4 Obstacles to Collecting Quality Data in Psychiatric Drug Trials

Sponsors often face significant challenges in collecting robust, quality data that demonstrates the safety and efficacy of investigative psychiatric drugs. However, before these problems can be solved, they must be understood. Be sure to keep these four issues in mind when mapping out strategies for ensuring quality data collection in psychiatric clinical trials.

1. There are a number of unique challenges inherent in working with a mentally ill population.

Manifestations of mental illness are varied and can be unpredictable. Additionally, these disorders are caused by complex interactions between the genetic, biological, social, and environmental factors. While the same is true for many physical conditions, psychiatric researchers must learn to work with a large number of interconnected unknowns.

2. Evaluation of a psychiatric condition has a large subjective component.

Observing changes in a patient’s condition over the course of a trial may have a significant subjective component. And with such subjectivity comes the risk of influence on the part of an investigator and/or patient that is largely unintentional and may not be easily recognizable. These issues must be anticipated through careful study design and mitigated when practical. Additionally, objective metrics should be included whenever possible.

3. Not all improvements to psychiatric symptoms may be attributable to the drug under investigation.

Of course, this is also an issue in trials testing efficacy for physical problems, but it’s one elevated in psychiatric trials. Mental illnesses often have a naturally unpredictable cycle. This means that a patient may be enrolled during a “flare” that subsides during the trial.

Additionally, over the last few years, psychiatric drug researchers have become less able to demonstrate treatment effects. For reasons that are not yet known, it’s getting harder to show that new psychiatric drugs work, which is most apparent when it comes to placebo response in acute schizophrenia trials. In 2007, a collaboration between the International Society for CNS Clinical Trials and Methodology and the International Society for CNS Drug Development focused on the apparent reduction in drug–placebo differences in multicenter trials for antipsychotic medications for the treatment of schizophrenia. It found evidence that there is now a higher placebo response. This is coupled with a lower response to therapeutic drugs, including previously established comparator products. These factors are introducing a novel challenge in treatment trials: Can a study be valid when control arms may not work as intended?

Factors that may be contributing to this problem include study design, subject selection, site capabilities, inclusion/exclusion criteria and diagnostic specificity. Currently, researchers are looking for answers to the placebo problem through considering questions such as:

  • Have study populations changed?
  • Are there any aspects of study design or execution that have inadvertently produced these results?
  • Have there been changes to the natural course of the disease being studied?
  • Are these the result of shifts in researchers’ diagnostic abilities?
  • Do newer compounds simply have lower efficacy compared to earlier drugs? If so, how does that account for the downward trend in established comparator efficacy?

4. Site factors play a major role in psychiatric study success.

Variability in study sites is yet another complicated factor that must be accounted for through good study design and monitoring practices.

The number of sites in a study has a major impact on study success. In general, the number of study sites is dependent on expected recruitment rate, available budget and desired timeline. However, there is risk associated with increasing the number of study sites: Doing so may lower the degree of drug–placebo differences. For example, one analysis of pediatric antidepressant studies found the number of sites to be the greatest predictor of placebo response rates. This is possibly caused by lower severity of illness, decreased site competence due to lower participant volume, and/or decreased standardization. Sponsors may employ centralized raters to mitigate this risk through greater uniformity and consistency of assessment.

The quality of site investigators and study staff is also crucial. This includes:

  • The degree of involvement and commitment of the principal investigator.
  • The staff’s experience, both in treating the indication being studied and in conducting clinical trials in general.
  • The past performance of the study site in a given indication with close attention given for the reasons behind dropouts and screen failures.
  • The availability and commitment of the staff toward timely data entry.

Research facility capabilities should also be properly evaluated. This may include the “feel” of the waiting room, the process for securing facility use agreements for third-party hospital space, and how subjects will be cared for beyond study-specific procedures. The study site’s ability to leverage networks, day-treatment programs, and local patient advocacy groups to recruit high-quality subjects most appropriate for a particular trial are other factors that should not be overlooked when picking sites.