Photograph by Photonica, Getty

Read Caption

Past friends or ex-girlfriends coming forward helped resolve 63 percent of the cleared cases, a new study found.

Photograph by Photonica, Getty

Solving Cold Cases Depends on New Witnesses, Not DNA

Real-life "cold case" murder convictions don't play out like they do in the movies.

Solving "cold case" homicides relies more on the emergence of new witnesses, a study suggests, than on the DNA analyses and other forensic techniques celebrated in crime dramas.

Cold cases are typically ones unsolved after a year or more of investigation. For homicides, their numbers have increased to almost 200,000 such cases nationwide, according to the Center for the Resolution of Unresolved Crime (CRUC) in East Longmeadow, Massachusetts.

The enormous backlog reflects the decline in convictions in recent decades, with 63 percent of murders solved in 2014, compared with 91 percent in 1965, a change attributed to rising drug gang violence. Such crimes are particularly difficult to prosecute because many witnesses, out of fear or gang allegiance, are unwilling to name names at the time of the murder. That means roughly 6,000 murders go unsolved every year.

In response, police departments have expanded cold case investigative teams, spurred by the promise of new forensic technologies, particularly DNA analysis. (See also: "Real-Life CSI.")

But in a new Journal of Forensic Sciences report, researchers take a first systematic look at such cases and conclude that old-fashioned police work and fresh witnesses look like the keys to solving cold case murders.

"Unless you have a good reason, particularly [evidence from] new witnesses, there is little reason to reopen a cold case," says study leader Robert Davis of the Police Executive Research Forum in Washington, D.C. "Cold case investigations are largely an exercise in making people feel good without new information."

Cold case squads might prioritize cases with higher odds of leading to a conviction in their review of old cases, he suggests, rather than working through the files chronologically.

Cold Hard Data

In the study, the team looked at the factors linked to successful convictions in 189 cold case investigations from the files of the District of Columbia's Metropolitan Police Department. The homicides dated back to the 1970s, but most had occurred in the past two decades.

Of those reopened cold cases, 24 percent ended in convictions and 24 percent were cleared by "exceptional" means, instances in which the culprit was already dead, in prison, or had gone missing. The rest remained unsolved.

What made for a cold case conviction? New witnesses helped resolve 63 percent of the cleared cases, the survey found. Often an ex-girlfriend or ex-friend of a murderer came forward years after a crime. DNA matches figured in only 3 percent of the cleared cases. "The worst reason to reopen a case was because of family pressure, if you want a conviction," Davis says.

Other factors make cold case convictions more likely. Study leader Davis suggests that gang-based crimes can in time be solved when gang members are arrested and feel motivated to testify against a rival gang. Another decisive factor in solving a cold case is the identification of a prime suspect within the first 72 hours of the investigation. Research suggests there is a sharp drop-off in cases being solved after that time.

"I think there is a lot of good information here, and we do need a more rational approach to criminal investigations," says criminologist James Adcock of the CRUC, who was not on the study team.

"But at the same time, there is a wider issue of not enough resources being given to do these investigations in the first place," Adcock adds. "We have a lot of resources being poured into finding DNA hits for crimes, which doesn't leave investigators with the time they need to do more thorough investigations."

In other words, he suggests that cold case investigators might turn up those ex-girlfriends or ex-friends of killers if they had more time to pursue them.

Another outside crime researcher, Michael White of Arizona State University's School of Criminology and Criminal Justice in Phoenix, cautioned that the survey results can be said to apply only to Washington, D.C.'s police department, although the results are in line with previous findings. "Clearly, this is a good starting point, but much more work needs to be done," White says.

CSI Effect

Funded by the U.S. Justice Department's National Institute of Justice, the report adds to existing evidence that old-fashioned investigative work, rather than the latest forensic technologies, matters the most in homicide cases, Davis says.

"Most of these people [78 percent] were shooting victims," he says. "That doesn't leave a lot of DNA" from the murderer. Delays in receiving DNA results often lead detectives to discount DNA's impact in solving murders, he adds.

That is a concern, Davis says, because some prosecutors and police officials have worried about a CSI effect (named after the popular police-drama television series) that may affect murder trials. Jurors may have come to expect complex forensic techniques to solve cases, instead of the often messy real-life details of investigations that center on interviews with witnesses.

"If anyone ever wanted to do a show about a real-life homicide investigation, I can guarantee it would be a lot less exciting and conclusive than a TV show," says Davis.

Follow Dan Vergano on Twitter.