OR WAIT 15 SECS
What leads to Patient dropout? Read more here.
Limited research has been conducted on factors that impact patient dropout; some research suggests that patients who are less physically active were 7.3 times more likely to drop out of a clinical trial, whereas unemployed patients were 4.7 times more likely to drop out. Other research indicates that clinical trial dropout factors may include age, gender, education, and that depressed patients are particularly at risk of attrition.
The subject of patient retention and engagement is starting to generate interest in the clinical research industry, however, due to the limitations of data explaining why patients dropout, study teams are implementing generalized programs in order to minimize subject attrition.
This article will introduce data collection methods and concepts that can result in predicting patients at risk of dropping out from a clinical trial. Presuming that depression is a risk factor for patient dropout, we will analyze the impact of income on depression rates, and then apply the concept towards clinical trial risk indicator development.
Does Money Make Us Happier?
Treatment Online is an online psychiatric platform that engages patients and clinicians. The platform is capable of collecting patient reported outcomes through validated measures including Beck scales, mood measurement, drug side effect reporting, and socioeconomic behaviors for a variety of psychiatric indications. Over the course of five years, Treatment Online ran a pilot on and collected data from nearly 2,000 patients in New York City.
In this case study, we took a subset of this data (982 White and Hispanic patients), and evaluated the impact of patient reported income on depression rates. Figure 1 illustrates that Hispanics exhibited higher rates of depression in higher income ranges, whereas Whites exhibited higher depression rates in lower income ranges. Nonetheless, the $50K – $74.9K income range demonstrated the highest depression levels amongst both Whites and Hispanics.
How Can We Use Income as a Risk Indicator for Patient Dropout?
If the notion that depression adversely impacts subject attrition is true and valid, we can use data from Figure 1 to develop risk indicators, and specific patient engagement programs. For example, Hispanics who have reported earning less than $50K may be at a lower risk of dropping out compared to whites earning under $50K. Alternatively patients fitting within the $50K – $74.9K income range will likely exhibit higher dropout risk.
For study teams, this data could mean optimizing specific engagement programs based on patient risk. For example, patients within the $50K – $74.9K income range will need higher impact engagement programs compared to those in other income categories. Naturally, study teams will need to obtain income data on patients prior to study enrollment in order to categorize and fit patients into the risk models.
When developing risk indicators, it is important to combine and weigh numerous indicators (i.e., factors affecting patient dropout) in order to aggregate patient risk and develop impactful analytical risk and predictive models.
Predictive Modeling in Risk Development
In order to develop solid predictive models, study teams need to validate and verify specific factors that impact subject dropout. Albeit depression was used as an example in this analysis, there is no solid evidence that suggests that depression is associated with subject dropout. There are several ways in which sponsors can develop predictive models, which include conducting empirical analyses on historical trial data, and including additional data collection requirements that evaluate patient geo-demographics during the consenting phase. This way, study teams can optimize their analytical risk approaches towards patient dropout, and create effective patient engagement initiatives.
 Azfar-e-Alam Siddiqi, Alla Sikorskii, Charles W Given and Barbara Given. Early participant attrition from clinical trials: role of trial design and logistics. Clin Trials 2008 5: 328
 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