Outsourcing Sample Testing

John Butler

Before any new technology is brought to bear on precious and irreplaceable samples like the victim remains in the World Trade Center disaster, validation testing must be performed to verify that it is capable of producing reliable results. Beyond the core issue of test reliability, we also assessed the results of new methods to determine their power to raise a profile to the level of an identification and for issues of compatibility—linkage—with other markers.

The best time to establish a good relationship with a vendor is during the planning phase for a mass fatality incident. Although outsourcing testing can be expensive—from $30–$60 for a reference buccal sample to hundreds of dollars for a disaster sample—it may be necessary or more effective to have another laboratory test some or all of the samples. For example, an outside laboratory may test certain sample types—family reference, personal items, or disaster samples—or a portion of the samples for quality control or conformation testing. An outside laboratory may be used for certain types of testing technologies—mitochondrial, single nucleotide polymorphism, or new technologies, for example—or for extraction and data analysis only. On the other hand, the entire testing process, from accessioning to data analysis, may be outsourced. Even in this situation, however, the managing laboratory is ultimately responsible for the quality and accuracy of the data.

The laboratory’s contracting office can ensure that contracting regulations are followed, and discussing this issue with a vendor in advance may prevent later problems. In an emergency situation, an agency may not be required to follow the typical, lengthy contracting procedures to obtain the best value but, rather, may be able to initiate contracts without competition. If a managing laboratory’s normal contracting process is not followed, however, it is very important to document the new process to reduce the potential for future problems.

To assist in the response to a mass fatality incident, a laboratory may contract with a current or new vendor, or seek help from another government agency. If the managing laboratory is already contracting with a vendor whose quality is satisfactory, it may be advantageous to use that vendor to process mass fatality samples, assuming the vendor’s capabilities and capacities can support the laboratory’s needs. For example, does the vendor have the capacity (e.g., equipment and staffing) to meet throughput and turn-around-time requirements, even while working on other contracts? If not, is the vendor able and willing to interrupt its regular work to take on the testing of mass fatality incident samples?

Does the vendor have experience in successfully typing samples from a mass fatality incident? The managing laboratory director needs to keep in mind that the volume may be larger and the samples more challenging than the vendor laboratory has previously experienced.

Meeting turnaround requirements in the face of expectations from victims’ families, the media, and policymakers likely will pose other challenges—and the laboratory director should not be afraid to ask for what is needed. For example, if a laboratory director is relatively inexperienced in contracting for testing services, he or she should enlist the support of laboratories that have extensive outsourcing experience. See appendix F, Issues to Consider When Outsourcing Reference Samples, for a discussion of issues that a laboratory director may want to consider when outsourcing sample testing to a vendor laboratory.

Government forensic laboratories may be able to provide assistance in a mass fatality incident identification response. Each agency that is helping in a mass fatality identification effort must understand its own and others’ roles and responsibilities, the scope of tasks, and the duration of expected services. It may be helpful to prepare a detailed Memorandum of Understanding (MOU), including a project point of contact for each agency.

Whether a laboratory director obtains the assistance of a private vendor laboratory or another government agency, it is important to review the testing procedures to be used. If more than one testing laboratory is used, for example, testing systems and results systems must be compatible with each other.

It is also critical to address how the samples will be numbered and how the data will be returned to the managing laboratory. The software package that evaluates the data is vital to managing this data exchange, and an MOU or vendor contract should specify how these issues will be handled.