial, however, the patient’s improvement was shocking: he obtained a steady job at a local bowling alley, met and started dating a woman, made some friends, and, by all accounts, was beginning to have what most people would consider a “normal” life. A month later, however, he withdrew from the trial.

At the time, the study monitor asked the coordinator why the patient dropped out: “Charles was a truly sad case, a patient who improved faster than his own ability to cope. Essentially, we had several long talks in which I tried to convince him to continue with the trial as it seemed to have helped him so much. Finally, though, he came in one day and said that he had to drop out. He asked me if I ever had mornings where I’d just want to sleep in and then other times when the alarm clock would go off and I’d bound out of bed. I told him I did, and that it was perfectly normal. He thought for awhile and then said that if that was ‘normal’, he didn’t want it. It was too much of a ‘roller coaster’ for him. Prior to taking the study medication, every day was part of a continuum with no one day better or worse, from his perspective, than any other. He said he wanted to go back to that, because being ‘normal’ was just too unpredictable.”

Patient support
There are trials in which counseling and other forms of intervention are discouraged because they complicate assessing the efficacy of the study drug. In some cases, the fact that pharmaceutical companies do not pay for such therapies means that sites cannot afford to offer them, even if they are permitted. Of course, this is not always a consideration: Patients who overcome migraines or reduce their blood pressure as a result of experimental medications rarely need emotional support. But understanding the patients and the lifestyle changes that may result from their involvement in a trial is critical to retaining them in long-term studies. Family structures and other support mechanisms may also play important roles that cannot be ignored. Thus, treating research patients can require a more holistic approach than many physicians are used to providing.

Finding patients to treat in a research trial is a big task; treating them is even bigger. To be successful at both, researchers must use a wealth of techniques that are unique to the research process. They must simultaneously be marketers, advertisers, medical practitioners, counselors, and educators. Finally, to truly be effective, they must learn not just about their patient populations, but about their patients as well.