Although patient-centered interventions for persons with multimorbidity are increasingly implemented in primary health care, evidence on effectiveness is still scarce and inconclusive.1 One potential explanation is the inconsistent use of outcome measures and a lack of a specific multimorbidity-adapted outcome measure.1,2 Using the 2015 North American Primary Care Research Group (NAPCRG) Annual Meeting, a forum was held with the goal of creating a list of relevant outcomes and to discussing methods of measurement.
Forum Process
The forum started with presentations on topics related to multimorbidity: concepts, definition, consequences, development of patient-centered outcome, and 2 intervention research examples. Results of a previous short survey on outcome relevance from the International Research Community on Multimorbidity platform were also presented.3 The online survey included 27 researchers. The main conclusions were that the most relevant outcome type was patient-reported outcome and most relevant domains of outcomes were self-management, quality of life, empowerment, and health behaviors.
Following the presentations, participants were divided into 3 small discussion groups and provided with 3 clinical vignettes (1 for each group) including 3 questions to initiate the discussion: (1) Have you experienced an intervention in multimorbidity and can you share that experience? (2) Which patient-perceived outcomes have the potential to be modified by the intervention? (3) If you had to build a single patient-perceived measure, what would be the outcomes to consider in order to capture the impact of the intervention?
Summaries of discussions were presented during a subsequent plenary session by each group and identified facilitators were invited to analyze the results on the spot to identify the consensual and relevant elements identified by the groups.
From the discussions, a list of relevant outcomes was created, grouped by categories and prioritized by the participants as the most important to consider when designing intervention for people with multimorbidity. Following the forum, the list of outcomes was reduced by conducting a thematic analysis. Outcomes that were related but named differently by the participants were grouped into constructs.
Results
Thirty-two participants from 6 different countries (Canada, United States, France, Belgium, Australia, United Kingdom) contributed to the discussions. They included general practitioners, nurses, social workers, and epidemiologists.
Thirteen outcome constructs (Table 1) were identified as important by the participants. Among these, 3 were identified as very relevant by all groups: quality of life; functional status; and goal attainment considering patient preferences. Three other outcomes were identified by at least 2 groups: general well-being; diseases knowledge and insight; and patient activation.
Outcomes Constructs Identified as Important by Participants
Participants identified that potential new measures should rely on a conceptual framework, include a variety of outcomes constructs and weight constructs to patients’ preferences.
Discussion
This forum gathered a sufficient number of knowledgeable participants from multiple fields and countries to allow a rich discussion. Furthermore, a post-NAPCRG blog posted in CMAJ by MacAuley, who participated in the forum discussion, reported that it was an insightful discussion on measurement by the world leaders in multimorbidity research.4
An extensive list of important outcomes was produced. The results offer an expert identification of multimorbidity-relevant outcomes, also suggesting that attempts to develop outcome measures should rely on a conceptual framework and be weighted to patients’ preferences.
- © 2017 Annals of Family Medicine, Inc.
