Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy Populations

Peter E Hickman; Gus Koerbin; Julia M Potter; Nicholas Glasgow; Juleen A Cavanaugh; Walter P Abhayaratna; Nic P West; Paul Glasziou

doi:10.1093/clinchem/hvaa208

Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy Populations

Peter E Hickman, Gus Koerbin, Julia M Potter, Nicholas Glasgow, Juleen A Cavanaugh, Walter P Abhayaratna, Nic P West, Paul Glasziou

Institute for Evidence-Based Healthcare

Research output: Contribution to journal › Article › Research › peer-review

15 Citations (Scopus)

91 Downloads (Pure)

Abstract

BACKGROUND: Reference intervals are an important aid in medical practice as they provide clinicians a guide as to whether a patient is healthy or diseased.Outlier results in population studies are removed by any of a variety of statistical measures. We have compared several methods of outlier removal and applied them to a large body of analytes from a large population of healthy persons.

METHODS: We used the outlier exclusion criteria of Reed-Dixon and Tukey and calculated reference intervals using nonparametric and Harrell-Davis statistical methods and applied them to a total of 36 different analytes.

RESULTS: Nine of 36 analytes had a greater than 20% difference in the upper reference limit, and for some the difference was 100% or more.

CONCLUSIONS: For some analytes, great importance is attached to the reference interval. We have shown that different statistical methods for outlier removal can cause large changes to reported reference intervals. So that population studies can be readily compared, common statistical methods should be used for outlier removal.

Original language	English
Pages (from-to)	1558-1561
Number of pages	4
Journal	Clinical Chemistry
Volume	66
Issue number	12
Early online date	24 Oct 2020
DOIs	https://doi.org/10.1093/clinchem/hvaa208
Publication status	Published - Dec 2020

Access to Document

10.1093/clinchem/hvaa208Licence: CC BY-NC-ND

Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy PopulationsFinal published version, 132 KBLicence: CC BY-NC-ND

Cite this

@article{797001789abb4781849b3f2cbf5f951f,

title = "Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy Populations",

abstract = "BACKGROUND: Reference intervals are an important aid in medical practice as they provide clinicians a guide as to whether a patient is healthy or diseased.Outlier results in population studies are removed by any of a variety of statistical measures. We have compared several methods of outlier removal and applied them to a large body of analytes from a large population of healthy persons.METHODS: We used the outlier exclusion criteria of Reed-Dixon and Tukey and calculated reference intervals using nonparametric and Harrell-Davis statistical methods and applied them to a total of 36 different analytes.RESULTS: Nine of 36 analytes had a greater than 20% difference in the upper reference limit, and for some the difference was 100% or more.CONCLUSIONS: For some analytes, great importance is attached to the reference interval. We have shown that different statistical methods for outlier removal can cause large changes to reported reference intervals. So that population studies can be readily compared, common statistical methods should be used for outlier removal.",

author = "Hickman, {Peter E} and Gus Koerbin and Potter, {Julia M} and Nicholas Glasgow and Cavanaugh, {Juleen A} and Abhayaratna, {Walter P} and West, {Nic P} and Paul Glasziou",

note = "{\textcopyright} American Association for Clinical Chemistry 2020.",

year = "2020",

month = dec,

doi = "10.1093/clinchem/hvaa208",

language = "English",

volume = "66",

pages = "1558--1561",

journal = "Clinical Chemistry",

issn = "0009-9147",

publisher = "AMER ASSOC CLINICAL CHEMISTRY",

number = "12",

}

TY - JOUR

T1 - Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy Populations

AU - Hickman, Peter E

AU - Koerbin, Gus

AU - Potter, Julia M

AU - Glasgow, Nicholas

AU - Cavanaugh, Juleen A

AU - Abhayaratna, Walter P

AU - West, Nic P

AU - Glasziou, Paul

PY - 2020/12

Y1 - 2020/12

N2 - BACKGROUND: Reference intervals are an important aid in medical practice as they provide clinicians a guide as to whether a patient is healthy or diseased.Outlier results in population studies are removed by any of a variety of statistical measures. We have compared several methods of outlier removal and applied them to a large body of analytes from a large population of healthy persons.METHODS: We used the outlier exclusion criteria of Reed-Dixon and Tukey and calculated reference intervals using nonparametric and Harrell-Davis statistical methods and applied them to a total of 36 different analytes.RESULTS: Nine of 36 analytes had a greater than 20% difference in the upper reference limit, and for some the difference was 100% or more.CONCLUSIONS: For some analytes, great importance is attached to the reference interval. We have shown that different statistical methods for outlier removal can cause large changes to reported reference intervals. So that population studies can be readily compared, common statistical methods should be used for outlier removal.

AB - BACKGROUND: Reference intervals are an important aid in medical practice as they provide clinicians a guide as to whether a patient is healthy or diseased.Outlier results in population studies are removed by any of a variety of statistical measures. We have compared several methods of outlier removal and applied them to a large body of analytes from a large population of healthy persons.METHODS: We used the outlier exclusion criteria of Reed-Dixon and Tukey and calculated reference intervals using nonparametric and Harrell-Davis statistical methods and applied them to a total of 36 different analytes.RESULTS: Nine of 36 analytes had a greater than 20% difference in the upper reference limit, and for some the difference was 100% or more.CONCLUSIONS: For some analytes, great importance is attached to the reference interval. We have shown that different statistical methods for outlier removal can cause large changes to reported reference intervals. So that population studies can be readily compared, common statistical methods should be used for outlier removal.

UR - http://www.scopus.com/inward/record.url?scp=85101654114&partnerID=8YFLogxK

U2 - 10.1093/clinchem/hvaa208

DO - 10.1093/clinchem/hvaa208

M3 - Article

C2 - 34214151

SN - 0009-9147

VL - 66

SP - 1558

EP - 1561

JO - Clinical Chemistry

JF - Clinical Chemistry

IS - 12

ER -

Choice of Statistical Tools for Outlier Removal Causes Substantial Changes in Analyte Reference Intervals in Healthy Populations

Abstract

Access to Document

Other files and links

Fingerprint

Cite this