Bayesian analysis of risk factors for anovulation

Yan Liu, Wesley O. Johnson, Ellen B Gold, Bill L. Lasley

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

Two algorithms for assessing ovulatory status using daily urinary levels of oestrogen and progesterone metabolites have been applied to non-clinic-based, free-living populations of women. These relatively new methods for assessing ovarian function have been used to assess the potential adverse effects of occupational and environmental exposures, such as smoking, on the reproductive health of women. One algorithm has been validated against serum hormone measurements and gives good sensitivity and specificity for anovulation. However, a gold standard is generally not available in epidemiologic field studies in which these daily urine samples are collected. In this paper, we used Bayesian methods to estimate: (i) the probability of occurrence of anovulation, (ii) the sensitivity and specificity of the two algorithms, and (iii) the association between anovulation and smoking and other risk factors in the absence of a perfect test. We evaluated the two published algorithms for assessing ovulatory status, based on their cross-classified results applied to one randomly selected cycle from each woman in a sample of 338 employed women. We first assumed that the algorithms were independent, conditional on ovulatory status. Then, we used a dependence model to allow for correlation between the results of the two algorithms. We implemented a Bayesian logistic regression analysis that allowed the outcome measurement to be partially imperfect. We incorporated the posterior distributions for algorithm accuracy obtained from the dependence model as prior distributions for this logistic regression model. Then, we compared the results with those obtained from a standard multiple logistic approach using the algorithm determination of ovulatory status as if it were perfect. Our results indicated that increasing physical activity was associated with a significantly increased risk of anovulation; and smokers had a potentially, but not statistically significant, increased occurrence of anovulation.

Original languageEnglish (US)
Pages (from-to)1901-1919
Number of pages19
JournalStatistics in Medicine
Volume23
Issue number12
DOIs
StatePublished - Jun 30 2004

Fingerprint

Anovulation
Bayes Theorem
Bayesian Analysis
Risk Factors
Smoking
Logistic Models
Specificity
Progesterone
Sensitivity and Specificity
Estrogen
Field Study
Logistic Regression Model
Reproductive Health
Environmental Exposure
Bayesian Methods
Occupational Exposure
Hormones
Logistic Regression
Prior distribution
Posterior distribution

Keywords

  • Anovulation
  • Bayesian approach
  • Gibbs sampling
  • Logistic regression
  • Sensitivity
  • Specificity

ASJC Scopus subject areas

  • Epidemiology

Cite this

Bayesian analysis of risk factors for anovulation. / Liu, Yan; Johnson, Wesley O.; Gold, Ellen B; Lasley, Bill L.

In: Statistics in Medicine, Vol. 23, No. 12, 30.06.2004, p. 1901-1919.

Research output: Contribution to journalArticle

Liu, Y, Johnson, WO, Gold, EB & Lasley, BL 2004, 'Bayesian analysis of risk factors for anovulation', Statistics in Medicine, vol. 23, no. 12, pp. 1901-1919. https://doi.org/10.1002/sim.1773
Liu, Yan ; Johnson, Wesley O. ; Gold, Ellen B ; Lasley, Bill L. / Bayesian analysis of risk factors for anovulation. In: Statistics in Medicine. 2004 ; Vol. 23, No. 12. pp. 1901-1919.
@article{a096153f9f92437fb7ec6efed73549c5,
title = "Bayesian analysis of risk factors for anovulation",
abstract = "Two algorithms for assessing ovulatory status using daily urinary levels of oestrogen and progesterone metabolites have been applied to non-clinic-based, free-living populations of women. These relatively new methods for assessing ovarian function have been used to assess the potential adverse effects of occupational and environmental exposures, such as smoking, on the reproductive health of women. One algorithm has been validated against serum hormone measurements and gives good sensitivity and specificity for anovulation. However, a gold standard is generally not available in epidemiologic field studies in which these daily urine samples are collected. In this paper, we used Bayesian methods to estimate: (i) the probability of occurrence of anovulation, (ii) the sensitivity and specificity of the two algorithms, and (iii) the association between anovulation and smoking and other risk factors in the absence of a perfect test. We evaluated the two published algorithms for assessing ovulatory status, based on their cross-classified results applied to one randomly selected cycle from each woman in a sample of 338 employed women. We first assumed that the algorithms were independent, conditional on ovulatory status. Then, we used a dependence model to allow for correlation between the results of the two algorithms. We implemented a Bayesian logistic regression analysis that allowed the outcome measurement to be partially imperfect. We incorporated the posterior distributions for algorithm accuracy obtained from the dependence model as prior distributions for this logistic regression model. Then, we compared the results with those obtained from a standard multiple logistic approach using the algorithm determination of ovulatory status as if it were perfect. Our results indicated that increasing physical activity was associated with a significantly increased risk of anovulation; and smokers had a potentially, but not statistically significant, increased occurrence of anovulation.",
keywords = "Anovulation, Bayesian approach, Gibbs sampling, Logistic regression, Sensitivity, Specificity",
author = "Yan Liu and Johnson, {Wesley O.} and Gold, {Ellen B} and Lasley, {Bill L.}",
year = "2004",
month = "6",
day = "30",
doi = "10.1002/sim.1773",
language = "English (US)",
volume = "23",
pages = "1901--1919",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "12",

