"There is no way to plan for training unless you control your budget," Sager says. "You can't have authority over your lab and over hiring and firing and the purchase of needed equipment unless you have complete control of your budgetary process."
Quality Assurance Quality assurance sounds like bunk culled from a Met Life TV ad, but the concept is central to the work of a crime lab. Criminalists produce hard data from the raw material of evidence, and the value of their data depends on their experience and knowledge, the accuracy of their equipment, the purity of their materials, and the rigor behind their techniques.
But all this hard data goes mushy if the equipment and materials are faulty and lab techniques are careless, which appears to be the case at the SFPD lab. The inspectors found the lab's criminalists to be generally qualified, but the absence of a "quality assurance" system sullied all of the lab's work.
In addition to the SFPD lab's horrendous chain of evidence problems, the inspectors decried other quality assurance deficiencies. The lab kept no record of staff training, nor did it document lab workers' skills with routine "proficiency" exams. Under the category of maintenance and calibration of equipment, inspectors learned that the "written procedures are neither complete nor well understood" by personnel, leaving it to anyone's guess whether the staffers knew what they were doing.
Every crime lab maintains essential "reference standards" against which illicit drugs are compared. At the SFPD lab, reference standards were "obviously degraded," the report states, and the reagents -- chemicals used in conducting tests -- were not routinely checked for freshness.
A well-run lab maintains the integrity of reference samples by centralizing them; at the SFPD lab, it appears that each technician had his own standards.
"This is not good," says Sager. "There is a real possibility of cross-contamination."
Ishii rates as low the likelihood that the SFPD lab sent innocent people to jail by scored false positives when testing for drugs. In fact, Ishii believes that it is more likely that the SFPD technicians failed to detect other drugs present -- hallucinogens, for example -- and let the guilty walk. "A small percentage of cases need more sophisticated testing," he says.
Sager is more skeptical. "Contamination could create false positives," he says. "Contaminants can come from dirty glassware, general dust in the lab, your skin, or your lab coat." He adds that the poor reagents in the lab could also produce false positives.
Assistant District Attorney John Dwyer, Smith's designated crime lab authority who read the report this week, is sanguine rather than surprised by its findings, saying that the DA, the defense bar, and judges have always known that the lab didn't perform confirmatory tests with machines and that lab notes were written on evidence envelopes.
"That unit has been an orphan in the Police Department for 27 years," he says. "It's not a priority to anybody."
Dwyer expresses confidence in the accuracy of the lab's work and the rigor of its chain of evidence practices, and notes that the lab now keeps paperwork separate from the evidence. He declines, however, to comment on the security issues raised by the report.
Retired lab director George Ishii is sanguine about the SFPD lab, too, but for different reasons.
"I think that out of this, something good will happen in San Francisco," he says. "One thing SFPD has to realize is that crime labs are expensive because you can't settle for 'almost' because you're sending citizens to prison. It's not fair to have anything less than the best."
Things will have to get better at the lab because it's hard to imagine them getting any worse. The inspectors gave the SFPD lab a score of 51.3 out of a possible 100 in the "essential criteria" for a good lab. To win accreditation, a lab must score 100.
Which brings us back to the much maligned Allison Lancaster: If she deliberately falsified drug evidence reports -- as her accusers maintain -- she had a leg up on her colleagues. At least she knew precisely what she was doing.
