To define research priorities to strategically inform the evidence base for physiotherapy practice.
A modified Delphi method using SurveyMonkey software identified priorities for physiotherapy research through national consensus. An iterative process of three rounds provided feedback. Round 1 requested five priorities using pre-defined prioritisation criteria. Content analysis identified research themes and topics. Round 2 requested rating of the importance of the research topics using a 1–5 Likert scale. Round 3 requested a further process of rating. Quantitative and qualitative data informed decision-making. Level of consensus was established as mean rating ≥3.5, coefficient of variation ≤30%, and ≥55% agreement. Consensus across participants was evaluated using Kendall's W.
Four expert panels (n=40–61) encompassing a range of stakeholders and reflecting four core areas of physiotherapy practice were established by steering groups (n=204 participants overall).
Response rates of 53–78% across three rounds were good. The identification of 24/185 topics for musculoskeletal, 43/174 for neurology, 30/120 for cardiorespiratory and medical rehabilitation, and 30/113 for mental and physical health and wellbeing as priorities demonstrated discrimination of the process. Consensus between participants was good for most topics. Measurement validity of the research topics was good. The involvement of multiple stakeholders as participants ensured the current context of the intended use of the priorities.
From a process of national consensus involving key stakeholders, including service users, physiotherapy research topics have been identified and prioritised. Setting priorities provides a vision of how research can contribute to the developing research base in physiotherapy to maximise focus.
Chartered Society of Physiotherapy's identification of national research priorities for physiotherapy using a modified Delphi technique Gabrielle Rankin, Alison Rushton, Pat Olver, Ann Moore
Physiotherapy - September 2012 (Vol. 98, Issue 3, Pages 260-272, DOI: 10.1016/j.physio.2012.03.002)