Advertisement
Research Article

School Playground Surfacing and Arm Fractures in Children: A Cluster Randomized Trial Comparing Sand to Wood Chip Surfaces

  • Andrew W. Howard mail,

    andrew.howard@sickkids.ca

    Affiliations: Division of Orthopaedic Surgery, Hospital for Sick Children, Toronto, Ontario, Canada, Child Health Evaluative Sciences, Hospital for Sick Children, Toronto, Ontario, Canada

    X
  • Colin Macarthur,

    Affiliation: Bloorview Kids Rehab, Toronto, Ontario, Canada

    X
  • Linda Rothman,

    Affiliation: Child Health Evaluative Sciences, Hospital for Sick Children, Toronto, Ontario, Canada

    X
  • Andrew Willan,

    Affiliation: Child Health Evaluative Sciences, Hospital for Sick Children, Toronto, Ontario, Canada

    X
  • Alison K. Macpherson

    Affiliation: Department of Kinesiology, York University, Toronto, Canada

    X
  • Published: December 15, 2009
  • DOI: 10.1371/journal.pmed.1000195

Reader Comments (11)

Post a new comment on this article

Much needed evidence on playground surfacing

Posted by bhagel on 25 Jan 2010 at 22:18 GMT

Howard and colleagues capitalized on a unique and rare opportunity to directly compare two playground surfaces for injury rates through a cluster randomized controlled trial – among the most rigorous designs for evaluating interventions in the health care field. The authors of the study carefully recorded the characteristics of the surfaces under actual conditions of use and performed a rigorous analysis of well documented outcomes demonstrating significantly lower rates of arm fractures involving falls from equipment on granitic sand surfacing compared with engineered wood fibre surfacing. The authors also presented an analysis of other, non-fall related injuries with no evidence of a difference in injury rates between the surfaces. The authors provide good evidence for the protective effect of sand surfacing over engineered wood fibre surfacing under the conditions reported. In my view, this study is among the strongest in the playground injury literature and represents a significant, and much needed, contribution to this area of research.

Competing interests declared: My membership on the Safe Kids Canada National Expert Advisory Committee overlapped with that of Dr. Howard and Dr. Macarthur and I have collaborated and published with Dr. Macpherson.

Pitfalls of Cluster Randomization Trials

Rogerwill replied to bhagel on 26 Jan 2010 at 14:05 GMT

Cluster randomization trials have a relatively short track record. Their popularity has only really developed over the past 2 decades and the development of well-accepted methodological foundations for this design has been relatively slow. It is well known that such trials may have substantially reduced statistical efficiency. The reduction in efficiency is a function of the variance inflation due to clustering. The increased administrative efficiencies (ease of performance) are often perceived by health researchers to outweigh the resulting loss in statistical precision.

The authors of the study carefully recorded the characteristics of the surfaces under well documented misuse of the product and performed a rigorous analysis of well documented outcomes that are subsequently invalid.

Competing interests declared: manufacturer of EWF