Return to CAP Home
Printable Version

  Hang on or move on? How to assess an LIS

title

 

 

 

cap today

 

 

November 2006
Feature Story

What should laboratorians expect from their laboratory information system, and what functionality should LIS vendors provide to address growing niche markets? CAP TODAY asked experts in the field for their opinions. Here's what they had to say.

CAP TODAY: When and how do you know that you’ve stayed with a legacy laboratory information system too long?

Gilbert Hakim, CEO, SCC Soft Computer, Clearwater, Fla.: You may have stayed with your legacy LIS too long and be saddled with a product that will soon be sunsetted if the vendor’s focus begins to shift from the primary product to other products, the vendor raises maintenance costs beyond that worthy of the services provided, or the vendor no longer offers upgrades, just software patches.

Cost of ownership is high with a legacy system, while productivity and flexibility are extremely low. A legacy system requires a greater number of support personnel and lab technologists than a modern LIS, so when your biggest challenge is trying to find enough full-time equivalent employees, you might want to consider making a move.

Another sign is difficulty when integrating other systems due to LIS limitations. Indicative of their name, legacy systems do not use state-of-the-art technology, such as Web services, and do not easily support robotics and lab workflow redesign. This often presents business limitations for the lab and keeps it in a cost center mindset instead of allowing it to become a potential revenue source for the hospital.

Curt Johnson, vice president of sales and marketing, Orchard Software, Carmel, Ind.:

First, DOS-based systems are a thing of the past, so if you are still working on a DOS-based system, be aware that you are working with an antiquated system and missing out on the many benefits of Windows-based technology, not to mention all the programming features that will improve efficiency and productivity in your lab.

Secondly, many legacy LIS vendors have ceased investing in their LIS to keep it up to date, which may be a sign that the vendor intends to sunset the system in the near future. If your LIS vendor is not sending out upgrades annually, it may be time to look for a new vendor.

The absence of upgrades is one reason many laboratories turn to middleware to fill in the gaps of their legacy LIS. Conversely, if you are looking to middleware to fill a gap in your LIS, it may be a sign that you’ve stayed with the product too long.

Jay Abrajano, CLS, MT(ASCP), product manager, information solutions, Aspyra, Calabasas, Calif.: Laboratories considering departmental improvements eventually evaluate their LIS as an integral piece of their daily operations. Laboratory managers should evaluate their systems and vendors with regard to future direction of the company, product functionality, and product development. Key steps in the information-gathering process include asking such questions as, Does the next release address up-and-coming regulatory and functional issues? Is the newest version in step with the latest platform, database, and technological options?

Consider the vendor’s vision for its products and its client base. Does the vendor contact clients about the upgrade path within reasonable time frames? Does the upgrade path include a complete audit/review of the laboratory’s current workflow? Take advantage of upgrades to review and make process or procedural improvements, as well as to incorporate new functionality that will benefit the laboratory and patient care. Consider quality of support as well as the amount of resources and cost required for a new system implementation.

The answers you receive to such questions will indicate whether you should rethink your current system vendor.

J.P. Fingado, vice president and general manager, Cerner Corp., Kansas City, Mo.: When your business growth is impeded by technology or automation, it is already too late. In an ideal situation, a lab will install a new LIS 12 months before tackling major growth opportunities. When a legacy LIS is no longer adequately serving laboratories, it is usually falling short in the areas of molecular diagnostics or resourcing or because of limited technology.

The ability to support the complex data structures and workflows of molecular diagnostics cannot be handled in numerous legacy laboratory information systems, as well as in many of the current systems. Many of the legacy LISs are limited by older technologies. Moving to an open-architecture format provides for better interconnectivity with other systems and greater research capabilities, and it allows users to integrate imaging technology throughout the laboratory.

Furthermore, shifting to a supplier-hosted model frees up laboratory staff to help grow the business. It provides for guaranteed uptime and locks in a pricing model that eliminates the risk associated with hardware purchases and staff replacements over the term of the contract. Integration between automation devices and the LIS also increases the lab’s efficiency and data-management capabilities.

Raymond Aller, MD, director of bioterrorism preparedness and response, Los Angeles County Public Health Acute Communicable Diseases, and contributing editor, CAP TODAY: You know that you’ve hung on to your legacy LIS too long when:

  • your LIS’s limited functionality interferes with your staff’s ability to serve the client.
  • you can’t get spare parts for the hardware.
  • the vendor can’t or won’t correct certain problems with the system, such as data corruption or lost records.

