Maybe this isn’t such a good idea.
A Houston police officer pulled Barry Demings over as he headed to work in Beaumont and plucked a spot of white powder off the floorboard of Demings’ year-old Ford Explorer.
Demings had just detailed the SUV – and wondered later if a speck of soap upended his life.
“I never even saw it,” he said, explaining how the officer dropped the speck into a small test kit and said “it came back for cocaine.”
Demings was charged with felony drug possession based on the results of the primitive test that costs about $2 and has been found to have a high error rate. He was told he could face a sentence as long as 30 years based on old prior convictions – no one mentioned waiting for a crime lab to verify the officer’s roadside result.
He insisted he was innocent but got scared and accepted a plea deal. He lost his job, his girlfriend and his Explorer. Upon release, he decided to leave Texas behind forever.
In 2015 – seven years later – the Harris County District Attorney’s Office notified him that Houston’s crime lab found no cocaine in the sample. He filed a writ of habeus corpus with the Texas Court of Criminal Appeals and was finally exonerated.
He is among 298 people convicted of drug possession even though crime lab tests later found no controlled substances in the samples, according to a far-reaching audit of drug cases by the Harris County District Attorney’s Office. So far, 131 of them, like Demings, have had their convictions overturned in cases that go back to 2004. About 100 other cases remain under review for potential dismissal.
In all 298 cases, prosecutors accepted both felony and misdemeanor plea deals before lab tests were performed. The $2 roadside tests, which officers use to help establish probable cause for an arrest, cannot be used at trial as evidence under Texas law.
The Harris County audit of drug possession convictions and related lab results going back to 2004 was launched in 2014 by Inger Chandler, an assistant district attorney in charge of the DA’s conviction integrity unit, after a reporter from the Austin American-Statesman called her about reversals of several drug convictions by the Court of Criminal Appeals.
The following year, Harris County District Attorney Devon Anderson changed her policies and directed prosecutors generally to stop accepting guilty pleas in felony drug cases before receiving lab reports confirming the evidence. Plea deals are still accepted prior to lab testing in misdemeanor drug cases, and in some felony cases in which jailed defendants can qualify for probation.
The forensic evidence problems uncovered by Chandler’s unit began around 2005, when Houston’s city crime lab – then overseen by HPD – lost several staff members and simultaneously saw a huge increase in drug cases, which created a backlog.
Lab officials implemented a triage system for drug testing with the DA’s office: Drug cases slated to go to trial would get processed first. For defendants who had accepted plea deals, the crime lab would later go back and test samples, often months or years after the guilty plea had been entered.
Chandler’s audit of wrongful convictions has been possible because the Houston Forensic Science Center, formerly HPD’s crime lab, preserved and tested the evidence even in the plea deal cases.
“We were keeping the evidence, and with the agreement with the District Attorney’s Office that we would continue to process even if it was pled,” said James Miller, manager of the center’s controlled substances section. “Because we both understood there was always the possibility that the substance may not actually be illegal.”
So far, prosecutors have identified and examined 456 flawed cases. Of those, 298 people had been convicted despite having no illegal controlled substances in their possession at all. In 29 of the 298 wrongful convictions, there had been no filing for relief because a defendant declined to pursue the case or faced other legal obstacles.
In other cases among the 456, the types or quantities of controlled substances were misidentified or there was too little evidence left to perform a confirmatory test.
About 78 percent of the 456 flawed cases came from the Houston Police Department, which still uses roadside tests that were developed in the 1970s. Chemicals in small vials turn colors when exposed to cocaine and other illegal drugs but can be easily misinterpreted by officers and can have high false positives, Miller and other experts said.
Emphasis mine. This article is a followup to a much longer ProPublica piece that explored the history and background of these roadside tests; another story, about the chemist who created these kits in 1973, is here. You should read them both – I don’t know about you, but I had no idea about any of this before now. We could have a debate about whether it’s reasonable for police officers to conduct roadside drug tests like this, but the high error rate for this test, which hasn’t been updated sine the 70s, makes it a particularly poor reason to hold people in jail or encourage them to plead out on a charge that is based on a crime that may never have existed. The point, again and again and again, is that there are way too many people in our jails who should not be there. The cost of this, both to the people who have been subjected to this and to us taxpayers who foot the bill for it, is unacceptable. When are we going to do something about it?