Determining the optimal number and location of cutoff points with application to data of cervical cancer

Chung Chang, Meng-Ke Hsieh, Wen-Yi Chang, An Jen Chiang and Jiabin Chen

PLOS ONE, 2017, vol. 12, issue 4, 1-13

Abstract: It is often helpful to classify biomarker values into groups of different risk levels to facilitate evaluation of a biological, physiological, or pathological state. Stratification of patients into two risk groups is commonly seen, but there is always need for more than two groups for fine assessment. So far, there are no standard methods or tools to help decide how many cutoff points are optimal. In this study, we developed a comprehensive package that included methods to determine both the optimal number and locations of cutoff points for both survival data and dichotomized outcome. We illustrated workflow of this package with data from 797 patients with cervical cancer. By analyzing several risk factors of cervical cancer such as tumor size, body mass index (BMI), number of lymph nodes involved and depth of stromal invasion, in relation to survival and clinical outcome such as lymph nodal metastasis and lymphovascular invasion, we demonstrated that the best choice for BMI and stromal invasion was two cutoff points and one for the others. This study provided a useful tool to facilitate medical decisions and the analyses on cervical cancer may also be of interest to gynecologists. The package can be freely downloaded.

Date: 2017
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0176231 (text/html)
https://journals.plos.org/plosone/article/file?id= ... 76231&type=printable (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:plo:pone00:0176231

DOI: 10.1371/journal.pone.0176231

Access Statistics for this article

More articles in PLOS ONE from Public Library of Science
Bibliographic data for series maintained by plosone ().