Adjustment of cancer incidence rates for ethnic misclassification

Susan L Stewart, Karen C. Swallen, Sally L. Glaser, Pamela L. Horn-Ross, Dee W. West

Research output: Contribution to journalArticle

34 Citations (Scopus)

Abstract

Although ethnic population counts measured by the United States Census are based on self-identification, the same is not necessarily true of cases reported to cancer registries. The use of different ethnic classification methods for numerators and denominators may therefore lead to biased estimates of cancer incidence rates. The extent of such misclassification may be assessed by conducting an ethnicity survey of cancer patients and estimating the proportion misclassified using double sampling models that account for sample stratification. For two ethnic categories, logistic regression may be used to model self-identified ethnicity as a function of demographic variables and the fallible classification method. Incidence rates then may be adjusted for misclassification using regression results to estimate the number of cancer cases of a given age, sex, and site in each self-identified ethnic group. An example is given using this method to estimate ethnic misclassification of San Francisco Bay area Hispanic cancer patients diagnosed in 1990. Results suggest that the number of cancer cases reported as Hispanic is an underestimate of the number of cases self- identified as Hispanic, resulting in an underestimate of Hispanic cancer rates.

Original languageEnglish (US)
Pages (from-to)774-781
Number of pages8
JournalBiometrics
Volume54
Issue number2
DOIs
StatePublished - Jun 1998
Externally publishedYes

Fingerprint

Misclassification
Incidence
Cancer
Adjustment
incidence
neoplasms
Hispanic Americans
nationalities and ethnic groups
Logistics
Neoplasms
Sampling
Estimate
Double Sampling
Numerator
San Francisco
Census
Denominator
Censuses
Logistic Regression
Stratification

Keywords

  • Double sampling
  • Hispanic
  • Logistic regression
  • Self-identification

ASJC Scopus subject areas

  • Agricultural and Biological Sciences(all)
  • Public Health, Environmental and Occupational Health
  • Agricultural and Biological Sciences (miscellaneous)
  • Applied Mathematics
  • Statistics and Probability

Cite this

Stewart, S. L., Swallen, K. C., Glaser, S. L., Horn-Ross, P. L., & West, D. W. (1998). Adjustment of cancer incidence rates for ethnic misclassification. Biometrics, 54(2), 774-781. https://doi.org/10.2307/3109783

Adjustment of cancer incidence rates for ethnic misclassification. / Stewart, Susan L; Swallen, Karen C.; Glaser, Sally L.; Horn-Ross, Pamela L.; West, Dee W.

In: Biometrics, Vol. 54, No. 2, 06.1998, p. 774-781.

Research output: Contribution to journalArticle

Stewart, SL, Swallen, KC, Glaser, SL, Horn-Ross, PL & West, DW 1998, 'Adjustment of cancer incidence rates for ethnic misclassification', Biometrics, vol. 54, no. 2, pp. 774-781. https://doi.org/10.2307/3109783
Stewart, Susan L ; Swallen, Karen C. ; Glaser, Sally L. ; Horn-Ross, Pamela L. ; West, Dee W. / Adjustment of cancer incidence rates for ethnic misclassification. In: Biometrics. 1998 ; Vol. 54, No. 2. pp. 774-781.
@article{1da93a4f59cc4bd88162beb02ba21a00,
title = "Adjustment of cancer incidence rates for ethnic misclassification",
abstract = "Although ethnic population counts measured by the United States Census are based on self-identification, the same is not necessarily true of cases reported to cancer registries. The use of different ethnic classification methods for numerators and denominators may therefore lead to biased estimates of cancer incidence rates. The extent of such misclassification may be assessed by conducting an ethnicity survey of cancer patients and estimating the proportion misclassified using double sampling models that account for sample stratification. For two ethnic categories, logistic regression may be used to model self-identified ethnicity as a function of demographic variables and the fallible classification method. Incidence rates then may be adjusted for misclassification using regression results to estimate the number of cancer cases of a given age, sex, and site in each self-identified ethnic group. An example is given using this method to estimate ethnic misclassification of San Francisco Bay area Hispanic cancer patients diagnosed in 1990. Results suggest that the number of cancer cases reported as Hispanic is an underestimate of the number of cases self- identified as Hispanic, resulting in an underestimate of Hispanic cancer rates.",
keywords = "Double sampling, Hispanic, Logistic regression, Self-identification",
author = "Stewart, {Susan L} and Swallen, {Karen C.} and Glaser, {Sally L.} and Horn-Ross, {Pamela L.} and West, {Dee W.}",
year = "1998",
month = "6",
doi = "10.2307/3109783",
language = "English (US)",
volume = "54",
pages = "774--781",
journal = "Biometrics",
issn = "0006-341X",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - Adjustment of cancer incidence rates for ethnic misclassification

AU - Stewart, Susan L

AU - Swallen, Karen C.

AU - Glaser, Sally L.

AU - Horn-Ross, Pamela L.

AU - West, Dee W.

PY - 1998/6

Y1 - 1998/6

N2 - Although ethnic population counts measured by the United States Census are based on self-identification, the same is not necessarily true of cases reported to cancer registries. The use of different ethnic classification methods for numerators and denominators may therefore lead to biased estimates of cancer incidence rates. The extent of such misclassification may be assessed by conducting an ethnicity survey of cancer patients and estimating the proportion misclassified using double sampling models that account for sample stratification. For two ethnic categories, logistic regression may be used to model self-identified ethnicity as a function of demographic variables and the fallible classification method. Incidence rates then may be adjusted for misclassification using regression results to estimate the number of cancer cases of a given age, sex, and site in each self-identified ethnic group. An example is given using this method to estimate ethnic misclassification of San Francisco Bay area Hispanic cancer patients diagnosed in 1990. Results suggest that the number of cancer cases reported as Hispanic is an underestimate of the number of cases self- identified as Hispanic, resulting in an underestimate of Hispanic cancer rates.

AB - Although ethnic population counts measured by the United States Census are based on self-identification, the same is not necessarily true of cases reported to cancer registries. The use of different ethnic classification methods for numerators and denominators may therefore lead to biased estimates of cancer incidence rates. The extent of such misclassification may be assessed by conducting an ethnicity survey of cancer patients and estimating the proportion misclassified using double sampling models that account for sample stratification. For two ethnic categories, logistic regression may be used to model self-identified ethnicity as a function of demographic variables and the fallible classification method. Incidence rates then may be adjusted for misclassification using regression results to estimate the number of cancer cases of a given age, sex, and site in each self-identified ethnic group. An example is given using this method to estimate ethnic misclassification of San Francisco Bay area Hispanic cancer patients diagnosed in 1990. Results suggest that the number of cancer cases reported as Hispanic is an underestimate of the number of cases self- identified as Hispanic, resulting in an underestimate of Hispanic cancer rates.

KW - Double sampling

KW - Hispanic

KW - Logistic regression

KW - Self-identification

UR - http://www.scopus.com/inward/record.url?scp=0031798980&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031798980&partnerID=8YFLogxK

U2 - 10.2307/3109783

DO - 10.2307/3109783

M3 - Article

VL - 54

SP - 774

EP - 781

JO - Biometrics

JF - Biometrics

SN - 0006-341X

IS - 2

ER -