(Image: Courtesy of the Sexual Assault Kit Initiative)

Editor’s Note: In this series “Rape Kits in America,” Forensic Magazine explores the pile of sexual-assault evidence that has been stored on crime lab and police department shelves for decades, even as some serial offenders remain free. From testing to investigation to catching the worst sexual offenders, the series looks at an issue that has yet to be fixed, despite more than a decade of work and a billion dollars spent. Part One looks at the forensic reality of finding DNA profiles in the various swabs and samples that go into a rape kit, through a new study conducted by the Ohio Attorney General’s Bureau of Investigation.

Sexual assault leaves a victim’s body a crime scene, with vital evidence mostly invisible to the eye. Depending upon the attacker’s modus operandi, and whether the victim has a chance to fight back, there could either be a forensic wealth of information leading to an arrest and conviction—or an evidentiary dead-end.

The DNA may be acquired through a vaginal or anal swab shortly after the crime, matching other samples of saliva or semen on the victim’s body. Or a nurse’s efforts could prove especially challenging days after the attack, when a victim has already bathed in the traumatic aftermath—or if the victim cannot recall the specifics of the physical contact.

No two victims—or assaults—are the same, according to investigators. But a new study of what works best forensically has been compiled by a group of scientists at the Ohio Attorney General’s office, in concert with statisticians from Bowling Green State University.

DNA is not assured in every case, they found—only some 65 percent of the rape kits yielded a DNA profile biologically complete enough that it could be entered into the national Combined DNA Index System, or CODIS, according to the findings published in the Journal of the Forensic Sciences.

Now they’re recommending the best forensic strategies possible, based on the thousands of kits they tracked.

Ohio Attorney General Mike DeWine speaks with a forensic analyst from the Ohio Bureau of Criminal Investigation. (Photo: Courtesy of the Ohio Attorney General)

“The primary focus of this study is to determine which samples collected ... tend to yield DNA profiles that are CODIS eligible and if specific case details impact the probability of obtaining CODIS eligible DNA profiles,” they write. “A statistical model was constructed using the case information from the Ohio Attorney General’s Sexual Assault Kit Testing Initiative to aid in the development of guidelines to reduce the cost and turnaround time of SAK processing.”

The sample of 2,500 kits was from the Ohio Attorney General’s massive Sexual Assault Kit Testing Initiative, which “cleared” a backlog of 14,000 kits since its inception in 2012.

While the Buckeye State is not alone in its untested evidence—there are still an estimated 200,000 sitting on evidence-room shelves in the U.S., even now—it has been one of the places where the results have perhaps been most striking. Many of the major cities in the state had backlogs in the thousands—and the rate of CODIS “hits” is higher than virtually every other state. While most other states reported a “hit rate” of between 20 and 25 percent, in Ohio it is 36 percent, mostly driven by three of the biggest cities: Cleveland, Toledo and Akron.  

(The investigative follow-up on those kits producing hits, and how they are actually investigated and then prosecuted, will be addressed in Part Two of this Forensic Magazine series.)

Two scientists discussing the case against Justin Christian, an Ohio man arrested and later convicted of rape, kidnapping and burglary in Cleveland. (Photo: Courtesy of the Ohio Attorney General)

The forensic scientists and statisticians have now attempted to determine the case-by-case factors determining whether the forensic evidence yields enough information to identify rapists in Ohio.

But many kits will never yield that identity—because the devil’s in the forensic details, as they report in the new analysis of kits mostly collected in and around the Cleveland metropolitan area.

Among the findings:

  • More time between the attack and the collection means less success. Most evidence was taken within 24 hours of the crime happening. When those kits were taken within that first day, more than two-thirds yielded a DNA profile that was complete enough to be uploaded into CODIS. One day after the attack still yields a 56 percent rate of getting a genetic profile. But after that one-day time frame is a sharp drop-off: the rate decreases to just 36 percent two days after the assault, and it drops still further after that.
  • Generally, the time between the sexual assault collection and testing—which has proven to be years in thousands of cases—does not have a huge effect on the success rate of DNA profiles. Most of the studied kits were held in storage between 3 and 15 years. The percentage of CODIS-eligible DNA profiles declines slightly from 59.5 percent at the 3 to 6 year range, to 56.6 percent at the 12 to 15 year range.
  • The body location of the swabs is a huge driver on the success of the testing. Swabs from the vagina are by far the most effective way to get DNA from an attack, producing useable DNA profiles nearly 50 percent of the time when a vaginal rape has been reported. But the opposite is true when an anal rape has been reported; when a victim recounts an anal rape it produces a readable DNA profile in only 27.4 of the cases. The scientists write this may be significantly lower because samples taken from that part of the body are prone to exposure to fecal bacteria, which may degrade samples more quickly.
  • Overall, the most successful swab locations were found in the vagina (44.4 percent of samples), the neck and ear area (42 percent), the breasts and chest (38 percent), general swabbing in skin areas that had not been specified in the nurses’ notes (35 percent) and in cuttings from the crotch of the victim’s underwear (30.6 percent).
  • The least successful swabs that went into the rape kits were oral samples (2.3 percent), fingernail scrapings (4.2 percent) and hairs with roots (4.3 percent).
  • Victim age has serious effects on the results. Victims younger than age 8 had the lowest probability of showing a DNA profile from swabbing, while victims between the ages of 23 and 50 had the greatest chances of success.
  • A victim’s report of consensual sex around the time of the reported attack complicates matters—but does mean a greater probability of producing at least one CODIS-eligible profile.
  • Skin and orifice swabs should be completely dried following collection to prevent degradation—and plastic storage containers should be avoided, since they encourage moisture accumulation.
  • In the ideal conditions (immediate collection, victim aged 23 to 50, with a kit immediately brought to the laboratory), the likelihood of producing a CODIS-eligible DNA profile is 74.1 percent, they found.

The results demonstrate that a statistical approach can help guide investigators, since not every kit is going to yield evidence—or “hit” on the rapist—according to the researchers.

“This will subsequently improve the rate at which probative DNA profiles are generated and uploaded into CODIS,” they conclude.

“Rape Kits in America” will continue with Part Two on the “backlog-on-backlog”—when testing kits results in CODIS hits, which end up “sitting in drawers.”