}

TY - JOUR

T1 - Bayesian analysis of risk factors for anovulation

AU - Liu, Yan

AU - Johnson, Wesley O.

AU - Gold, Ellen B

AU - Lasley, Bill L.

PY - 2004/6/30

Y1 - 2004/6/30

N2 - Two algorithms for assessing ovulatory status using daily urinary levels of oestrogen and progesterone metabolites have been applied to non-clinic-based, free-living populations of women. These relatively new methods for assessing ovarian function have been used to assess the potential adverse effects of occupational and environmental exposures, such as smoking, on the reproductive health of women. One algorithm has been validated against serum hormone measurements and gives good sensitivity and specificity for anovulation. However, a gold standard is generally not available in epidemiologic field studies in which these daily urine samples are collected. In this paper, we used Bayesian methods to estimate: (i) the probability of occurrence of anovulation, (ii) the sensitivity and specificity of the two algorithms, and (iii) the association between anovulation and smoking and other risk factors in the absence of a perfect test. We evaluated the two published algorithms for assessing ovulatory status, based on their cross-classified results applied to one randomly selected cycle from each woman in a sample of 338 employed women. We first assumed that the algorithms were independent, conditional on ovulatory status. Then, we used a dependence model to allow for correlation between the results of the two algorithms. We implemented a Bayesian logistic regression analysis that allowed the outcome measurement to be partially imperfect. We incorporated the posterior distributions for algorithm accuracy obtained from the dependence model as prior distributions for this logistic regression model. Then, we compared the results with those obtained from a standard multiple logistic approach using the algorithm determination of ovulatory status as if it were perfect. Our results indicated that increasing physical activity was associated with a significantly increased risk of anovulation; and smokers had a potentially, but not statistically significant, increased occurrence of anovulation.

AB - Two algorithms for assessing ovulatory status using daily urinary levels of oestrogen and progesterone metabolites have been applied to non-clinic-based, free-living populations of women. These relatively new methods for assessing ovarian function have been used to assess the potential adverse effects of occupational and environmental exposures, such as smoking, on the reproductive health of women. One algorithm has been validated against serum hormone measurements and gives good sensitivity and specificity for anovulation. However, a gold standard is generally not available in epidemiologic field studies in which these daily urine samples are collected. In this paper, we used Bayesian methods to estimate: (i) the probability of occurrence of anovulation, (ii) the sensitivity and specificity of the two algorithms, and (iii) the association between anovulation and smoking and other risk factors in the absence of a perfect test. We evaluated the two published algorithms for assessing ovulatory status, based on their cross-classified results applied to one randomly selected cycle from each woman in a sample of 338 employed women. We first assumed that the algorithms were independent, conditional on ovulatory status. Then, we used a dependence model to allow for correlation between the results of the two algorithms. We implemented a Bayesian logistic regression analysis that allowed the outcome measurement to be partially imperfect. We incorporated the posterior distributions for algorithm accuracy obtained from the dependence model as prior distributions for this logistic regression model. Then, we compared the results with those obtained from a standard multiple logistic approach using the algorithm determination of ovulatory status as if it were perfect. Our results indicated that increasing physical activity was associated with a significantly increased risk of anovulation; and smokers had a potentially, but not statistically significant, increased occurrence of anovulation.

KW - Anovulation

KW - Bayesian approach

KW - Gibbs sampling

KW - Logistic regression

KW - Sensitivity

KW - Specificity

UR - http://www.scopus.com/inward/record.url?scp=2942735334&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942735334&partnerID=8YFLogxK

U2 - 10.1002/sim.1773

DO - 10.1002/sim.1773

M3 - Article

C2 - 15195323

AN - SCOPUS:2942735334

VL - 23

SP - 1901

EP - 1919

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 12

ER -