NEWARK, N.J. (AP) — Raquel Vanderpool found out the hard way that while fingerprints don't lie, they don't always tell the whole truth.
Vanderpool had been working as a nurse's aide for nearly a decade when a change in state law required her Michigan employer to conduct FBI fingerprint background checks on employees. It turned up a conviction for falsifying a prescription for painkillers years earlier, when she was 19.
Except the charge actually had been dismissed and the record sealed after Vanderpool went through a diversionary program for youthful offenders. That didn't prevent the mother of two from losing her job and, eventually, her house.
"I completed my end, and unfortunately the state failed to remove that from my record," said Vanderpool, now 34. "Because somebody didn't do their job, I lost mine and a whole lot more."
Her story is far from unique. According to studies by the federal Government Accountability Office and the Department of Justice, depending on which state you live in, as many as one-half of the criminal records in the FBI's fingerprint system could have missing or inaccurate information on how criminal cases were resolved.
It worries civil rights and employment rights advocates at a time when job-related requests for fingerprint background checks have soared since 9/11. The nonprofit National Employment Law Project estimated the FBI performed about 2.8 million checks in 2002; an FBI spokesman said recently the bureau has processed roughly 24 million since October.
And the issue of reliability in fingerprint records has come into sharper focus recently as ride-hailing companies battle states and municipalities over fingerprinting of their drivers.
Uber pulled out of Austin, Texas, over the checks, and recently the Chicago City Council left fingerprinting out of its new ride-hailing regulations after Uber and Lyft threatened to leave the city.
The companies' concerns that fingerprint data can omit the disposition, or resolution, of a criminal case — whether it ended in a conviction, acquittal or dropping of charges — are grounded in truth, said Maurice Emsellem, program director for the nonprofit National Employment Law Project, which did a study on the issue in 2013 and updated it last year.
"You have to work with what you've got, but what we've seen is that in states like California where they track down the missing dispositions, then the information is more reliable," Emsellem said. "Otherwise, without that kind of follow-up, you're going to continue to see a lot of workers, mostly workers of color, fall through the cracks."
Most galling for Vanderpool, she said, was that she was able to have the record changed in the state criminal database within a few days.
Convincing the state department in charge of licensing took far longer. Although she still works in the same field, Vanderpool lost her state certification and now earns about $3 per hour less than she earned before her troubles with the fingerprint system began in 2009.
The National Employment Law Project estimates hundreds of thousands of job applicants in fields such as education, health care and security that routinely rely on fingerprint checks have been affected by incomplete or inaccurate records.
For example, the Transportation Security Administration's post-9/11 screening program for port workers led to more than 50,000 appeals of inaccurate or incomplete criminal history information, according to a 2013 NELP report.
It generally works this way: An applicant for a job or license provides fingerprints that are checked against state criminal records and the FBI's database. If the FBI check yields a positive match, the FBI sends the criminal record back to the state, which then determines if the offense disqualifies the applicant. The applicant can challenge and correct inaccurate information.
The FBI created a task force in 2009 to address the problem and set a goal in 2012 of creating a national strategy for reporting case dispositions. That hasn't happened, according to the 2015 GAO report, and although some progress has been made, performance can vary greatly from state to state.
For example, the number of states that reported providing more than 75 percent of their arrest records with final dispositions increased from 16 in 2006 to 20 in 2012, according to the Government Accountability Office. Yet in 2012, 10 states reported that 50 percent or less of their arrest records had final dispositions.
Where you live — or where you're arrested — can make a huge difference. A 2015 Justice Department study found more than 3 million unprocessed or partially processed case disposition forms in 19 states, ranging from 200 in North Dakota to more than 1 million in Nevada.
According to the same study, eight states took one day or less to provide final felony case dispositions to their records repositories, while Missouri took 164 days. Similarly, 20 states took a day or less to input a felony case disposition into the repository's database, while Oregon took more than 100 days.
The FBI is "implementing multi-levels of addressing specific and general disposition submission problems" and is working toward producing a best practices guide, spokesman Stephen Fischer said in an email.