Permitted index matching

43.8 Part 1D of the Crimes Act provides an index matching table that specifies which indexes of a DNA database system may be matched against each other.^[12] The Inquiry has heard concerns that these index matching provisions may be unduly permissive, contrary to the information privacy rights of those providing forensic material in the context of a criminal investigation.

Issues and concerns

43.9 The main privacy concern expressed with the index matching rules is the provision for unlimited index matching between the suspects index and the crime scene index—to generate ‘cold hits’ that may identify a suspect for an unresolved offence.

43.10 This form of index matching constitutes a use of suspects’ genetic information that is different from the purpose for which it was collected. For example, a person may be one of a number of suspects in relation to an offence. The suspect might readily submit to a forensic procedure for the purpose of eliminating him or herself from suspicion in relation to that offence. However, despite being eliminated from suspicion the person’s DNA profile currently could remain on the suspects index of the DNA database system for a period of 12 months—during which time it could be subject to unlimited matching against the crime scene index.

43.11 The Model Criminal Code Officers Committee (MCCOC) opposed unrestricted matching between these indexes while it was developing the Model Forensic Procedures Bill 2000 (Model Bill). The MCCOC discussion paper proposed that a suspect’s profile should be matched only against the crime scene profiles relating to the particular investigation in which the person is a suspect:

The suspects profile can be matched against anything on the crime scene index but unlike the serious offenders index, should not be available for unrestricted comparison as part of a pool of suspects that can be matched with profiles from any index. For example, it is not intended that the whole index of suspects could be compared with all crime scene profiles. To do so would go far beyond the purpose for which the forensic material was obtained in the first place and may expose suspects to random searchings by police anywhere in the country who are quite separate from the particular investigation and who are just fishing for matches on the crime scene index.^[13]

43.12 However, the final draft of the Model Bill permitted unrestricted matching between these indexes.^[14] MCCOC did not provide any explanation for this change in approach, but the rationale appears to have been to maximise the resolution of unsolved crimes by increasing the potential for ‘cold hits’ between suspect and crime scene profiles. Dr Jeremy Gans has commented on this process as follows:

An extreme instance of the doubtful decision-making process underlying the matching provisions is the major policy reversal between May 1999 and February 2000 on the question of whether profiles taken from unconvicted suspects can be compared on mass to crime scene profiles to generate ‘cold hits’ … This question is, arguably, the most important policy issue in the contemporary politics of DNA databases.^[15]

Submissions and consultations

43.13 In DP 66, rather than prohibiting cold hit matching in relation to suspects, the Inquiry proposed that opportunities for such matching should be limited by minimising the period of retention for suspects’ genetic information.^[16]DP 66 proposed that forensic procedures legislation should provide that forensic material taken from a suspect, and any information obtained from its analysis, must be destroyed as soon as practicable after the person has been eliminated from suspicion, or police investigators have decided not to proceed with a prosecution in relation to that investigation.^[17]

43.14 A number of the submissions supported the proposal.^[18] The Law Institute of Victoria commented that to retain an innocent suspect’s profile on a national DNA database in an index that permits comparison with unrelated criminal investigations constitutes ‘an unjustifiable double standard’.^[19]

43.15 Several submissions supported the proposal but raised certain concerns with it. For example, the Commonwealth Attorney-General’s Department submitted that the proposal raised a number of practical concerns:

The 12 months period is a guarantee that the material will be destroyed after a set time period. The proposal would allow the police to retain material for an indefinite period provided they were able to establish that an investigation was still on foot or a decision not to prosecute had not been made. In essence, such a proposal means that the decision when to destroy material is left entirely in the hands of the police.^[20]

43.16 The Office of the Federal Privacy Commissioner (OFPC) noted that:

It may not always be possible … to find and destroy all records relating to that sample and the profile, which are capable of identifying the suspect. Hence it is recommended that, in the case of ‘eliminated suspects’, the investigative officers and laboratory staff undertake to use their best endeavours to destroy all DNA-related information capable of identifying that suspect. In the event that the personal information of an eliminated suspect is improperly obtained or used or disclosed, that information should remain subject to the laws of evidence relating to improperly obtained evidence. The person mishandling that information should also be the subject of criminal sanctions.^[21]

43.17 Dr Jeremy Gans commented that the proposal did not achieve much on its own unless ‘cold hit’ matching were abolished. He also noted the difficulty in defining the ‘end of suspicion’ for the purpose of managing destruction of the genetic sample.^[22]

43.18 The AFP noted that its forensic laboratory already has difficulty managing the destruction dates for the genetic samples it holds. The AFP advised that, in practice, suspects’ profiles are entered into the AFP’s DNA database system with a default destruction date of 12 months from the date the sample was obtained. Two months before the destruction date, the computer reminds the laboratory of the coming date. The laboratory checks the file and contacts the police investigator, who has two months to inform them of the status of the sample and profile. If they do not do so the profile will be destroyed on the destruction date. For reasons of workload, the AFP considers it would be easier to manage an automatic destruction date—for example, ‘12 months after the sample is taken’—or at least to monitor destruction dates on a periodic basis.^[23]

43.19 The New South Wales Police Service (NSW Police Service) opposed the proposal on the basis that monitoring investigations to determine appropriate destruction dates would be extremely time-consuming. The Police also noted that:

In certain cases, investigators may decide not to proceed with a prosecution because of lack of sufficient evidence. However, it is impossible to predict what additional evidence may come to light in subsequent months. It is considered that the retention of forensic material for 12 months is justified in such circumstances as the suspect has not been conclusively eliminated from the investigation and 12 months is sufficient time for additional evidence to come to light or for a matter to be re-opened, if necessary, and further investigated.^[24]

Inquiry’s views

43.20 While the submissions generally supported the Inquiry’s proposal, several submissions raised concerns regarding its practical implementation. The Commonwealth Attorney-General’s Department noted that the proposal appeared to permit the retention of the material for a period longer than the current legislative maximum. In order to clarify this ambiguity, the Inquiry has amended the recommendation to reflect the current maximum retention period. Therefore, a suspect’s profile must be destroyed by the end of 12 months from the date of the forensic procedure, subject to any extension granted by a magistrate under s 23YD of the Crimes Act.

43.21 The AFP and the NSW Police Service expressed concerns about the potential increase in workload for laboratories. The Inquiry acknowledges these concerns but considers that the recommendation adequately accommodates them. The Inquiry understands that, at the federal level, computer databases currently alert laboratories of an upcoming deadline for destruction in relation to a suspect’s sample. The investigator must then notify the laboratory whether the destruction date has been extended pursuant to the Crimes Act. Under the Inquiry’s recommended approach, a police investigator would instead be required to notify the laboratory when the suspect has been eliminated from suspicion, or the investigator has otherwise decided not to prosecute the person in relation to that offence. As the police investigator would have carriage of the investigation, the Inquiry considers this would not be an onerous responsibility.

43.22 Once the laboratory has been notified of the destruction date in relation to a suspect’s forensic material and profile, the Inquiry considers that regular dates for destruction of batches of profiles—for example, fortnightly or monthly—would satisfy the requirement that they be destroyed ‘as soon as practicable’.

43.23 The Inquiry recommends that the Commonwealth should amend the Crimes Act to provide that forensic material taken from a suspect, and any information obtained from its analysis, must be destroyed as soon as practicable after the person has been eliminated from suspicion, or police investigators have decided not to proceed with a prosecution against that person in relation to that investigation. However, in any event, the forensic material and information must be destroyed no later than: (a) 12 months after the material was taken or the information obtained; or (b) the period stipulated in an order made under s 23YD of the Crimes Act.