RbM Data Reveals Site Screening Due Diligence Improves Patient Dropout and Retention

Clinical trial patient retention and dropout continues to be an issue amongst biopharmaceutical sponsors, as patient dropouts minimize the statistical power of clinical trial data, requiring study teams to enroll additional patients.  This outcome not only expends project management and patient enrollment resources, but also causes study timeline slippage, and increases the costs associated with developing additional investigational medical product.

There has been some research on factors that may impact subject dropout; for example, one research study evaluated that only 70% of enrolled subjects completed a Phase III clinical trial, and factors that impacted retention included age, gender, and education level ^[1].  Other research studies have concluded that depression ^[2], physical ability, employment status, and access to transportation contributed towards subject dropout ^[3].  Unfortunately most of this research is based on qualitative analysis, and does not definitively prove factors affecting patient dropout through statistical means.

With access to RbM data, study teams can now better understand analytical factors that impact subject dropout.  Additionally, with access to study performance data, study teams can create predictive models to mitigate pre-inherent study risks and change future outcomes.   In this article, we will statistically analyze the impact of screen failure rates on clinical trial subject dropout in a global Phase III clinical trial.

Cyntegrity’s EarlyBird System

Cyntegrity’s EarlyBird System is an RbM technology platform that uses data from all possible clinical recording systems (Electronic Data Capture systems, CTMS, Lab data, etc.) to evaluate clinical trial risk from numerous standpoints. The system incorporates a variety of statistical analyses and algorithms to gauge clinical trial performance, data quality and risk evaluation. In this case study, we leveraged aggregated data from a large global Phase III trial with more than 106 study sites and 1,700 enrolled patients and extrapolated Cyntegrity’s data-mining algorithms to assess study site engagement levels in different countries.

The Enrollment and Screen Failure Process

Figure 1 delineates the subject screening process for this Phase III clinical trial.

Figure 1: Subject Screening and Enrollment Process

Figure 1 illustrates that an inclusion/exclusion criteria assessment is conducted. Once the patient is qualified, the patient undergoes a consenting process, where both the patient and study site deem that the patient understands the clinical trial risks and benefits, and is committed towards complying with study requirements. Once the study team determines that a patient is a good match and is available for the clinical trial, the patient is enrolled. Alternatively, if the study team is concerned that a patient may not be committed towards the study, they screen fail the patient.

Impact of Screen Failure Rates on Patient Dropout Rates

Figure 2 delineates a statistical analysis that evaluates the impact of screen failure rates on patient dropout (or patient retention) rates.

Figure 2: Impact of Screen Failure on Subject Retention Rates (P<0.01, Adj. R² = 0.42) (Courtesy of Cyntegrity, Annex Clinical)

Figure 2 confirms that there is a statistical association between screen failure rates and patient dropout rates. In other words, study sites that screen failed more patients relative to total patients enrolled also tended to be statistically associated with higher patient retention rates.

What Might Be Causing This Trend?  The Polaris Perspective

Many may view this trend as very unusual and counterintuitive, however, Lauren Kelley of Polaris Compliance Consultants explains the potential cause of this trend.  “A really good screener will ask a lot of questions to make sure the subject understands the time and effort involved in the clinical study,” said Kelley. “They (research staff) ask about vacations or business trips that might interfere with visit schedules (that’s the number one thing subjects forget about), and they make sure that the subject is a good candidate for the study,” added Kelley. 

In this case, site staff that took more due diligence in screening and disqualifying/screen failing patients that lack commitment also exhibited better patient retention in the long run.

A Reason for Study Site Incentives?

In this study, whose screening procedures are depicted in Figure 1, we have confirmed a statistical association between screen failures and subject retention (Figure 2).  While studies and screening procedures vary widely, sponsors may want to consider incentivizing sites for conducting more thorough due diligence during the screening process, as the cost of patient dropouts can be very high. 

Investing a little time and resources upfront to prevent subjects with a high dropout risk from enrolling in a clinical trial would significantly mitigate the risk of dropouts during a trial.  Moreover, training research coordinators on asking dropout related questions and providing them with appropriate toolkits for interviewing and qualifying patients can also reduce patient dropout risk.  However, it is essential to emphasize that every study type requires a categorized enrollment analysis (i.e., Figure 1) in order to determine and execute the best subject retention strategies.

References

[1] Seeking Predictable Subject Characteristics That Influence Clinical Trial Discontinuation. Jai Shankar K.B. Yadlapalli and Irwin G. Martin. Drug Information Journal, May 2012; vol.46:pp.313-319, first published on April 9, 2012

[2] Marian U.C. Worcester, Barbara M. Murphy, Virginia K. Mee, Susan B. Roberts and Alan J. Goble. Cardiac rehabilitation programmes: predictors of non-attendance and drop-out. European Journal of Cardiovascular Prevention & Rehabilitation 2004 11: 328

[3] Azfar-e-AlamSiddiqi, AllaSikorskii, Charles W Given and Barbara Given. Early participant attrition from clinical trials: role of trial design and logistics. ClinTrials 2008 5: 328