The American Society for Microbiology collaboration with the CDC Laboratory Medicine Best Practices initiative for evidence-based laboratory medicine

Abstract

SUMMARYClinical medicine has embraced the use of evidence for patient treatment decisions; however, the evaluation strategy for evidence in laboratory medicine practices has lagged. It was not until the end of the 20th century that the Institute of Medicine (IOM), now the National Academy of Medicine, and the Centers for Disease Control and Prevention, Division of Laboratory Systems (CDC DLS), focused on laboratory tests and how testing processes can be designed to benefit patient care. In collaboration with CDC DLS, the American Society for Microbiology (ASM) used an evidence review method developed by the CDC DLS to develop a program for creating laboratory testing guidelines and practices. The CDC DLS method is called the Laboratory Medicine Best Practices (LMBP) initiative and uses the A-6 cycle method. Adaptations made by ASM are called Evidence-based Laboratory Medicine Practice Guidelines (EBLMPG). This review details how the ASM Systematic Review (SR) Processes were developed and executed collaboratively with CDC's DLS. The review also describes the ASM transition from LMBP to the organization's current EBLMPG, maintaining a commitment to working with agencies in the U.S. Department of Health and Human Services and other partners to ensure that EBLMPG evidence is readily understood and consistently used.

Keywords: evidence-based medicine; meta-analysis; outcomes; systematic review.

Fig 1

The original LMBP A-6 cycle.

Fig 2

Analytic framework. Adapetd from LaRocco et al. (2). Review question: Are there pre-analytic practices related to the collection, storage, preservation, and transport of urine for microbiological culture that improve the diagnosis and management of patients with urinary tract infection?

Fig 3

Ranking in terms of scientific rigor, with meta-analysis representing the highest form of scientific rigor.

Fig 4

Example of forest plot format, adapted from LaRocco et al. (2). In the figure, each row in the table represents the summary of statistics from each study. Typically, these are organized into comparison groups (arms). The diamond indicates the location of the pooled effect, and the vertical line is the line of equivalence between groups; if the diamond touches the line of equivalence, then the groups are not statistically different. (i) Pooled statistics (the pooled odds ratio) are found at the bottom of the table or below a group of studies if the studies are subgrouped by characteristic. (ii) Heterogeneity (variation between studies not due to chance) statistics are also at the bottom of the figure. (iii) The forest plot (on the right side) is a visual representation of the values on the left portion of the figure. Each study is represented by a box where the size of the box represents the study’s weight, and the whiskers to each side of the box represent within-study error. (iv) The diamond represents the value of the pooled effect size. The width of the diamond indicates the error in the pooled effect measure. (v) Typically, a line of equivalence will be placed down the middle of the forest plot. This line represents the value at which the comparison arms are perfectly equivalent (1 in odds ratios, 0 in comparison of arm means). (vi) When the comparison groups differ on the outcome measure, the pooled effect diamond will fall to one side or the other of the line. (vii) If the pooled effect diamond does not touch the line of equivalence, then the effects in the two comparison groups (arms) are significantly different. The references in the figure are Blake & Doherty (84), Bradbury (85), Holliday (86), Schlager (87), and Schneeberger (88).

Fig 5

Diagnostic accuracy of midstream clean-catch (MSCC) versus suprapubic aspiration (SPA) for the diagnosis of urinary tract infections in children. Adapted from LaRocco et al. (2). The references in the figure are Hardy (91), Aronson (92), Pylkkanen (93), Ramage (94), and Morton (95).

Fig 6

HSROC curve for diagnosis of rheumatoid arthritis using anti‐cyclic citrullinated peptide antibody (anti‐CCP) compared with the reference standard (created using metandi command in STATA 14) (data from reference 89).

Fig 7

A-6 method adaptations by ASM.