The Houston Forensic Science Center seeks to streamline fingerprint analysis by cutting unnecessary testing of prints irrelevant to the crime. (Photo: Courtesy of the Houston Forensic Science Center)

When investigating a burglary, is it necessary to do a full comparison of a homeowner’s fingerprints on a broken front door? How about the dozens of latents left on the knob by the homeowner’s spouse, children and friends?

The Houston Forensic Science Center says there is a better way. They have found a common sense workflow to focus on the important prints at a crime scene—thereby reducing both unnecessary work and the backlogs that come with it.

“We have backlogs in latent prints, and that certainly is a driver for why we’re trying to do things more efficiently,” said Peter Stout, the president and chief executive officer of the HFSC. “But even if we didn’t have backlogs … we need to be looking at doing this as efficiently as we can, regardless.

“It is always an exercise in finding the route to the most efficient answer—by efficient I mean both right and quick,” he added.

The forensic workflow at the HFSC traditionally had two latent fingerprint examiners conduct complete comparisons on any potential matches returned by the Automated Fingerprint Identification System (AFIS). That meant every print at a burglary prompted a painstaking process of matches that took weeks before being presented to a lead detective—only for a majority to have no investigative value, like the homeowner’s prints in a burglary case.

The new system prioritizing the matches of value has now streamlined the entire fingerprint unit, according to HFSC officials.

A whopping 86 percent of the full comparisons analysts completed prior to the change were probably unnecessary, the forensic experts estimate.

The change is in processing what comes out of AFIS. Instead of immediately starting full comparisons of all the on-screen matches, two analysts instead send the preliminary list of names and IDs right to the investigators.

Those investigators send back only the names they think would be important.

The protocol change cuts out a large majority of needless matches, said the experts.

A process that previously took up to several months on a case can now potentially be done instead in mere days, the HFSC leadership explained.

Prior to the change, the fingerprint experts didn’t even know just how much of their work was being used to make arrests or prosecute criminals, said Tim Schmahl, manager of the HFSC’s latent print section.

“A lot of the times we don’t know what happened to the comparison—whether it was worth anything to the investigation or not. (So) we asked, what if we just stop it right there on the screen?” said Schmahl. “(We told them), ‘You tell us if this name we gave you is important to this cold case, and we’ll work it all the way through.’”

The transition came gradually. First, the fingerprints unit changed the workflow for cold cases in the early part of 2015. Hardly any requests came back from the lead detectives on the various cases—meaning that the prints were mostly irrelevant to the investigations, said Schmahl. Property crimes—which make up about 95 percent of the fingerprint backlog at HFSC—were added next. But the system proved itself quickly, and all no-suspect crimes (including violent crimes) were included by mid-2016.

Some of the print analysts had concerns about the new workflow, the HFSC leaders conceded. But it has been more accepted as time has gone on, due to its efficiency, they added.

“It has been a productive discussion, certainly they’ve had their heated moments,” said Stout. “But it comes down to the question of, how do we get this stuff out the door?”

One of the concerns was the possibility of getting the wrong person, based on the expedited process. But the investigative lead report includes clear highlighted language telling detectives that the AFIS matches are only an investigative lead until the full comparison is returned, the HFSC said.

The new way of doing things has essentially doubled output, they said. For the second half of 2016, the section completed 1,477 latent print comparison requests, compared with 763 requests with the same time frame in 2015.

Concurrently, more long-term work is getting done. The latent fingerprint backlog has shrunk from 3,700 requests to 2,800 requests over the last year, said Stout.

The workflow concept has also caught the attention of some other law enforcement agencies that are seeking to cut back on unnecessary comparisons. Schmahl said the Royal Canadian Mounted Police, the Dallas Police and Raleigh’s department have all expressed interest. Schmahl will also present some of the findings of the latent fingerprint unit at the August conference of the International Association for Identification.