Reason: Restricted by author. A copy can be supplied under Section 51(2) of the Australian Copyright Act 1968 by submitting a document delivery request through your library or by emailing firstname.lastname@example.org
Optimising the utility of injury surveillance systems to inform injury prevention in active populations
In order to distinguish essays and pre-prints from academic theses, we have a separate category. These are often much longer text based documents than a paper.
posted on 15.05.2017by McKinnon, Adam David
Background: Injury is the most common disbenefit of participation in physical activity
and has substantial related personal, social and economic costs. An understanding of
where, when, how and to whom injuries occur is critical for the development of
interventions designed to prevent injuries. Injury surveillance contributes to this
understanding, and despite the importance of such systems, limited research exists
examining their optimisation from a user perspective.
Aim: To investigate strengths and weaknesses, and identify shortcomings of injury
surveillance systems with regard to factors that influence system operation, and thereby
identify enhancements that optimise system utility from a user perspective.
Methods: Six studies utilising a series of complementary research designs were
undertaken, some using more than one analysis method. Two qualitative studies, directed
by the tenets of Grounded Theory, were conducted with key informants interacting with
injury surveillance systems in the Australian Defence Force, and in the Australian state of
Victoria, to determine system experiences and future operational expectations. The results
of these qualitative studies informed the objectives and designs of the later research.
A Discrete Choice Modeling study was then performed with 225 users of the Victorian
Injury Surveillance System data output to determine user preferences toward
predetermined current and hypothetical information dissemination mediums. Finally, three
studies trialed and evaluated five novel analytical methods, previously untried in the
analysis of injury surveillance data. The methods were: (1) the 11 and EWMA statistical process control charts, (2) Kohonen Self Organising Maps, and (3) the A priori Association
Rule and SPSS Clementine Sequence Analysis algorithms. These analytical techniques
were applied to an historical ADF injury data set. The process of applying each technique
and interpreting the results obtained were evaluated using preliminary criteria designed to
assess utility through the evaluation of usefulness and user responsiveness.
Results: A range of factors relating to injury data collection, analysis and interpretation,
and information dissemination that act as barriers to optimal injury surveillance in terms of
quality, efficiency and usefulness were identified in the qualitative studies. Sociocontextual
factors were also identified which were unique to each research setting. These
have received limited attention in previous research. The Discrete Choice Modeling study
indicated user preference, in order of preference strength, toward online dissemination of
injury data and information, a willingness to pay for system information products (inverse
relationship); an online information repository; and dissemination of a regular publication.
The five data analysis methods each identified previously unidentified injury problems
within the ADF data, and each met several of the criteria associated with utility. The
strength and weaknesses of these methods are discussed.
Conclusions: There is a scarcity of critical research directed at optimising the performance
and utility of injury surveillance systems, particularly from a user perspective. This
research identifies some potential means of achieving enhancements. Future research is
required in two major categories: (l) the human interaction with all phases of injury
surveillance systems; and (2) improving the empirical evidence and knowledge regarding
optimal methods of analysis, interpretation and dissemination of injury surveillance data