Thursday, July 24, 2008 Improving Quality in Laboratory Medicine Home Page    Contact Us
Links:

Skip Navigation Links
CDC Division of Laboratory Systems
Healthcare Innovations Exchange
Guide to Community Preventive Services
2007 Status Report on Laboratory Medicine

Identifying Best Practices in Laboratory Medicine

CDC and Battelle are conducting a pilot test of methods for evidence-based evaluation of best practices in laboratory medicine. This pilot test will add to a larger effort spearheaded by DLS that has three major goals:

  • Improve patient safety and health care outcomes by improving the use of laboratory testing in diagnosis, monitoring, and management of disease
  • Reduce redundancy and waste in laboratory services
  • Provide tools that laboratories can use to improve quality of service to clinicians and patients

A comprehensive approach to evaluating best practices is needed in laboratory medicine because laboratory test results influence a large percentage of diagnostic and patient care decisions in modern medical practice. Conversely, quality in laboratory medicine is affected by multiple processes and factors, many of which are not under the control of laboratorians. Laboratory errors and inappropriate laboratory testing (underuse, overuse, misuse) thus have ramifications that extend beyond the walls of the laboratory and affect many aspects of medical care. Opportunities for error exist at many points in the total laboratory testing process, including selecting and ordering appropriate clinical tests, obtaining patient specimens, reporting results to clinicians, and interpreting laboratory results. Error rates are higher in the pre- and post-analytical stages than within the laboratory itself.

In response to the systemic shortcomings in health care quality identified by the IOM and others, evidence-based recommendations, guidelines, and quality measures have been developed in many fields of medicine. The Agency for Healthcare Research and Quality (AHRQ) compiles this information in its National Guidelines Clearinghouse and National Quality Measures Clearinghouse. Many disease-specific guidelines and quality measures for screening, diagnosis, treatment, and management include recommendations for laboratory testing. Laboratory medicine guidelines and measures that satisfy AHRQ’s inclusion criteria may be included in these databases, however relatively few have met the criteria.

In laboratory medicine, previous efforts to develop guidelines, standards, policies, and best practice recommendations served specific fields or professions in the laboratory community. The College of American Pathologists (CAP), the Clinical and Laboratory Standards Institute (CLSI), and the Clinical Laboratory Management Association (CLMA), among others, have developed approaches to recommending and disseminating quality practices. Recognizing the need for a more systematic, comprehensive, and transparent approach to identifying, evaluating, and recommending best practices for the field as a whole, CDC convened an expert multi-disciplinary Laboratory Medicine Best Practices Workgroup in October 2006. DLS charged the Workgroup to identify systematic review frameworks and methods that are currently in use in medicine and public health, and to adapt these approaches to the field of laboratory medicine. The Workgroup’s efforts were supported by a team of researchers from DLS and Battelle.

Phase I
In 2006 and 2007, the study team developed methods for evaluating candidate best practices which included an analytic framework, search strategy, practice and study inclusion criteria for selecting practices and developing an evidence base, and systematically abstracting and summarizing the evidence.  Reference sources included the US Preventive Services Task Force and the Task Force on Community Preventive Services, as well as others, for assessing the strength of individual studies and of a body of evidence that have been used to develop evidence-based recommendations.

To test these methods, the research team and Workgroup conducted a “proof of concept” exercise that focused on practices to reduce patient/specimen identification errors in laboratory testing. As the proof-of-concept exercise proceeded, the limitations of the scant peer-reviewed published evidence base for patient/specimen identification practices (and laboratory medicine practices generally) became increasingly apparent. Workgroup members believed that laboratories conduct case studies, evaluations, and quality improvement studies. Typically, these studies are not published and thus are not accessible by conventional bibliographic searches. The Workgroup recommended adding an “investigational component” to the process to identify, obtain, and evaluate unpublished evidence from practitioners, laboratories, and expert groups. As recommended by the Workgroup, the criteria and methods for evaluating this expanded evidence base should parallel those applied to evidence from published sources, and should be specified beforehand.

The final report of Phase 1 is available: Laboratory Medicine Best Practices: Developing an Evidence-Based Evaluation Process: Final Technical Report.

Phase II
For the project’s second phase, CDC and Battelle are seeking laboratories willing to share unpublished studies or assessments for a pilot test of the investigational component, evidence review, and evaluation methods. Exploratory work suggests that relevant unpublished evidence is produced from quasi-experimental studies (e.g., non-blinded trials), surveillance of error rates or other such objective outcomes over time, case studies, failure mode and effects analysis (FMEA), Six Sigma studies, continuous quality improvement (CQI), and other kinds of “in-house” studies. The pilot test will focus identifying and evaluating specific practices in two topic areas: patient/ specimen identification and communication of critical laboratory test values.

How the Pilot Test will work:

  1. Battelle and CDC will assemble two Expert Panels with appropriate training and experience in evidence evaluation and laboratory operations, one for each topic area. The Expert Panels will define specific parameters for research questions and practices to be evaluated in each topic area.


  2. CDC will invite laboratories with previously completed unpublished studies or assessments in the two topic areas, to participate in the methods pilot test. These laboratories will provide unpublished evidence on practices that will be assessed using the same methods as the published studies.


  3. CDC and Battelle scientists supporting the two Expert Panels and pilot study sites, will identify, assemble, abstract, and evaluate evidence—published and unpublished—concerning the effectiveness of the candidate practices.


  4. The Expert Panels will review the entire body of evidence for each practice, published and unpublished, and draft a summary evaluation of the strength of evidence and effect size to support an evidence-based recommendation.


  5. For the purposes of the pilot test, the Laboratory Medicine Best Practices Workgroup will act as a “Recommending Body.” The Workgroup will review the draft summary statements and supporting documentation provided by the Expert Panels, to decide on recommendations concerning each practice (i.e., strongly recommend, recommend, no recommendation for or against, recommend against).

Data provided to CDC and Battelle
No “new” data collection is requested from participating laboratories; the pilot test will focus entirely on previously completed data collection and evaluation efforts that provide information about the effectiveness of specific practices. No patient-specific data or personal health information will be requested. Individual laboratories, hospitals, hospital systems, clinics, or other institutions will have the option to be individually identified or remain anonymous in publications or summaries describing the pilot test and its findings.

Responsibilities for participating laboratories
Participation in the pilot test will be by invitation from the CDC. The single most important criterion for selecting laboratories to participate will be availability of relevant unpublished information, preferably in the form of a study on practices in one or both of the topic areas

Participating organizations will be asked to:

  • Designate a senior laboratory scientist, laboratory director, or other appropriate staff as the primary contact for requesting information regarding availability of potential studies
  • Identify and manage requirements necessary to provide requested information, potentially including an Institutional Review Board process
  • Submit requested specific, de-identified information to CDC

Costs of participating
The primary cost to organizations participating in the pilot test will be the time of the laboratory director or senior scientist who is designated as the primary study contact, and the time of other personnel involved in identifying, and submitting unpublished studies. No other significant costs are anticipated.

Additional information
For additional information about CDC’s initiative to define best practices in laboratory medicine, visit the DLS website. To volunteer your laboratory to participate in the pilot test, or if you have specific questions about the project, contact:

     Pamela Thompson pamela.thompson@cdc.hhs.gov 404-718-1021
     Anne Pollock anne.pollock@cdc.hhs.gov 404-718-1023

     

Battelle conducts the project described on this website for the Centers for Disease Control under contract W911NF-07-D-0001/DO 0191/TCN 07235