Contrary to what Barry argues1 the paper presenting results from the Prospective Outcomes of Injury Study (POIS)2 is not a contradiction of arguments Colin Cryer and I have advanced against seeking to prevent all injuries irrespective of severity.3 ,4

The methods section of POIS paper clearly states that severity of injury was assessed according to the New Injury Severity Score.2 This is a threat to life assessment. We describe this as an anatomical threat to life measure to distinguish it from measures that are physiologically based.

What the POIS paper demonstrates is that injury severity as measured by threat to life is inadequate as the sole means for determining prevention priorities. It was this fact that was the primary stimulus for mounting POIS.

As stated in the paper by Wilson et al,2 prior to POIS's inception, the few injury outcome studies that had been undertaken have recruited participants from those perceived or assessed to have moderate to high threat to life injuries. Typically study participants were recruited from those seeking hospital treatment (usually inpatients) in the acute phase of their injury. The shortcoming of this approach is that a significant portion of injuries do not require hospital treatment in the acute phase, yet result in moderate to long term disability. As Colin Cryer and I have argued, priorities for injury prevention should be determined by a consideration of injuries that result in death, or are a high threat to life, or high threat of disability, or are costly.3 ,4

As described in detail elsewhere,5 ,6 POIS recruited study participants from New Zealand's national comprehensive no-fault injury compensation scheme. For the period July 2012–June 2013 there were 3.6 million new and active injury claims which cost the scheme $NZ2.4 billion.7 Approximately 8% of these claims were entitlement claims and they accounted for 80% of the costs. Entitlement claims are those that require some form of assistance (eg, compensation for time off work and/or rehabilitation) beyond the acute treatment phase. It was from this relatively small population of entitlement claimants, but at an earlier point in time, that POIS participants were recruited. So put another way, POIS in its recruitment focused on the more important cases—in terms of cost and likely ongoing disability. This shows I have been entirely consistent in focusing on important injury and considering multiple dimensions of severity in doing so. As POIS and the statistics show, this approach is necessary since these dimensions are poorly correlated.

Clearly Barry is not to be swayed from the view that we should be seeking to prevent all injuries irrespective of risk of death, risk of disability or cost. Accordingly, I intend to nominate him as a Patron of the about to be formed International Society for the Prevention of Paper Cuts.