Re-examining so-called 'secondary identifiers' in Disaster Victim Identification (DVI): Why and how are they used?

Disaster victim identification (DVI) refers to the identification of multiple deceased persons following an event that has a catastrophic effect on human lives and living conditions. Identification methods in DVI are typically described as either being primary, which include nuclear genetic markers (DNA), dental radiograph comparisons, and fingerprint comparisons, or secondary, which are all other identifiers and are ordinarily considered insufficient as a sole means of identification. The aim of this paper is to review the concept and definition of so-called 'secondary identifiers" and draw on personal experiences to provide practical recommendations for improved consideration and use. Initially, the concept of secondary identifiers is defined and examples of publications where such identifiers have been used in human rights violation cases and humanitarian emergencies are reviewed. While typically not investigated under a strict DVI framework, the review highlights the idea that non-primary identifiers have proven useful on their own for identifying individuals killed as a result of political, religious, and/or ethnic violence. The use of non-primary identifiers in DVI operations in the published literature is then reviewed. Because there is a plethora of different ways in which secondary identifiers are referenced it was not possible to identify useful search terms. Consequently, a broad literature search (rather than a systematic review) was undertaken. The reviews highlight the potential value of so-called secondary identifiers but more importantly show the need to scrutinise the implied inferior value of non-primary methods which is suggested by the terms "primary" and "secondary". The investigative and evaluative phases of the identification process are examined, and the concept of "uniqueness" is critiqued. The authors suggest that non-primary identifiers may play an important role in providing leads to formulating an identification hypothesis and, using the Bayesian approach of evidence interpretation, may assist in establishing the value of the evidence in guiding the identification effort. A summary of contributions non-primary identifiers may make to DVI efforts is provided. In conclusion, the authors argue that all lines of evidence should be considered because the value of an identifier will depend on the context and the victim population. A series of recommendations are provided for consideration for the use of non-primary identifiers in DVI scenarios.