Vol. 16 Issue 11
The Molecular Edge
Information Systems for MDx Labs, Part 1
Part one of this two-part series examines storing molecular diagnostic data in a manner you can use. The conclusion will appear in the January 2008 issue and will address information systems' challenges and opportunities.
The demand for laboratory tests performed by molecular methods is increasing at a rapid rate; several hospitals have reported increases from 6 percent to 25 percent per year.1 In some institutions, this rate exceeds the rate of increase in laboratory tests as a whole.
To meet the demand, more laboratorians are seeking subspecialty certification in molecular techniques. Recently, the American Board of Pathology reported that more than twice as many people have applied for the Molecular Genetic Pathology subspecialty board examination in 2007 than for the last examination in 2005.2
The unique nature and newness of molecular techniques have resulted in a relative lack of adequate information management tools. However, the expectation that this rise in test numbers will continue for the foreseeable future is putting pressure on vendors to supply laboratories with information systems that can meet their specific needs. Several molecular information systems are commercially available, each with varying capability to meet these needs. For those laboratories considering the purchase of a molecular laboratory information system (LIS), the following is a description of information challenges and how information system tools could be used to improve patient care and increase efficiency.
Molecular laboratories perform testing for a broad range of analytes using myriad testing methods that are often complex. For example, most molecular tests require some analysis of the quality and quantity of the sample, specifically its nucleic acid. Reactions may require loading with a specific quantity or concentration of nucleic acid, the volumes and dilutions for which then have to be calculated. Because of these additional data-generating steps, molecular laboratories have had difficulty in re-purposing LISs for their use. Molecular laboratories that do not use an LIS designed with molecular tests in mind have to use a hybrid system by which paper is used to maintain test batch records, calculations and lot numbers while reporting of results occurs electronically. Additionally, the vast majority of tests require multiple steps, technologists, instruments and/or reagents, which makes the generation of a fully functional molecular LIS difficult. Tracking this information in discrete fields would reduce the time and improve the accuracy of troubleshooting assays not performing properly. Searching through paper worksheets for questionable reagent lots or other culprits can be time consuming and prone to error. Appropriate queries of properly and easily stored electronic data could make a determination many times faster.
The lack of standards associated with data exchange between instruments and information systems in combination with the wide variety of molecular techniques available have left vendors with little choice but to develop their own proprietary software that generates and stores the data from the instrument. This difficulty has caused some to advocate for standardization in molecular methods, but the disadvantage of that is some of the most common molecular methods (and therefore most likely to become a standard) are patented and associated with significant royalties. A better way is to standardize how molecular data is stored and exchanged. Some preliminary work has been accomplished to this end for tissue banks.3,4 These standards, as some have argued, should require that all molecular instruments are able to communicate with the LIS via a standard interface5 and that the data should be stored in a set of common data elements.3 Such standards would not only improve connectivity, but facilitate multi-institutional research studies.6
Other areas of complexity include the diversity of the data generated by molecular laboratories. They can include qualitative as well as quantitative results. Reports may include calculated values, including quantitation of target per unit volume of sample, log calculations and textual interpretations. Re-purposed information systems often cannot handle large numerical results (i.e., 500,000,000 IU/mL), log data (8.7 log IU/mL) or scientific notation (5 x 109 IU/mL). In addition, common calculations used by molecular labs are not internal to the system and must be performed on paper (such as determination of DNA quantity from fluorimetric or spectrophotometric analysis or conversion of copies per reaction to IU/mL). The results themselves may require interpretation by a PhD scientist, pathologist or medical geneticist. This is especially true of tests for genetic disorders, hematologic malignancies and solid tumors.
In those laboratories that store results in their LISs solely by transcribing them into text, mining for information can be quite difficult. This is not generally the case when data has been segregated into discrete fields.7 Molecular laboratories have been described as being expensive and technology-driven with the exception of having tools to mine the data they generate,8 but the issue may be more related to how the data is stored in the LIS in the first place.
This becomes particularly clear when performing quality assurance. One example relates to the requirement for a determination and comparison of allele frequencies in the tested population to the general population for each gene tested in the laboratory.7 Another example is the requirement to periodically survey testing results to detect epidemiologically significant changes in infectious disease rates.9
Clinical or Anatomic Pathology?
While most molecular laboratories operate within clinical pathology departments, they have some information needs that have traditionally been aligned with anatomic pathology systems. These include image storage and management of solid tissue specimens.
Image storage is becoming an increasingly important component of all information systems, and an LIS for molecular laboratories is no exception. Many molecular laboratories still take photographs of gels for documentation and store those photos either on paper or in an imaging system that does not communicate with the general LIS. This enables staff to look up more than just the result on a patient when searching the patient's history in the system.
The ability to incorporate images into reports is starting to take off; molecular laboratories will benefit from this capability by incorporating gel photographs as well as images of sequencing or real-time PCR data into reports. Because more molecular testing is being performed on paraffin blocks of tissue or on samples of tissue from a surgical pathology case, the molecular LIS must be able to track and manage these types of specimens.6 Data stored in an easily retrievable manner will ease laboratory workflow, analysis and output of results.
Dr. Carter is director of Pathology Informatics for Emory University Hospital in Atlanta, GA.
1.Carter AB. Increase in numbers of molecular laboratory tests performed per year. Atlanta: Emory University and University of Pittsburgh Medical Center; 2007.
2.Number of applicants taking Molecular Genetic Pathology board examination. In: Carter AB, ed. Tampa, FL; 2007.
3.Patel AA, Kajdacsy-Balla A, Berman JJ, et al. The development of common data elements for a multi-institute prostate cancer tissue bank: The Cooperative Prostate Cancer Tissue Resource (CPCTR) experience. BMC Cancer 2005;5:108.
4.Berman JJ, Edgerton ME, Friedman BA. The tissue microarray data exchange specification: A community-based, open source tool for sharing tissue microarray data. BMC Med Inform Decis Mak May 23, 2003;3:5.
5.Smith TJ. Molecular diagnostics: Outpacing informatics solutions? Advance for Administrators of the Laboratory. 2007;16(6):58.
6.Patel AA, Gilbertson JR, Parwani AV, et al. An informatics model for tissue banks: Lessons learned from the Cooperative Prostate Cancer Tissue Resource. BMC Cancer 2006;6:120.
7.McGinniss MJ, Chen R, Pratt VM, et al. Development of a Web-based query tool for quality assurance of clinical molecular genetic test results. J Mol Diagn 2007;9(1):95-98.
8.Elkin PL. Primer on medical genomics part V: Bioinformatics. Mayo Clin Proc 2003;78(1):57-64.
9.Amadoz A, Gonzalez-Candelas F. epiPATH: An information system for the storage and management of molecular epidemiology data from infectious pathogens. BMC Infect Dis 2007;7:32.