Introduction: Appropriate outcome selection is essential if research is to guide decision-making for patients, professionals and policy makers. Systematic reviews evaluating the clinical, cosmetic and patient-reported outcomes of breast reconstruction, however, have demonstrated marked heterogeneity of outcome reporting such that results from individual studies cannot be compared or combined. Standardising end-points by developing and using core outcome sets - an agreed minimum set of outcomes that should be measured and reported in all research and audit studies – is one way by which outcome reporting may be improved. We therefore report the initial results of the BRAVO (Breast Reconstruction and Valid Outcomes) Study which aims to use a scientifically rigorous Delphi consensus process to develop a core outcome set for reconstructive breast surgery.
Methods: The Delphi process involves the sequential completion of questionnaires to allow stakeholder opinions to be synthesised using item responses to prioritise outcome domains.
The questionnaire was developed from a long list of 148 outcomes generated from literature reviews and qualitative work with stakeholders. The outcomes were categorised into 34 domains in six categories (short-term complications; late complications; symptoms; psychosocial issues; practical issues and cosmesis) and each domain operationalised.
Key stakeholders were identified as patients, surgeons, specialist nurses and psychologists and participants were sampled purposively to ensure a breadth of perspectives. Each participant was sent a questionnaire and asked to prioritise the outcomes on a nine-point likert scale from 1(not important) to 9(extremely important).
The number of respondents in each group rating each outcome as not important(scores 1–3); equivocal(scores 4–6) or very important(score 7–9) were calculated for each item and compared between groups. The proportions of respondents rating each item as very important(score 7–9) was used to rank the items.
Results: 213 of the 430 questionnaires were returned(126/274 patients and 87/156 professionals) giving a response rate of 49.5%.
Patient participants had a median age of 53.4 years(range 34–76) and had undergone a full range of reconstructive procedures. The professional group included 39 breast surgeons, 20 plastic surgeons and 18 clinical nurse specialists.
There was agreement between 7 of the 10 outcomes that each group rated most highly. Items with consensus included patient-reported cosmesis, cosmetic satisfaction and early complications. Patients, but not professionals, considered generic complications such as bleeding to be important while professionals valued psychosocial issues such as self-esteem more highly than patients.
Conclusions: Patients and professionals prioritise similar outcomes, but areas of discrepancy with regard to complications and psychosocial outcomes remain. A further Delphi round asking participants to re-prioritise outcomes and a consensus meeting to ratify the final decisions will be necessary to determine a final core outcome set for reconstructive breast surgery.
Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P4-17-03.