Tschopp M1, Caliesch R1, Sattelmayer M1, Hilfiker R1
1HES-SO Valais-Wallis. University of Applied Sciences and Arts Western Switzerland Valais, Physiotherapy, Leukerbad, Switzerland
Background: Transfer from research to practice is known to be too slow. Despite the existence of several meta-analyses for almost every topic, decision making concerning interventions is not easy. First, there are several new randomised controlled trials being published every month. Second, most of these articles are not freely available. Third, exact and detailed information on how to perform the interventions are lacking in these publications, especially for interventions in physiotherapy. Fourth, results for specific subgroups are not yet available and health professionals cannot easily perform own analyses on published data, such as results for different endpoints (for example, the published meta-analysis reported outcomes after six months, but the researcher would like to see the results at the end of the intervention, or the meta-analysis reports only results of questionnaire B but the researcher would like to see the results for questionnaire C, etc.). Most important, clinicians do not have the possibility to access detailed information on the interventions, such as the interventional manuals or the instruction given to the study staff. Furthermore, information on adverse events may only arise later and might remain unnoticed, hence they should be collected centrally in a database.
Purpose: To reduce these problems, we set out to co-create with users and researchers a freely accessible webpage which will host different living systematic reviews (LSR) and we performed one LSR with a network meta-analysis on interventions for cancer related fatigue, incorporated interactive data analyses system.
Methods: We used a co-creation approach and adapted methods from software development (i.e. agile development/ scrum), implemented strategies for the prevention and detection of errors (such as data-extraction error, errors in primary literature, etc.) and we defined necessary future developments (e.g. machine learning approaches for semi-automation of review processes) to implement a sustainable maintenance of the system.
Results: Our literature search and the co-creation process with users and researchers resulted
i) in the implementation of project management and development method not yet used in physiotherapy, such as agile development/ scrum,
ii) shortened “production” cycles, i.e. instead of publishing a deliverable at the end of the project, each months a so called “potentially shippable product increment“ was published on the project webpage; and
iii) the implementation of test-routines for the detection of errors, for example in included publications or in extracted data).
Conclusion(s): Modern physiotherapy hugely benefits from methods of other professions, such as software developers. However, additional strategies need to be further developed, such as machine learning approaches for a true automation of searches which, in a best-case scenario, sidesteps existing subscription-based databases.
Implications: Our strategy circumvents existing problems such as paywalls, outdated evidence-syntheses, errors in publications that are not corrected, or rigid, non-interactive, presentations of data.
Keywords: strategy, living systemativ review, cancer related fatigue
Funding acknowledgements: This study was funded by the Swiss Cancer Research and the HES-SO.
Purpose: To reduce these problems, we set out to co-create with users and researchers a freely accessible webpage which will host different living systematic reviews (LSR) and we performed one LSR with a network meta-analysis on interventions for cancer related fatigue, incorporated interactive data analyses system.
Methods: We used a co-creation approach and adapted methods from software development (i.e. agile development/ scrum), implemented strategies for the prevention and detection of errors (such as data-extraction error, errors in primary literature, etc.) and we defined necessary future developments (e.g. machine learning approaches for semi-automation of review processes) to implement a sustainable maintenance of the system.
Results: Our literature search and the co-creation process with users and researchers resulted
i) in the implementation of project management and development method not yet used in physiotherapy, such as agile development/ scrum,
ii) shortened “production” cycles, i.e. instead of publishing a deliverable at the end of the project, each months a so called “potentially shippable product increment“ was published on the project webpage; and
iii) the implementation of test-routines for the detection of errors, for example in included publications or in extracted data).
Conclusion(s): Modern physiotherapy hugely benefits from methods of other professions, such as software developers. However, additional strategies need to be further developed, such as machine learning approaches for a true automation of searches which, in a best-case scenario, sidesteps existing subscription-based databases.
Implications: Our strategy circumvents existing problems such as paywalls, outdated evidence-syntheses, errors in publications that are not corrected, or rigid, non-interactive, presentations of data.
Keywords: strategy, living systemativ review, cancer related fatigue
Funding acknowledgements: This study was funded by the Swiss Cancer Research and the HES-SO.
Topic: Research methodology & knowledge translation; Research methodology & knowledge translation
Ethics approval required: No
Institution: n.a.
Ethics committee: n.a.
Reason not required: systematic review
All authors, affiliations and abstracts have been published as submitted.
