Re: Assessing the value of diagnostic tests: a framework for designing and evaluating trials
Lavinia Ferrante di Ruffano et al are absolutely right in calling for studies which determine a diagnostic test’s true clinical value.
As the authors point out, such studies can be difficult to execute and evaluate. However the few such studies available in the field of clinical microbiology are robust enough to guide and direct the microbiology service as a whole. Doern et al in a prospective controlled trial, (J. Clin. Microbiol. 1994;32:1757-1762), showed that a group of hospitalised patients assigned to receive rapidly reported bacterial identification and antimicrobial susceptibility tests had a lower mortality rate, underwent significantly fewer procedures and incurred significantly lower laboratory and pharmacy costs than a control group assigned to receive routine overnight reports. Numerous complementary studies show that early use of the appropriate antibiotic improves survival following bacteraemia. If these studies were taken into consideration, microbiologists would seek to implement this rapid reporting protocol as a priority. However, laboratory costs of rapid reporting are relatively high and are only affordable if non-laboratory savings are reinvested in the laboratory, and if inappropriate tests are rigorously excluded- all within a reasonably estimated and flexible capitation type budget. A microbiology service procured on a cost per test basis cannot pursue this clinical agenda. Consolidation, with the loss of on-site microbiology facilities further hampers rapid reporting.Demonstration of improved patient outcomes should be included in all laboratory service specifications.
Rapid Response:
Re: Assessing the value of diagnostic tests: a framework for designing and evaluating trials
Lavinia Ferrante di Ruffano et al are absolutely right in calling for studies which determine a diagnostic test’s true clinical value.
As the authors point out, such studies can be difficult to execute and evaluate. However the few such studies available in the field of clinical microbiology are robust enough to guide and direct the microbiology service as a whole. Doern et al in a prospective controlled trial, (J. Clin. Microbiol. 1994;32:1757-1762), showed that a group of hospitalised patients assigned to receive rapidly reported bacterial identification and antimicrobial susceptibility tests had a lower mortality rate, underwent significantly fewer procedures and incurred significantly lower laboratory and pharmacy costs than a control group assigned to receive routine overnight reports. Numerous complementary studies show that early use of the appropriate antibiotic improves survival following bacteraemia. If these studies were taken into consideration, microbiologists would seek to implement this rapid reporting protocol as a priority. However, laboratory costs of rapid reporting are relatively high and are only affordable if non-laboratory savings are reinvested in the laboratory, and if inappropriate tests are rigorously excluded- all within a reasonably estimated and flexible capitation type budget. A microbiology service procured on a cost per test basis cannot pursue this clinical agenda. Consolidation, with the loss of on-site microbiology facilities further hampers rapid reporting.Demonstration of improved patient outcomes should be included in all laboratory service specifications.
Competing interests: No competing interests