Understanding Patient Preferences for Medication Process Attributes: A Systematic Literature Review

Patient preference studies have been used to identify aspects of medical products that matter to patients. This paper sought to summarise patient preferences for treatment processes (mode and frequency).

Search strategies based on existing validated search filters were applied to databases including MEDLINE, Embase, Scopus, EconLit and Health Preference Research. Additional references were obtained by searching the reference lists of systematic reviews. English-language peer-reviewed studies involving patients that quantitatively elicited preferences and compared different modes or frequencies were eligible. Data extraction was performed by two reviewers using Covidence. As a heuristic, we considered if process attributes were ranked in the top two attributes. The Purpose, Respondents, Explanation, Findings, Significance (PREFS) checklist (0-5, higher = better) was used to evaluate study quality.

Data were extracted from 164 articles of an identified 16,357 (12,872 screened by title/abstract and 795 full text screened). Mean study quality was 3.48 on PREFS (standard deviation = 0.70), but patient engagement was rarely reported in studies (n = 20, 12.2%). Most studies included both mode and frequency attributes, either presented as a combined attribute (n = 65, 39.6%) or as individual separate attributes (n = 43, 26.2%). Regardless of how attributes were presented, mode/frequency attributes were infrequently in the top two attributes. In terms of level preference, when an oral mode was included as a level, the oral level was the most preferred level in nearly all studies. Patients with less severe disease or who were less experienced with treatment tended to prefer oral modes or have preferences against injection.

Mode/frequency attributes are not necessarily the most important attributes, but patients, particularly those with less treatment experience or more mild disease, may have strong preferences. Although PREFS scores indicated high study quality, future studies would benefit from deeper patient engagement. In line with ISPOR's definition, patient engagement should extend beyond participation to meaningful involvement across the research lifecycle, including opportunities for co-authorship and partnership in study design, conduct and dissemination.