Prioritizing Untested Sexual Assault Kits by Victim–Offender Relationship

Jiaqi Lu, University of New Haven

Large amounts of untested sexual assault kits (SAKs) have troubled jurisdictions across the country, and debates regarding processing untested SAKs have continued for decades. SAKs are containers of biological specimens and related physical evidence collected from sexual assault crimes for DNA profiling. After collection, SAKs usually are submitted to government forensic laboratories for extraction, quantification, amplification, separation, and interpretation. As soon as a DNA profile is identified, it may be eligible for uploading to the Combined DNA Index System (CODIS) and compared to the Convicted Offender Index and the Forensic Index. The occurrence of DNA evidence backlog has directly led to thousands of SAKs being unexamined and untested nationwide. 

Current Research

The policy debate over whether SAKs should be prioritized by victim-offender relationship was examined in the research article by Rebecca Campbell, Steven Pierce, Dhruv Sharma, Hannah Feeney, and Giannina Fehler-Cabral (2016) and policy essays by William Wells (2016) and by Kevin Strom and Matthew Hickman (2016). This scholarship discussed the analysis of DNA forensic testing outcomes from multiple jurisdictions. 

Campbell, Pierce, Sharma, Feeney, and Fehler-Cabral (2016) presented an analysis of DNA testing outcomes from 894 previously untested SAKs in Detroit, Michigan. A CODIS hit is the objective of DNA testing for SAKs, and after the DNA Profile is uploaded to CODIS, a comparison will automatically occur to the reference samples in different indexes. The offender and arrestee index contains known DNA profile collected from convicted offenders and arrestees, and the forensic index contains unknown DNA profile collected at crime scenes and on related evidence. If a DNA profile with stranger involved CODIS hits with the offender index, it would link the case with prior sexual assault crimes and convicted offenders. If the DNA profile hits with the forensic index, it will link the case with other cases and unknown individuals. Hence, stranger involved DNA evidence from SAKs has evidential value and potential investigative value for serial crimes. If a DNA profile with non-stranger evidence is submitted, CODIS would indicate confirmatory results of physical contact or sexual activity, which supplements other circumstantial evidence.      

From previous research, Campbell et al. (2016) found results of studies from 602 SAKs from five jurisdictions (one in Los Angeles and four in Indiana), 830 SAKs from New Orleans, Louisiana, and 1,320 SAKs randomly sampled from 10,895 “backlogged/untested” SAKs from Los Angeles that did not present a clear, consistent picture of what practitioners and policy makers should expect when previously unsubmitted SAKs are tested for DNA. Campbell et al. (2016) randomly selected a sample of SAKs from 11,219 rape kits in the Detroit police department storage facility for DNA testing, and compared the CODIS entry rates, CODIS hit rates, and serial sexual assault hit rates as a function of the victim–offender relationship. Four sampling criteria were used to select cases from the population for this study, which included SAKs that never were examined, SAKs still within the statute of limitations (SOL), SAKs still eligible for prosecution, and SAKs including information regarding victim–offender relationships (stranger vs. non-stranger). Stranger-perpetrated assaults were coded as Testing Group 1, and all other assaults were coded as non-stranger Testing Group 2. Within the non-stranger group, victim–offender relationships were subcoded into three categories: known by sight/nickname/street name; friend/associate/family member (but not intimate partner); and current/past intimate partner. 

After random selection, the final sample size for Testing Group 1 (Stranger) was 445; the final sample size for Testing Group 2 (Non-stranger) was 449, and SAKs were outsourced for forensic testing in two laboratories.  The unconditional CODIS entry rates for non-stranger and stranger SAKs were 40.1% and 53.7%, indicating that the difference in the CODIS entry rates was small enough to consider them functionally equivalent. The conditional CODIS hit rates, among SAKs with CODIS entries, were 57.2% for non-stranger SAKs and 65.3% for stranger SAKs, indicating that the conditional CODIS hit rates were equivalent. The conditional serial sexual assault hit rate for non-stranger SAKs with CODIS hits was 17.5% and 32.7% for stranger SAKs, which was significantly different. The unconditional CODIS hit rate for non-stranger SAKs was 22.9%, whereas the corresponding rate for stranger SAKs was 35.1%, which again was significantly different. Finally, the unconditional serial sexual assault hit rate for non-stranger SAKs was 4.0%, as compared to 11.5% for stranger SAKs.

Campbell et al. (2016) found the results from this study are similar to those from Los Angeles (Peterson et al., 2012), which suggests that at the very least, the findings are not strikingly different from another major urban city that has had a large number of previously unsubmitted SAKs. Although this study lacked sufficient variability in the sample to examine the importance of racial composition, the forensic testing outcome results are similar to other cities that have more racial heterogeneity (e.g., Los Angeles; Peterson et al., 2012). At this stage, it’s not known if race/ethnicity might be influential to forensic testing outcomes. The study discusses CODIS hit rates as one of the ways to define the utility of SAK testing, but the findings of this research only could evaluate the utilization of DNA testing for stranger and non-stranger SAKs, not for the actual utilization and other probative values. This study statistically compared forensic outcomes from stranger and non-stranger SAKs to examine variations in the results, but the findings cannot support a policy recommendation that prioritizing the SAKs in cases involving strangers should occur over cases involving non-strangers. 

Policy Analysis

William Wells (2016) responded to Campbell et al. (2016) and raised some considerations about prioritizing SAKs for DNA testing. Different purposes of testing SAKs require a different prioritization system. Each jurisdiction should determine which prioritization system would match its core purpose, and the value of a CODIS hit should help determine when cases are handled by different level of jurisdictions. Wells (2016) stated that certain information is necessary for agencies to obtain reliable data for prioritization, such as victim–offender relationship, victim age, severity of the attack, statute of limitations, and adjudication status. Wells (2016) also presented the assumption that additional workload after DNA testing may produce new information and potential leads. Finally, Wells (2016) illustrated how a broader sexual assault response system benefits jurisdictions by increasing efficiency and effectiveness.

Kevin Strom and Matthew Hickman (2016) also responded to Campbell et al. (2016), first by describing current SAKs submission statues by different regions and programs. They then discussed the value of forensic evidence processing, with selection of prioritization based on the impact of such data on criminal justice case outcomes. Untested SAKs were considered in the context of victims’ rights and due process rights of the accused, but their essay does not put forward any constructive suggestions on the treatment of untested SAKs. Finally, Strom and Hickman (2016) argued uniform forensic policy and standardized procedures should be developed to solve the SAKs backlog issue, but this may be easier said than done. 

References 

• Campbell, Rebecca, Steven J. Pierce, Dhruv B. Sharma, Hannah Feeney, and Giannina Fehler-Cabral. 2016. Should rape kit testing be prioritized by victim–offender relationship? Empirical comparison of forensic testing outcomes for stranger and nonstranger sexual assaults. Criminology & Public Policy, 15: 555–583.
• Wells, William. 2016. Some considerations when making decisions about prioritizing sexual assault kits for forensic testing. Criminology & Public Policy, 15: 585–592. 
• Strom, Kevin J. and Matthew J. Hickman. 2016. Untested sexual assault kits: Searching for an empirical foundation to guide forensic case processing decisions. Criminology & Public Policy, 15: 593–601.