Documenting Clinical Microbiology Impact: Performing Clinical Research to Document the Value of the Microbiology Laboratory

https://doi.org/10.1016/j.clinmicnews.2018.04.001Get rights and content

Abstract

For those intending to achieve the maximum validity and usefulness of microbiology results, the interface between clinical medicine and diagnostic methods in laboratory medicine is a critical focus. As such, method verification studies and quality improvement projects in clinical microbiology should optimally be performed with the rigor of scientific research, even if the project intent is not clinical research. This review aims to translate the concepts of clinical research into laboratory strategies that are relevant to clinical microbiologists who are attempting to document the value of their laboratory programs. This review focuses on the overall sequence of events that help to ensure a successful project that generates clinical impact that is data driven by laboratory results. We describe the steps needed to fully document laboratory impact after what we call a “diagnostic intervention.”

Section snippets

The new complexity of diagnostic testing

All medical laboratory scientists know that before being used for patient care diagnostic tests should be thoroughly evaluated and their accuracy and reproducibility should be documented. An equally well-known but rarely developed concept is that health care decision makers, physicians, and other providers require more than a simple transaction from the laboratory; we are clearly much more than “sample in, answer out.” In the next decade, health care organizations will expect that diagnostic

The typical sequence of an investigation or a study

The sequence of POR typically involves 10 steps; (i) observation of current circumstances or challenges that can define a research question, (ii) literature review and information gathering, (iii) generation of a testable idea for improvement, (iv) creation of study aims and a study hypothesis, (v) preparation of a study design that will test the hypothesis, (vi) performance of the clinical study with data collection, (vii) data validation, (viii) data analysis, (ix) data interpretation, and

Create and comply with procedures for data handling and storage

Designate your source data (all information in original records of clinical findings, observations, or other activities necessary for the reconstruction and evaluation of the study). Then, create databases and file directories in which to store the data. Carefully select units of measurement, data collection tools, analytical tools, and biostatistics software. Develop data collection forms and develop or select a statistical analysis plan to document technical details and elaboration of the

Data Validation

If there are data entry errors or extraction errors or incomplete extraction the data will not be correct or useful; therefore, the process of data validation is performed. Data validation is the process of ensuring that data have undergone assessment, often called data cleaning or cleansing, to ensure the quality of the data set. Sometimes validation rules are created to check for correctness and meaningfulness of data. The rules may be manually applied to the data set, for example checking

Data Analysis – Testing the Hypothesis

The data analysis plan is a key element in the design of a clinical research study. It helps to determine the number of subjects or samples needed to sufficiently power the study. It may be necessary to consult with a biostatistician to properly assess power, study design, endpoints, and method of data analysis, as well as to assess and analyze the data and perform biostatistical analysis. The type of data analysis used depends on the number and the type of dependent (outcome variables) and

Data Interpretation

Interpret the data and display the results using graphical representation and establish whether the hypothesis was rejected, thus proving the alternate hypothesis. Deal with unexpected results, consider the limitations of the study, and determine the generalizability of the study results (i.e., the extent to which the findings of a study can be reliably extrapolated from the subjects who participated in the study to a broader patient population and a broader range of clinical settings).

Formulation of Conclusions and Future Action Plans

After data analysis and graphical representation of the study results are complete, formulate a conclusion and establish whether the hypothesis was proven. Forming accurate conclusions that do not go beyond the data is a critical aspect of research. Researchers must ask if the data are definitive, i.e., strong enough to stand alone for recommending a change in practice, or whether further studies are required. Consider the need for future studies and to develop an action plan based on the study

Other Considerations

Continuing medical education and habits of lifelong learning are key to the future of clinical laboratory science. For laboratory scientists who are working in quality control, quality assurance, or evidence-based medicine; who are considering a research career or graduate school; or who are taking course work with a focus on biostatistics, regulatory issues, grant writing, experimental design, and clinical trial development, the knowledge gained from clinical laboratory data can be extremely

Summary

In the future of health care laboratory medicine, clinical research needs will expand and clinical laboratories will be asked provide evidence of their population health and organizational impact to improve patient care. The topics in this review are only a small portion of the skills and strategies required for research track positions, study coordinators, quality assurance and improvement teams, laboratory managers, supervisors, and laboratory directors of the future.

References (23)

  • PM Bossuyt et al.

    Comparative accuracy: assessing new tests against existing diagnostic pathways

    BMJ

    (2006)

    • Cited by (0)

    View full text