Background: Approximately half of people prescribed medications do not take them as prescribed. There is a significant unmet need regarding the barriers to medication adherence not being addressed in primary care. There is no agreement on which outcomes should be measured and reported in trials of medication adherence interventions.
Objective: To develop a core outcome set (COS) for trials of medication adherence interventions in primary care for adults prescribed medications for long-term health conditions.
Methods: A list of potentially relevant outcomes from the literature was developed. Using a two-round Delphi survey of stakeholder groups representing patients and their carers; primary care staff; and academic researchers with an interest in medication adherence; each outcome was scored in terms of importance for determining the effectiveness of medication adherence interventions in primary care. This was followed by two consensus workshops, where importance, as well as feasibility and acceptability of measurement, were considered in order to finalise the COS.
Results: One hundred and fifty people took part in Delphi Round 1 and 101 took part in Round 2. Eight people attended the workshops (four attendees per workshop). Seven outcomes were identified as most important, feasible and acceptable to collect in medication adherence trials: Health-related quality of life, number of doses taken, persistence with medicines, starting (initiating) a medicine, relevance of the medication adherence intervention for an individual, mortality, and adverse medicine events.
Conclusions: This COS represents the minimum outcomes that should be collected and reported in all medication adherence trials undertaken in primary care. When developing and finalizing the COS, feasibility and acceptability of collection of outcomes has been considered. In addition to the COS, medication adherence trials can choose to include outcomes to suit their specific context such as the health condition associated with their
medication adherence intervention.
Prof Debi Bhattacharya (Principal Investigator), University of Leicester
Dr Sion Scott (Principal Investigator), University of Leicester
Prof David Wright, University of Leicester
Katherine Murphy, PPI representative
David Turner, University of East Anglia
Kumud Kantilal, University of Leicester
Dr Allan Clark, University of East Anglia
Dr Jackie Martin-Kerry, University of Leicester
Disease Category: Other
Disease Name: Medication adherence
Age Range: 18 - 120
Sex: Either
Nature of Intervention: Drug
- Clinical experts
- Consumers (caregivers)
- Consumers (patients)
- Patient/ support group representatives
- Policy makers
- Researchers
- Service commissioners
- Service providers
- Service users
- COS for clinical trials or clinical research
- COS for practice
- Consensus meeting
- Delphi process
- Literature review
- Semi structured discussion
- Survey
The COS will be developed using a modified Delphi process, involving 3 phases:
Phase 1: Identify and review potentially relevant outcomes
Phase 2: Modified Delphi survey - 2 rounds
Phase 3: Stakeholder consensus workshop
Phase 1 will involve identifying all potential outcomes for medicines adherence from systematic reviews related to medicines adherence interventions and an existing COS for medication review trials. The list will be reviewed to identify all relevant outcomes, including definitions and organised into domains.
Phase 2 will involve reaching stakeholder consensus about which outcomes identified in phase 1 are the most important. We will conduct two Delphi rounds of stakeholder questionnaires. In round 1, for each outcome, participants will be asked to indicate whether an outcome is important on a 3-point scale: YES; NO and UNSURE/I DO NOT KNOW. Participants will be able to add comments for each outcome to explain their response. There will also be an opportunity to suggest any additional outcomes. Consensus will be met if the outcome is rated as important by 70% or more of participants of each stakeholder group. Outcomes reaching consensus to include or reject will be removed at round 1. Outcomes which fail to reach consensus and any new proposed outcomes will progress to round 2 for re-rating/rating. In round 2, for each outcome participants will be presented with the distribution of responses across the three options for each participant group and from all participant groups. They will be asked to re-rate whether the outcomes are important on a 2-point scale: YES or NO.
After round 2, the research team will collate responses from the two Delphi rounds. The outcomes that have reached consensus to include will progress to the workshop. Any outcomes that reached consensus to reject will be removed as well as those where all stakeholder groups failed to reach consensus. Outcomes will progress to the phase 3 workshop if a sub-group analysis demonstrates that at least one stakeholder group reached consensus to include (partial consensus). This is to ensure that no outcomes are removed that are important to one group e.g., patients, but are not considered important to the remaining groups.
Phase 3 will involve convening a 120-minute online consensus workshop with 2-4 representatives of each stakeholder group, up to a maximum of 20 participants. The aim of the workshop is to finalise the outcomes that should be included in the COS, explore the barriers and enablers to implementing the COS, and identify trial design features (e.g., using routinely collected data) and data collection tools and processes (e.g., patient follow-up) to facilitate this.