The Haddon matrix has been used both to conceptualize etiologic factors for injury and to identify potential preventive strategies.^1,2 In this issue, the matrix has been used by Eddleston et al to identify strategies to prevent death after pesticide self-poisoning (see page 333).³ I certainly agree with the authors of this paper that intentional pesticide self-harm is an important but neglected public health problem, not only because of its size but also because of its preventability.

I do not, however, entirely agree with some of the depictions of host factors in their matrix. In this editorial, I want to illustrate how, compared with traditional epidemiologists, social epidemiologists might have different depictions of the same host factor or identify different host factors. I argue that many of the so-called “unalterable” host factors can be viewed as “alterable”, depending on how the theoretical conceptual framework is used. These differences have great implications for injury prevention. Any host factor defined as “unalterable” or “not modifiable” will never be a candidate in the agenda of prevention strategies.

PRE-EVENT PREVENTION AND A BROADER VIEW OF ENVIRONMENT

Eddleston et al appear to be pessimistic about the primary prevention of self-poisoning using pesticides. They suggest that strategies to prevent deaths by restricting access to toxic pesticides may be effective but that there are considerable economic and regulatory barriers to their implementation. They repeatedly state that the substantial delay after the act of self-harm offers several opportunities to minimize damage. Accordingly, they invest most attention on event and post-event phases of prevention.

Taiwan, a newly industrializing country, also experienced many unnecessary pesticide poisoning …