Application of measurement error models to correct for systematic differences among readers and vendors in echocardiography measurements: the CARDIA study

Aisha Betoko, Chike Nwabuo, Bharath Ambale Venkatesh, Erin P. Ricketts, Sejong Bae, Colin Wu, Samuel S. Gidding, Kiang Liu, João A.C. Lima, Christopher Cox

Research output: Contribution to journalArticle

Abstract

We illustrate the application of linear measurement error models to calibrate echocardiography measurements acquired 20 years apart in the CARDIA study. Of 4242 echocardiograms acquired at Year-5 (1990–1991), 36% were reread 20 years later. Left ventricular (LV) mass and 8 other measurements were assessed. A machine reproducibility study including 96 additional patients also compared Year-5 and Year-25 equipment. A linear measurement error model was developed to calibrate the original Year-5 measurements, incorporating the additional Year-5 reread and machine reproducibility study data, and adjusting for differences among readers and machines. Median (quartiles) of original Year-5 LV mass was 144.4 (117.6, 174.2) g before and 129.9 (103.8, 158.6) g, after calibration. The correlation between original and calibrated LV mass was 0.989 (95% confidence interval: 0.988, 0.990). The original and calibrated measurements had similar distributions. Additional comparisons of original and calibrated data supported the use of the model. We conclude that systematic differences among readers and machines have been accounted for, and that the calibrated Year-5 measurements can be used in future longitudinal comparisons. It is hoped that this paper will encourage the wider application of measurement error models.

Original languageEnglish (US)
JournalJournal of Applied Statistics
DOIs
StateAccepted/In press - Jan 1 2019

Fingerprint

Echocardiography
Measurement Error Model
Reproducibility
Quartile
Confidence interval
Calibration
Vendors
Measurement error

Keywords

  • Bias
  • calibration
  • echocardiography
  • linear measurement error models
  • systematic differences

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

Application of measurement error models to correct for systematic differences among readers and vendors in echocardiography measurements : the CARDIA study. / Betoko, Aisha; Nwabuo, Chike; Ambale Venkatesh, Bharath; Ricketts, Erin P.; Bae, Sejong; Wu, Colin; Gidding, Samuel S.; Liu, Kiang; Lima, João A.C.; Cox, Christopher.

In: Journal of Applied Statistics, 01.01.2019.

Research output: Contribution to journalArticle

Betoko, Aisha ; Nwabuo, Chike ; Ambale Venkatesh, Bharath ; Ricketts, Erin P. ; Bae, Sejong ; Wu, Colin ; Gidding, Samuel S. ; Liu, Kiang ; Lima, João A.C. ; Cox, Christopher. / Application of measurement error models to correct for systematic differences among readers and vendors in echocardiography measurements : the CARDIA study. In: Journal of Applied Statistics. 2019.
@article{0a592908dcd548e79efd469a646f2f6f,
title = "Application of measurement error models to correct for systematic differences among readers and vendors in echocardiography measurements: the CARDIA study",
abstract = "We illustrate the application of linear measurement error models to calibrate echocardiography measurements acquired 20 years apart in the CARDIA study. Of 4242 echocardiograms acquired at Year-5 (1990–1991), 36{\%} were reread 20 years later. Left ventricular (LV) mass and 8 other measurements were assessed. A machine reproducibility study including 96 additional patients also compared Year-5 and Year-25 equipment. A linear measurement error model was developed to calibrate the original Year-5 measurements, incorporating the additional Year-5 reread and machine reproducibility study data, and adjusting for differences among readers and machines. Median (quartiles) of original Year-5 LV mass was 144.4 (117.6, 174.2) g before and 129.9 (103.8, 158.6) g, after calibration. The correlation between original and calibrated LV mass was 0.989 (95{\%} confidence interval: 0.988, 0.990). The original and calibrated measurements had similar distributions. Additional comparisons of original and calibrated data supported the use of the model. We conclude that systematic differences among readers and machines have been accounted for, and that the calibrated Year-5 measurements can be used in future longitudinal comparisons. It is hoped that this paper will encourage the wider application of measurement error models.",
keywords = "Bias, calibration, echocardiography, linear measurement error models, systematic differences",
author = "Aisha Betoko and Chike Nwabuo and {Ambale Venkatesh}, Bharath and Ricketts, {Erin P.} and Sejong Bae and Colin Wu and Gidding, {Samuel S.} and Kiang Liu and Lima, {Jo{\~a}o A.C.} and Christopher Cox",
year = "2019",
month = "1",
day = "1",
doi = "10.1080/02664763.2019.1686133",
language = "English (US)",
journal = "Journal of Applied Statistics",
issn = "0266-4763",
publisher = "Routledge",