However, jumping to a different LIS because you think your present system is obsolete or doesn’t have the latest hardware and software or the most fashionable database can be a serious mistake.

In the early 1980s, LIS vendor X was selling systems on the claim that vendor Y’s database was obsolete. Today vendor Y continues to grow and serve its customers, while vendor X has sunsetted its system. Numerous successful information systems are based on the same database vendor Y uses.

CAP TODAY: What efforts should LIS vendors be undertaking in such emerging markets as molecular diagnostics, drugs-of-abuse testing, and home testing?

Curt Johnson (Orchard Software): Laboratory information system vendors address molecular testing in several ways. Most systems can accommodate molecular methods for improving the specificity or sensitivity of existing analyses. However, in the case of DNA testing, interpreting results can be very complex, and there are a limited number of testing centers. Few laboratories have the ability to deal with the complex analyses or large amounts of raw data, or to interpret the results. A few LIS companies have established relationships with these testing centers to develop software that will spread testing to a wider range of labs. In doing so, the LIS will go beyond merely presenting results and will provide additional information to physicians and others working with affected individuals.

Drugs-of-abuse testing largely has become cyber-testing. The LIS must support Web-enabled communications that provide prompt and secure results to many different off-site users, including medical personnel, the police, and the courts.

Home testers make up yet another family of cyber-users. Again, the LIS must provide remote users of differing knowledge levels not only data, but also other types of useful information. The LIS can display testing trends, provide alerts, and, with physician guidance, suggest treatment actions. Many of these activities are already second-nature to an LIS. It is moving beyond the confines of the laboratory that creates the greatest challenge.

Gilbert Hakim (SCC Soft Computer): Molecular diagnostics testing will become more prevalent as technologies are improved and packaged in kit form and as users are able to perform this testing at the community hospital level. Experts estimate that up to 35 percent of tests will be performed at the molecular level in the next five years, replacing tests performed in hospital laboratories.

The user needs a consolidated picture of all the patients’ laboratory results, including clinical lab, microbiology, anatomic pathology, blood bank, and genetics. Patient and specimen identification is easiest with a consolidated, integrated system. The pathology and genetics modules need to be able to easily integrate results from all laboratory departments and render a consolidated diagnosis. Molecular testing is rapidly growing in volume and can provide a lab with a high return on investment.

Raymond Aller, MD (Los Angeles County Public Health Acute Communicable Diseases): Well-designed laboratory information systems can provide considerable functionality to these emerging markets by creatively adapting capabilities that are already built into the systems.

Wherever possible, vendors should add support for new testing types as a seamless extension of existing functions, rather than creating new modules. Applying this to a molecular diagnostics module, for example, the feature-capturing results of a large array of molecular probes can be used in the future for a large array of chemical sensors in a sports-doping application. For home testing, the challenge is to make the user interface as simple and patient-proof as possible. The need for simplicity and clarity also applies to use of the LIS.

Brian Keefe, MT(ASCP), director of product marketing for clinical systems, Psyche Systems Corp., Milford, Mass.: New types of testing will place new demands on laboratories, requiring them to consider new equipment, new recording and reporting methods, and new approaches to client service—all of which have implications for the information system. LIS vendors must support the laboratory in its drive to capture these opportunities by moving quickly to address interfacing needs, data capture and retrieval, and client-reporting capabilities in these emerging markets.

It is critical that vendors keep an open dialog with laboratory professionals about new opportunities and other changes affecting the industry.

J.P. Fingado (Cerner Corp.): Laboratory information systems suppliers should be rapidly developing innovations in each of these areas to help meet demand. Molecular diagnostics is a complex area that requires very specific data management and workflow design, and it is impossible to force these requirements into an existing LIS. The development of a molecular diagnostics solution combined with a complex ontology for managing data will allow the solution to stay current with advancements in this field.

CAP TODAY: What factors should labs interested in rules or algorithms consider when deciding whether to purchase a rules engine from an LIS vendor or a middleware vendor? And what are the benefits of choosing one type of vendor over another?

