Exploring multivariate clinical chemical routine data concerning three major disease groups
1988

Exploring Clinical Chemistry Data for Disease Groups

Sample size: 142 publication Evidence: moderate

Author Information

Author(s): Jan B. Hemel, Hilko van tier Voet, Roll Hendriks, Frans R. Hindriks, Willem van tier Slik

Primary Institution: University Hospital Groningen

Hypothesis

Can multivariate analysis improve the separability of major disease groups based on clinical chemistry data?

Conclusion

The study found that multivariate analysis provides better separation of disease groups compared to univariate analysis, although some overlap remains.

Supporting Evidence

  • Multivariate analysis revealed better distinction between disease groups than univariate analysis.
  • Three-dimensional plots provided more insight than two-dimensional plots.
  • Non-linear mapping retained distances between data points effectively.

Takeaway

This study looked at blood tests from patients with liver, kidney, and heart diseases to see if we could tell them apart better using fancy math techniques.

Methodology

The study used multivariate analysis techniques on clinical chemistry data from patients with liver, kidney, and heart diseases.

Potential Biases

The selection of patients and assays may introduce bias, as not all relevant assays were performed on every patient.

Limitations

Some valuable information was lost due to the deletion of incomplete variables, and the analysis was limited to patients admitted within 10 days.

Participant Demographics

Patients included 46 with liver diseases, 50 with kidney diseases, and 46 with heart diseases.

Statistical Information

P-Value

p<0.0005

Statistical Significance

p<0.0005

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication