Fourth Annual DNA Grantees' Workshop

Tuesday, June 24, 2003

MORNING SESSION

Management Perspectives on Validation
Susan Hart Johns
Biography

MS. HART JOHNS: Everybody talked about validation, but as a manager I want to explain my perspectives on validation. I went to my group and asked them to validate robotics for use in the indexing laboratory. Months later we came up with a procedure and implemented it, but in 30 days we shut it down, even though we had implemented a perfectly valid procedure—validation meaning that results would be acceptable in court. Sozer: Slide 9

Why did we shut it down? We had a validated procedure all right, but not by my definition. I want to produce at least 150 samples per week for a minimum of 4 weeks with a 10 percent repeat rate. In the first example, we had a perfectly great result when we went to court, but 50 percent of the samples that were run had to be repeated, so it was useless for a high-throughput laboratory.

From a manager's point of view, I need to have something that's going to work for the purpose it is intended for. When we do these things, we have to make sure we're defining very clearly what we want. When we talk about validations, it's not always clear.

I need a procedure. I need something in writing that I can give to 60 different people and get one result, because if I give it to 60 different people and get 60 different results, then what good is the procedure.

The procedure must address quality assurance issues. I have to monitor this if it's in an operational setting. I have to address safety issues and interpretational guidelines for a sizable laboratory system with multiple laboratories and multiple analysts. We can't have true independence in reporting. There has to be some agreement, some consensus, among our laboratory personnel so that a report going out on our letterhead is reasonably the same when whoever signs it.

I also need report wording guidelines. As a manager, that helps me monitor and review things from an operational perspective. All of these things need to be considered when we talk about the procedure that we're going to implement.

Then we can talk about the implementation itself. If you want to put this online, I need to know how much it is going to cost, what instrumentation needs purchased, when it needs purchased, and how many instruments are needed.

I have to talk about technology transfer. How are we going to train people? I have to ask about lab criteria that require that I have training for procedures in the laboratory. In our system, we pretty much always pilot test on a small scale, so we don't implement a new procedure in every lab with every person. We'll pick one laboratory and implement it on a small scale and try to work out the bugs before we get all eight or nine laboratories involved.

I need to address competency testing to be prepared when ASCLD (American Society of Crime Laboratory Directors) staff ask if individuals were competency tested prior to the implementation of the procedure.

I have all of those issues to address when it comes to the whole implementation plan.

DR. SOZER: Thank you, Susan.