Is there an "unhealthy volunteer effect" in mental health outpatient research?

The ability to generalize clinical research findings depends on whether patients who agree to participate in studies have more or less severe illness than applicable clinical populations. We analyzed data from standardized clinical interviews given to patients seeking outpatient mental health treatment to determine whether there were differences between those who were willing to consider research participation and those who were not. Univariate and multivariable logistic regression analyses were conducted using measures of demographics, symptom and illness severity, and social environment. Factors associated with increased likelihood of willingness to consider research participation in the multivariable analysis were: moderate to severe depressive symptoms (OR=1.26; 95%CI: 1.07,1.48), hazardous drinking (OR=1.44, 95%CI: 1.19,1.75), having 1-15 days of partial disability due to mental illness (OR=1.19, 95%CI: 1.00,1.42), having any mental health outpatient visits in the last 2 years (OR=1.24; 95%CI: 1.09,1.42), and having a family member with mental illness (OR=1.57, 95%CI: 1.37,1.79). These results suggest an "unhealthy volunteer effect"-that patients interested in volunteering for outpatient mental health research have more severe illness, longer treatment history, and possibly greater genetic loading for mental illness than those not included in research trials. The possibility of such an effect should be considered when conducting and interpreting clinical trials.