Interobserver variation in breast cancer grading: A statistical modeling approach

Nilotpal Chowdhury, Muktha R. Pai, Flora D. Lobo, Hema Kini, Rebecca Varghese

Research output: Contribution to journalArticle

13 Citations (Scopus)

Abstract

OBJECTIVE: To study random and the systematic error in breast cancer grading, to find the source of disagreements and measure the reliability of graders so that appropriate corrective action can be taken. STUDY DESIGN: Five independent observers graded 50 breast carcinoma slides from 50 consecutive breast cancer specimens according to the Nottingham criteria. The polychoric correlation was used to measure association. Stuart-Maxwell and McNemar tests were used to measure equality of thresholds. RESULTS: The polychoric correlation among observers was high (mean = 0.803,0.712, 0.797 and 0.602 for the final grade, tubule formation, nuclear pleomorphism and mitotic figures, respectively). However, there were significant differences in thresholds (6, 7, 7 and 9 pairs of 10 showing significant differences in classification of grades/scores for final grade, tubule formation, nuclear pleomorphism and mitotic counts, respectively). CONCLUSION: The high polychoric correlations suggest that random error in grading breast cancers in this study was low, confirming the underlying reliability of grading and graders. However, significant differences in the thresholds lowers raw agreement. Such a scenario may be rectified by increased intradepartmental discussion.

Original languageEnglish
Pages (from-to)213-218
Number of pages6
JournalAnalytical and Quantitative Cytology and Histology
Volume28
Issue number4
Publication statusPublished - 01-08-2006

Fingerprint

Observer Variation
Breast Neoplasms

All Science Journal Classification (ASJC) codes

  • Anatomy
  • Histology

Cite this

@article{b16442e2b8904a648b08dd09f4e269f1,
title = "Interobserver variation in breast cancer grading: A statistical modeling approach",
abstract = "OBJECTIVE: To study random and the systematic error in breast cancer grading, to find the source of disagreements and measure the reliability of graders so that appropriate corrective action can be taken. STUDY DESIGN: Five independent observers graded 50 breast carcinoma slides from 50 consecutive breast cancer specimens according to the Nottingham criteria. The polychoric correlation was used to measure association. Stuart-Maxwell and McNemar tests were used to measure equality of thresholds. RESULTS: The polychoric correlation among observers was high (mean = 0.803,0.712, 0.797 and 0.602 for the final grade, tubule formation, nuclear pleomorphism and mitotic figures, respectively). However, there were significant differences in thresholds (6, 7, 7 and 9 pairs of 10 showing significant differences in classification of grades/scores for final grade, tubule formation, nuclear pleomorphism and mitotic counts, respectively). CONCLUSION: The high polychoric correlations suggest that random error in grading breast cancers in this study was low, confirming the underlying reliability of grading and graders. However, significant differences in the thresholds lowers raw agreement. Such a scenario may be rectified by increased intradepartmental discussion.",
author = "Nilotpal Chowdhury and Pai, {Muktha R.} and Lobo, {Flora D.} and Hema Kini and Rebecca Varghese",
year = "2006",
month = "8",
day = "1",
language = "English",
volume = "28",
pages = "213--218",
journal = "Analytical and Quantitative Cytology",
issn = "0884-6812",
publisher = "Science Printers and Publishers Inc.",
number = "4",

}

Interobserver variation in breast cancer grading : A statistical modeling approach. / Chowdhury, Nilotpal; Pai, Muktha R.; Lobo, Flora D.; Kini, Hema; Varghese, Rebecca.

In: Analytical and Quantitative Cytology and Histology, Vol. 28, No. 4, 01.08.2006, p. 213-218.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Interobserver variation in breast cancer grading

T2 - A statistical modeling approach

AU - Chowdhury, Nilotpal

AU - Pai, Muktha R.

AU - Lobo, Flora D.

AU - Kini, Hema

AU - Varghese, Rebecca

PY - 2006/8/1

Y1 - 2006/8/1

N2 - OBJECTIVE: To study random and the systematic error in breast cancer grading, to find the source of disagreements and measure the reliability of graders so that appropriate corrective action can be taken. STUDY DESIGN: Five independent observers graded 50 breast carcinoma slides from 50 consecutive breast cancer specimens according to the Nottingham criteria. The polychoric correlation was used to measure association. Stuart-Maxwell and McNemar tests were used to measure equality of thresholds. RESULTS: The polychoric correlation among observers was high (mean = 0.803,0.712, 0.797 and 0.602 for the final grade, tubule formation, nuclear pleomorphism and mitotic figures, respectively). However, there were significant differences in thresholds (6, 7, 7 and 9 pairs of 10 showing significant differences in classification of grades/scores for final grade, tubule formation, nuclear pleomorphism and mitotic counts, respectively). CONCLUSION: The high polychoric correlations suggest that random error in grading breast cancers in this study was low, confirming the underlying reliability of grading and graders. However, significant differences in the thresholds lowers raw agreement. Such a scenario may be rectified by increased intradepartmental discussion.

AB - OBJECTIVE: To study random and the systematic error in breast cancer grading, to find the source of disagreements and measure the reliability of graders so that appropriate corrective action can be taken. STUDY DESIGN: Five independent observers graded 50 breast carcinoma slides from 50 consecutive breast cancer specimens according to the Nottingham criteria. The polychoric correlation was used to measure association. Stuart-Maxwell and McNemar tests were used to measure equality of thresholds. RESULTS: The polychoric correlation among observers was high (mean = 0.803,0.712, 0.797 and 0.602 for the final grade, tubule formation, nuclear pleomorphism and mitotic figures, respectively). However, there were significant differences in thresholds (6, 7, 7 and 9 pairs of 10 showing significant differences in classification of grades/scores for final grade, tubule formation, nuclear pleomorphism and mitotic counts, respectively). CONCLUSION: The high polychoric correlations suggest that random error in grading breast cancers in this study was low, confirming the underlying reliability of grading and graders. However, significant differences in the thresholds lowers raw agreement. Such a scenario may be rectified by increased intradepartmental discussion.

UR - http://www.scopus.com/inward/record.url?scp=33746684302&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746684302&partnerID=8YFLogxK

M3 - Article

C2 - 16927641

AN - SCOPUS:33746684302

VL - 28

SP - 213

EP - 218

JO - Analytical and Quantitative Cytology

JF - Analytical and Quantitative Cytology

SN - 0884-6812

IS - 4

ER -