Informing a priori Sample Size Estimation in Qualitative Concept Elicitation Interview Studies for Clinical Outcome Assessment Instrument Development

• Published in: Value in health Vol. 21; no. 7; pp. 839 - 842
• Main Authors: Turner-Bowker, Diane M., Lamoureux, Roger E., Stokes, Jonathan, Litcher-Kelly, Leighann, Galipeau, Nina, Yaworsky, Andrew, Solomon, Jeffrey, Shields, Alan L.
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.07.2018; Elsevier Science Ltd
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Child; Clinical assessment; clinical outcome assessment; Clinical outcomes; concept elicitation interview; Concept Formation; Concepts; Elicitation; Estimates; Female; Humans; Interviews as Topic; Male; Middle Aged; Patient Outcome Assessment; Patients; Qualitative Research; Research Subjects - psychology; Retrospective Studies; Sample Size; Saturation; saturation of concept; Surveys and Questionnaires; Young Adult; clinical outcome assessment; sample size; concept elicitation interview; saturation of concept
• DOI: 10.1016/j.jval.2017.11.014
• ISSN: 1098-3015; 1524-4733

Evidence-based recommendations for the a priori estimation of sample size are needed for qualitative concept elicitation (CE) interview studies in clinical outcome assessment (COA) instrument development. Saturation is described as the point at which no new data is expected to emerge from the conduct of additional qualitative interviews. A retrospective evaluation of 26 CE interview studies conducted with patients between 2006 and 2013 was completed to assess the point at which saturation of concept was achieved in each study. For each of the 26 interview studies, saturation of symptom concepts was assessed by dividing the sample into quartiles and then comparing the number of responses elicited from the first 25% of participants to the next 25% of participants, from the first 50% of participants to the next 25% of participants, and then from the first 75% of participants to the last 25% of participants. The number of interviews required to achieve saturation was documented for each study and then summarized across studies. Findings indicate that 84% of symptom concepts emerged by the 10th interview, 92% emerged by the 15th interview, 97% emerged by the 20th interview, and 99% by the 25th interview. Results provide practical guidance for estimating the number of interviews that may be needed to achieve saturation in a qualitative CE interview study for COA instrument development; address an important gap in qualitative research for the development of COAs in the context of medical product development; and offer useful information for study design and implementation.