Raymond Aller, MD (Los Angeles County Public Health Acute Communicable Diseases): Can an interface engine from an LIS vendor access all of the data elements in the database? If so, then going through an LIS vendor may offer a significant advantage.

A middleware-based rules engine will only be able to drive decisions off data elements that are sent to the middleware. On the flip side, middleware vendors have often devoted more attention to creating a fully functional rules engine. Middleware-based systems have to deal only with the software complexity or interaction present within that system. LIS-based systems may have to contend with interactions involving other areas of the LIS, as well as with other clinical information system modules operating off the same database.

LIS-based rules systems will likely be better attuned to the way results and other data are represented within the database. LISs may send novel or unexpected formats and codes to the middleware system that the middleware rules engine cannot handle.

One should also consider the training program, telephone support, and user groups available from each type of vendor. Over the past two decades, the principal impediment to implementing rules-based systems has been that no one in the laboratory feels comfortable enough using them.

Gilbert Hakim (SCC Soft Computer): Facilities running a modern LIS probably have an integrated rules engine, so there is no need to add an additional level of complexity. But with a legacy LIS, middleware can be a cost-effective way to add advanced functionality, such as rules and instrument grouping. However, by using middleware, you are limited to the instruments on that piece of middleware, or data manager. Rules may only apply to tests run on those attached instruments. Data retention on data managers and legacy systems is much shorter than on a modern LIS, which may preserve clinical lab results for several years.

Before deciding whether to purchase an LIS with a rules engine versus middleware with rules, consider the following:

  • Does the middleware application require additional hardware and interfaces to function with the existing LIS?
  • Can the rules engine middleware application access the history of the patient or the master patient index?
  • Can auto-verification access all instrument results and patient diagnoses?

Brian Keefe, MT(ASCP) (Psyche Systems Corp.): Rules and algorithms should be standard functionality in any LIS. However, if a laboratory must use a system that lacks this functionality, third-party software can fill the gaps.

It’s important to consider the third-party vendor’s experience working with many different types of LISs to ensure that data can be transferred properly between the systems. When using third-party software, data must reside in two places to apply rules to various parameters recorded in the LIS. For this reason, data integrity could become an issue if the systems aren’t working together properly.

Additionally, many rules engines cannot access data once they’ve released it, which can also make it difficult to run algorithms on such information as historical patient data. When selecting a rules engine, the laboratory must note how much data redundancy is needed between the third-party software and the LIS to run various rules and algorithms. Also important is whether the software is user friendly or if it requires some technical proficiency to create a new algorithm.

J.P. Fingado (Cerner Corp.): The seamless integration of a rules engine with the LIS provides greater flexibility and promotes increased quality of care and patient safety. Attempting to use middleware from a third-party supplier rather than the LIS vendor will result in the need for complex interfaces, which will increase support costs and limit capabilities as compared with an integrated solution.

An integrated rules engine achieves greater benefits because it offers a more complete laboratory record for automated assessment, whereas middleware has a narrower view. Where the LIS and rules engine share a unified platform with the patient electronic medical record, the scope of clinical data available to the rules engine far exceeds that available to middleware solutions. The middleware can contribute highly sophisticated analytical data but often cannot “see” the patient’s full status, thereby limiting the effectiveness of the rules.

Jay Abrajano, CLS, MT(ASCP) (Aspyra): Increased use of auto-verification and goals to improve efficiency and reduce errors have made rules a necessity. Laboratories should ascertain if the rules engine handles the rules that are needed and whether they are flexible, scalable, and easy to modify and activate. Advantages of using rules within the LIS are that they are already in the system and linked to additional result review/ release functionality. Some of these features include critical/abnormal call reporting alerts, tracking/monitoring reports, date/time/technologist stamps for all actions available for historical inquiries within one system, and result comments usually entered in the LIS.

Middleware vendors designed rules and reports for review/release of instrument results. However, LIS vendors have increased their focus in this area and incorporated much of this functionality and flexibility into their applications.

Regardless of whether you are leaning toward an LIS or middleware vendor, you should research interface, implementation, and maintenance processes and costs for each. Note that with the development of rules in LISs, vendors have adjusted interface pricing to be more attractive.

In evaluating middleware, laboratories should research their true rules requirements and those already available in their LIS. Furthermore, they should consider any functional losses within the LIS in terms of their current result review standard operating procedures.