Interpretation of change over time in patient-reported outcomes requires appropriate responder definitions. This study compares responder definitions for the short-form version of the Quality of Life Enjoyment and Satisfaction Questionnaire [Q-LES-Q(SF)] in populations with generalized anxiety disorder (GAD) and bipolar disorder. A review of the Q-LES-Q(SF) literature published in English from 1993 through May 2009 identified publications using the Q-LES-Q(SF) in GAD or bipolar disorder clinical trials. In six relevant articles reporting Q-LES-Q(SF) responder definitions in GAD or bipolar disorder, two methods for defining responders emerged: (i) return to a score within 10% of community norms for the Q-LES-Q(SF); and (ii) a change score at or greater than the condition-specific mean change achieved by patients with minimal improvement on the Clinical Global Impression–Improvement (CGI-I) at study endpoint or a 1-point decrease on the CGI–Severity scale between baseline and study endpoint. The magnitude of the CGI-I based responder thresholds differed across mental health conditions. Use of the Q-LES-Q(SF) community norms as a responder definition is discouraged. A responder definition needs to be investigated within each condition or disease using appropriate anchors, and may not be generalizable from one condition or disease to another.
aUnited BioSource Corporation, Bethesda, Maryland
bAstraZeneca Pharmaceuticals LP, Wilmington, Delaware
cNew York State Psychiatric Institute, New York City, New York, USA
dAstraZeneca R&D, Mölndal, Sweden
Correspondence to Kathleen W. Wyrwich, PhD, United BioSource Corporation, 7101 Wisconsin Avenue, Suite 600, Bethesda, MD 20814, USA Tel/fax: +1 314 773 8000; e-mail: firstname.lastname@example.org
Received September 8, 2010
Accepted November 10, 2010