Iterative outlier removal: A method for identifying outliers in laboratory recalibration studies

Christina M. Parrinello, Morgan E. Grams, Yingying Sang, David Couper, Lisa M. Wruck, Danni Li, John H. Eckfeldt, Elizabeth Selvin, Josef Coresh

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

BACKGROUND: Extreme values that arise for any reason, including those through nonlaboratory measurement procedure-related processes (inadequate mixing, evaporation, mislabeling), lead to outliers and inflate errors in recalibration studies. We present an approach termed iterative outlier removal (IOR) for identifying such outliers. METHODS: We previously identified substantial laboratory drift in uric acid measurements in the Atherosclerosis Risk in Communities (ARIC) Study over time. Serum uric acid was originally measured in 1990-1992 on a CoulterDACOSinstrument using an uricase-based measurement procedure. To recalibrate previous measured concentrations to a newer enzymatic colorimetric measurement procedure, uric acid was remeasured in 200 participants from stored plasma in 2011-2013 on a Beckman Olympus 480 autoanalyzer. To conduct IOR, we excluded data points >3 SDs from the mean difference. We continued this process using the resulting data until no outliers remained. RESULTS: IOR detected more outliers and yielded greater precision in simulation. The original mean difference (SD) in uric acid was 1.25 (0.62) mg/dL. After 4 iterations, 9 outliers were excluded, and the mean difference (SD) was 1.23 (0.45) mg/dL. Conducting only one round of outlier removal (standard approach) would have excluded 4 outliers [mean difference (SD) = 1.22 (0.51) mg/dL]. Applying the recalibration (derived from Deming regression) from each approach to the original measurements, the prevalence of hyperuricemia (>7 mg/dL) was 28.5% before IOR and 8.5% after IOR. CONCLUSIONS: IOR is a useful method for removal of extreme outliers irrelevant to recalibrating laboratory measurements, and identifies more extraneous outliers than the standard approach.

Original languageEnglish (US)
Pages (from-to)966-972
Number of pages7
JournalClinical Chemistry
Volume62
Issue number7
DOIs
StatePublished - Jul 1 2016

Fingerprint

Uric Acid
Urate Oxidase
Hyperuricemia
Atherosclerosis
Serum
Evaporation
Plasmas

ASJC Scopus subject areas

  • Clinical Biochemistry
  • Biochemistry, medical

Cite this

Parrinello, C. M., Grams, M. E., Sang, Y., Couper, D., Wruck, L. M., Li, D., ... Coresh, J. (2016). Iterative outlier removal: A method for identifying outliers in laboratory recalibration studies. Clinical Chemistry, 62(7), 966-972. https://doi.org/10.1373/clinchem.2016.255216

Iterative outlier removal : A method for identifying outliers in laboratory recalibration studies. / Parrinello, Christina M.; Grams, Morgan E.; Sang, Yingying; Couper, David; Wruck, Lisa M.; Li, Danni; Eckfeldt, John H.; Selvin, Elizabeth; Coresh, Josef.

In: Clinical Chemistry, Vol. 62, No. 7, 01.07.2016, p. 966-972.

Research output: Contribution to journalArticle

Parrinello, CM, Grams, ME, Sang, Y, Couper, D, Wruck, LM, Li, D, Eckfeldt, JH, Selvin, E & Coresh, J 2016, 'Iterative outlier removal: A method for identifying outliers in laboratory recalibration studies', Clinical Chemistry, vol. 62, no. 7, pp. 966-972. https://doi.org/10.1373/clinchem.2016.255216
Parrinello, Christina M. ; Grams, Morgan E. ; Sang, Yingying ; Couper, David ; Wruck, Lisa M. ; Li, Danni ; Eckfeldt, John H. ; Selvin, Elizabeth ; Coresh, Josef. / Iterative outlier removal : A method for identifying outliers in laboratory recalibration studies. In: Clinical Chemistry. 2016 ; Vol. 62, No. 7. pp. 966-972.
@article{1761a065043d46f48d15eda15ae10927,
title = "Iterative outlier removal: A method for identifying outliers in laboratory recalibration studies",
abstract = "BACKGROUND: Extreme values that arise for any reason, including those through nonlaboratory measurement procedure-related processes (inadequate mixing, evaporation, mislabeling), lead to outliers and inflate errors in recalibration studies. We present an approach termed iterative outlier removal (IOR) for identifying such outliers. METHODS: We previously identified substantial laboratory drift in uric acid measurements in the Atherosclerosis Risk in Communities (ARIC) Study over time. Serum uric acid was originally measured in 1990-1992 on a CoulterDACOSinstrument using an uricase-based measurement procedure. To recalibrate previous measured concentrations to a newer enzymatic colorimetric measurement procedure, uric acid was remeasured in 200 participants from stored plasma in 2011-2013 on a Beckman Olympus 480 autoanalyzer. To conduct IOR, we excluded data points >3 SDs from the mean difference. We continued this process using the resulting data until no outliers remained. RESULTS: IOR detected more outliers and yielded greater precision in simulation. The original mean difference (SD) in uric acid was 1.25 (0.62) mg/dL. After 4 iterations, 9 outliers were excluded, and the mean difference (SD) was 1.23 (0.45) mg/dL. Conducting only one round of outlier removal (standard approach) would have excluded 4 outliers [mean difference (SD) = 1.22 (0.51) mg/dL]. Applying the recalibration (derived from Deming regression) from each approach to the original measurements, the prevalence of hyperuricemia (>7 mg/dL) was 28.5{\%} before IOR and 8.5{\%} after IOR. CONCLUSIONS: IOR is a useful method for removal of extreme outliers irrelevant to recalibrating laboratory measurements, and identifies more extraneous outliers than the standard approach.",
author = "Parrinello, {Christina M.} and Grams, {Morgan E.} and Yingying Sang and David Couper and Wruck, {Lisa M.} and Danni Li and Eckfeldt, {John H.} and Elizabeth Selvin and Josef Coresh",
year = "2016",
month = "7",
day = "1",
doi = "10.1373/clinchem.2016.255216",
language = "English (US)",
volume = "62",
pages = "966--972",
journal = "Clinical Chemistry",
issn = "0009-9147",
publisher = "American Association for Clinical Chemistry Inc.",
number = "7",

