Mediterr J Rheumatol 2021;32(2):104-11
A Systematic Review of Discrete Choice Experiment Studies in Rheumatoid Arthritis Biological Medicines
Authors Information

1. Pharmaceutical Management & Economic Research Center and Department of Pharmacoeconomics and Pharmaceutical Administration, Faculty of Pharmacy, Tehran University of Medical Sciences, Tehran, Iran
2. Rheumatology Research Center, Tehran University of Medical Sciences, Tehran, Iran
3. Department  of  Economics,  School  of  Economics  and  Political  Sciences,  Shahid Beheshti University, Tehran, Iran

Zartab S, Nikfar S, Karimpour-Fard N, Jamshidi A, Varahrami V , Homayouni A , Kebriaeezadeh A


Objective: Rheumatoid arthritis is a chronic disease with various clinical characteristics. The introduction of biological drugs has enhanced the efficacy and increased diversity of treatment options. Considering the patients' preferences in decision-making about treatment can improve their adherence. A discrete choice experiment is a type of conjoint method that can elicit preferences in more realistic scenarios. This article reviewed discrete choice experiment (DCE) studies to extract which attributes and levels were included in surveys. In addition, we focused on the process of designing surveys and the method that they used. Method: PubMed, EMBASE, Web of Science, Scopus, Ovid (Medline) and ProQuest were systematically searched in order to find studies that evaluated rheumatoid arthritis patients' preferences about biological medicines. Studies published in peer-reviewed journals between 1/1/1990 and 12/31/2019 were included. The included studies were analyzed using a narrative synthesis method and descriptive statistics. ResultsA total of 7733 studies were initially found. After deleting irrelevant and duplicate studies, 15 studies were included. The most common attributes that were used in surveys were efficacy, adverse effect, route of administration, frequency of administration, and cost. Most studies used a literature review for developing attributes and levels. The median number of included attributes and levels were seven and three, respectively. Eight studies explained their experimental design while seven studies did not. Conditional logit and mixed logit were the most common methods for modeling reciprocally. Conclusion: Several aspects of DCE studies investigating biological drugs in RA were assessed. Explaining the sample size, experimental design, and qualitative work for developing attributes can improve this type of study.

Article Submitted: 20 Oct 2020; Revised Form: 29 Jan 20201; Article Accepted: 5 Feb 2021; Available Online: 30 Jun 2021


This work is licensed under a Creative Commons Attribution 4.0 International License. 

©Zartab S, Nikfar S, Karimpour-Fard N, Jamshidi A, Varahrami V, Homayouni A, Kebriaeezadeh A.