The number of risk prediction models developed for orthopaedic patients has increased in recent decades and continues to rise. This trend has been further accelerated by the rise in availability of large healthcare datasets, advanced machine learning algorithms, and computing power. However, only a small proportion of these models have been implemented in clinical practice where they might be able to make a positive impact on patient experience and outcomes through improved shared clinical decision-making. This highlights the fact that it has become a relatively simple task to develop risk prediction models, while implementation is more challenging. Part of the challenge is the broad stakeholder involvement required to take a model from the computer to the live clinical environment. Such stakeholders include hospital administrative and information technology staff, clinicians, patients, and research staff. Furthermore, clinically meaningful implementation is only possible if the impact of implementation is monitored closely and consistently over time, with adjustments being made to the risk prediction model and the workflow in which it is embedded as necessary.
The purpose of this research topic is to encourage the submission of studies reporting on implementation of risk prediction models in the orthopaedic clinical environment. Clinical impact evaluation is encouraged but not necessary – if author have plans for how they will evaluate clinical impact of risk prediction model implementation then they are encouraged to include this in implementation studies, even if the impact evaluation has not yet commenced. As with all good research, transparency of reporting is critical and as such authors should not hesitate to submit studies in which ‘negative’ results were found, for example the risk prediction model have no impact on the outcome of interest. Well-conducted studies should be submitted, regardless of the findings.
The purpose of this research topic is to encourage the submission of studies reporting on implementation of risk prediction models in the orthopaedic clinical environment. Clinical impact evaluation is encouraged but not necessary – if author have plans for how they will evaluate clinical impact of risk prediction model implementation then they are encouraged to include this in implementation studies, even if the impact evaluation has not yet commenced. As with all good research, transparency of reporting is critical and as such authors should not hesitate to submit studies in which ‘negative’ results were found, for example the risk prediction model have no impact on the outcome of interest. Well-conducted studies should be submitted, regardless of the findings.
Keywords:
Implementation; risk prediction; machine learning; decision-making; clinical impact
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
The number of risk prediction models developed for orthopaedic patients has increased in recent decades and continues to rise. This trend has been further accelerated by the rise in availability of large healthcare datasets, advanced machine learning algorithms, and computing power. However, only a small proportion of these models have been implemented in clinical practice where they might be able to make a positive impact on patient experience and outcomes through improved shared clinical decision-making. This highlights the fact that it has become a relatively simple task to develop risk prediction models, while implementation is more challenging. Part of the challenge is the broad stakeholder involvement required to take a model from the computer to the live clinical environment. Such stakeholders include hospital administrative and information technology staff, clinicians, patients, and research staff. Furthermore, clinically meaningful implementation is only possible if the impact of implementation is monitored closely and consistently over time, with adjustments being made to the risk prediction model and the workflow in which it is embedded as necessary.
The purpose of this research topic is to encourage the submission of studies reporting on implementation of risk prediction models in the orthopaedic clinical environment. Clinical impact evaluation is encouraged but not necessary – if author have plans for how they will evaluate clinical impact of risk prediction model implementation then they are encouraged to include this in implementation studies, even if the impact evaluation has not yet commenced. As with all good research, transparency of reporting is critical and as such authors should not hesitate to submit studies in which ‘negative’ results were found, for example the risk prediction model have no impact on the outcome of interest. Well-conducted studies should be submitted, regardless of the findings.
The purpose of this research topic is to encourage the submission of studies reporting on implementation of risk prediction models in the orthopaedic clinical environment. Clinical impact evaluation is encouraged but not necessary – if author have plans for how they will evaluate clinical impact of risk prediction model implementation then they are encouraged to include this in implementation studies, even if the impact evaluation has not yet commenced. As with all good research, transparency of reporting is critical and as such authors should not hesitate to submit studies in which ‘negative’ results were found, for example the risk prediction model have no impact on the outcome of interest. Well-conducted studies should be submitted, regardless of the findings.
Keywords:
Implementation; risk prediction; machine learning; decision-making; clinical impact
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.