}

TY - JOUR

T1 - Iterative outlier removal

T2 - A method for identifying outliers in laboratory recalibration studies

AU - Parrinello, Christina M.

AU - Grams, Morgan E.

AU - Sang, Yingying

AU - Couper, David

AU - Wruck, Lisa M.

AU - Li, Danni

AU - Eckfeldt, John H.

AU - Selvin, Elizabeth

AU - Coresh, Josef

PY - 2016/7/1

Y1 - 2016/7/1

N2 - BACKGROUND: Extreme values that arise for any reason, including those through nonlaboratory measurement procedure-related processes (inadequate mixing, evaporation, mislabeling), lead to outliers and inflate errors in recalibration studies. We present an approach termed iterative outlier removal (IOR) for identifying such outliers. METHODS: We previously identified substantial laboratory drift in uric acid measurements in the Atherosclerosis Risk in Communities (ARIC) Study over time. Serum uric acid was originally measured in 1990-1992 on a CoulterDACOSinstrument using an uricase-based measurement procedure. To recalibrate previous measured concentrations to a newer enzymatic colorimetric measurement procedure, uric acid was remeasured in 200 participants from stored plasma in 2011-2013 on a Beckman Olympus 480 autoanalyzer. To conduct IOR, we excluded data points >3 SDs from the mean difference. We continued this process using the resulting data until no outliers remained. RESULTS: IOR detected more outliers and yielded greater precision in simulation. The original mean difference (SD) in uric acid was 1.25 (0.62) mg/dL. After 4 iterations, 9 outliers were excluded, and the mean difference (SD) was 1.23 (0.45) mg/dL. Conducting only one round of outlier removal (standard approach) would have excluded 4 outliers [mean difference (SD) = 1.22 (0.51) mg/dL]. Applying the recalibration (derived from Deming regression) from each approach to the original measurements, the prevalence of hyperuricemia (>7 mg/dL) was 28.5% before IOR and 8.5% after IOR. CONCLUSIONS: IOR is a useful method for removal of extreme outliers irrelevant to recalibrating laboratory measurements, and identifies more extraneous outliers than the standard approach.

AB - BACKGROUND: Extreme values that arise for any reason, including those through nonlaboratory measurement procedure-related processes (inadequate mixing, evaporation, mislabeling), lead to outliers and inflate errors in recalibration studies. We present an approach termed iterative outlier removal (IOR) for identifying such outliers. METHODS: We previously identified substantial laboratory drift in uric acid measurements in the Atherosclerosis Risk in Communities (ARIC) Study over time. Serum uric acid was originally measured in 1990-1992 on a CoulterDACOSinstrument using an uricase-based measurement procedure. To recalibrate previous measured concentrations to a newer enzymatic colorimetric measurement procedure, uric acid was remeasured in 200 participants from stored plasma in 2011-2013 on a Beckman Olympus 480 autoanalyzer. To conduct IOR, we excluded data points >3 SDs from the mean difference. We continued this process using the resulting data until no outliers remained. RESULTS: IOR detected more outliers and yielded greater precision in simulation. The original mean difference (SD) in uric acid was 1.25 (0.62) mg/dL. After 4 iterations, 9 outliers were excluded, and the mean difference (SD) was 1.23 (0.45) mg/dL. Conducting only one round of outlier removal (standard approach) would have excluded 4 outliers [mean difference (SD) = 1.22 (0.51) mg/dL]. Applying the recalibration (derived from Deming regression) from each approach to the original measurements, the prevalence of hyperuricemia (>7 mg/dL) was 28.5% before IOR and 8.5% after IOR. CONCLUSIONS: IOR is a useful method for removal of extreme outliers irrelevant to recalibrating laboratory measurements, and identifies more extraneous outliers than the standard approach.

UR - http://www.scopus.com/inward/record.url?scp=84976463909&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84976463909&partnerID=8YFLogxK

U2 - 10.1373/clinchem.2016.255216

DO - 10.1373/clinchem.2016.255216

M3 - Article

AN - SCOPUS:84976463909

VL - 62

SP - 966

EP - 972

JO - Clinical Chemistry

JF - Clinical Chemistry

SN - 0009-9147

IS - 7

ER -