}

TY - JOUR

T1 - Application of measurement error models to correct for systematic differences among readers and vendors in echocardiography measurements

T2 - the CARDIA study

AU - Betoko, Aisha

AU - Nwabuo, Chike

AU - Ambale Venkatesh, Bharath

AU - Ricketts, Erin P.

AU - Bae, Sejong

AU - Wu, Colin

AU - Gidding, Samuel S.

AU - Liu, Kiang

AU - Lima, João A.C.

AU - Cox, Christopher

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We illustrate the application of linear measurement error models to calibrate echocardiography measurements acquired 20 years apart in the CARDIA study. Of 4242 echocardiograms acquired at Year-5 (1990–1991), 36% were reread 20 years later. Left ventricular (LV) mass and 8 other measurements were assessed. A machine reproducibility study including 96 additional patients also compared Year-5 and Year-25 equipment. A linear measurement error model was developed to calibrate the original Year-5 measurements, incorporating the additional Year-5 reread and machine reproducibility study data, and adjusting for differences among readers and machines. Median (quartiles) of original Year-5 LV mass was 144.4 (117.6, 174.2) g before and 129.9 (103.8, 158.6) g, after calibration. The correlation between original and calibrated LV mass was 0.989 (95% confidence interval: 0.988, 0.990). The original and calibrated measurements had similar distributions. Additional comparisons of original and calibrated data supported the use of the model. We conclude that systematic differences among readers and machines have been accounted for, and that the calibrated Year-5 measurements can be used in future longitudinal comparisons. It is hoped that this paper will encourage the wider application of measurement error models.

AB - We illustrate the application of linear measurement error models to calibrate echocardiography measurements acquired 20 years apart in the CARDIA study. Of 4242 echocardiograms acquired at Year-5 (1990–1991), 36% were reread 20 years later. Left ventricular (LV) mass and 8 other measurements were assessed. A machine reproducibility study including 96 additional patients also compared Year-5 and Year-25 equipment. A linear measurement error model was developed to calibrate the original Year-5 measurements, incorporating the additional Year-5 reread and machine reproducibility study data, and adjusting for differences among readers and machines. Median (quartiles) of original Year-5 LV mass was 144.4 (117.6, 174.2) g before and 129.9 (103.8, 158.6) g, after calibration. The correlation between original and calibrated LV mass was 0.989 (95% confidence interval: 0.988, 0.990). The original and calibrated measurements had similar distributions. Additional comparisons of original and calibrated data supported the use of the model. We conclude that systematic differences among readers and machines have been accounted for, and that the calibrated Year-5 measurements can be used in future longitudinal comparisons. It is hoped that this paper will encourage the wider application of measurement error models.

KW - Bias

KW - calibration

KW - echocardiography

KW - linear measurement error models

KW - systematic differences

UR - http://www.scopus.com/inward/record.url?scp=85075215412&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85075215412&partnerID=8YFLogxK

U2 - 10.1080/02664763.2019.1686133

DO - 10.1080/02664763.2019.1686133

M3 - Article

AN - SCOPUS:85075215412

JO - Journal of Applied Statistics

JF - Journal of Applied Statistics

SN - 0266-4763

ER -