Skip to content
Licensed Unlicensed Requires Authentication Published by De Gruyter June 1, 2022

Performance of four regression frameworks with varying precision profiles in simulated reference material commutability assessment

Corey Markus ORCID logo, Rui Zhen Tan, Chun Yee Lim, Wayne Rankin, Susan J. Matthews, Tze Ping Loh and William M. Hague

Abstract

Objectives

One approach to assessing reference material (RM) commutability and agreement with clinical samples (CS) is to use ordinary least squares or Deming regression with prediction intervals. This approach assumes constant variance that may not be fulfilled by the measurement procedures. Flexible regression frameworks which relax this assumption, such as quantile regression or generalized additive models for location, scale, and shape (GAMLSS), have recently been implemented, which can model the changing variance with measurand concentration.

Methods

We simulated four imprecision profiles, ranging from simple constant variance to complex mixtures of constant and proportional variance, and examined the effects on commutability assessment outcomes with above four regression frameworks and varying the number of CS, data transformations and RM location relative to CS concentration. Regression framework performance was determined by the proportion of false rejections of commutability from prediction intervals or centiles across relative RM concentrations and was compared with the expected nominal probability coverage.

Results

In simple variance profiles (constant or proportional variance), Deming regression, without or with logarithmic transformation respectively, is the most efficient approach. In mixed variance profiles, GAMLSS with smoothing techniques are more appropriate, with consideration given to increasing the number of CS and the relative location of RM. In the case where analytical coefficients of variation profiles are U-shaped, even the more flexible regression frameworks may not be entirely suitable.

Conclusions

In commutability assessments, variance profiles of measurement procedures and location of RM in respect to clinical sample concentration significantly influence the false rejection rate of commutability.


Corresponding author: Corey Markus, Flinders University International Centre for Point-of-Care Testing, Flinders Health and Medical Research Institute, Sturt Road, Bedford Park, South Australia 5042, Australia; and GPO Box 2100, Adelaide, SA, 5001, Australia, E-mail:

  1. Research funding: None declared.

  2. Author contributions: CM, TL, WR and WH conceived the requirement for simulation. CM performed simulation and produced numerical and graphical summary results. RT and CL had oversight and approval of statistical analysis. CM drafted the original manuscript, with all authors contributing to the content and editorial review of manuscript. All authors have read and approved the final manuscript.

  3. Competing interests: Authors state no conflict of interest.

  4. Ethical approval: Simulated data is exempt from review by the local Institutional Ethics Board.

References

1. Vesper, HW, Miller, WG, Myers, GL. Reference materials and commutability. Clin Biochem Rev 2007;28:139–47.Search in Google Scholar

2. Braga, F, Panteghini, M. Commutability of reference and control materials: an essential factor for assuring the quality of measurements in laboratory medicine. Clin Chem Lab Med 2019;57:967-73. https://doi.org/10.1515/cclm-2019-0154.Search in Google Scholar PubMed

3. Clinical Laboratory Standards Institute. EP14-A3: Evaluation of commutability of processed samples; approved guideline — third edition. Pennsylvania: Clinical Laboratory Standards Institute; 2014.Search in Google Scholar

4. Clinical Laboratory Standards Institute. EP30-A: Characterization and qualification of commutable reference materials for laboratory medicine; approved guideline — first edition. Pennsylvania: Clinical Laboratory Standards Institute; 2010.Search in Google Scholar

5. Sadler, WA. Variance function program version 17.0 (for Windows XP and later). Christchurch: Nuclear Medicine Department, Christchurch Hospital; 2020. Available from: https://www.aacb.asn.au/documents/item/164.Search in Google Scholar

6. Sadler, WA. Imprecision profiling. Clin Biochem Rev 2008;29:S33–36.Search in Google Scholar

7. Rigby, RA, Stasinopoulos, DM. Generalized additive models for location, scale and shape. J Roy Stat Soc C Appl Stat 2005;54:507–54, https://doi.org/10.1111/j.1467-9876.2005.00510.x.Search in Google Scholar

8. de Onis, M. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ 2007;85:660–7. https://doi.org/10.2471/BLT.07.043497.Search in Google Scholar PubMed PubMed Central

9. Gunapalasingham, G, Frithioff-Bøjsøe, C, Lund, MAV, Hedley, PL, Fonvig, CE, Dahl, M, et al.. Reference values for fasting serum concentrations of thyroid-stimulating hormone and thyroid hormones in healthy Danish/North-European white children and adolescents. Scand J Clin Lab Invest 2019;79:129–35. https://doi.org/10.1080/00365513.2019.1581945.Search in Google Scholar PubMed

10. Markus, C, Flores, C, Saxon, B, Osborn, K. Pregnancy-specific continuous reference intervals for haematology parameters from an Australian dataset: a step toward dynamic continuous reference intervals. Aust N Z J Obstet Gynaecol 2021;61:223–31. https://doi.org/10.1111/ajo.13260.Search in Google Scholar PubMed

11. Altman, N, Krzywinski, M. Simple linear regression. Nat Methods 2015;12:999–1000. https://doi.org/10.1038/nmeth.3627.Search in Google Scholar PubMed

12. Koenker, R, Bassett, G. Regression quantiles. Econometrica 1978;46:33. https://doi.org/10.2307/1913643.Search in Google Scholar

13. Konstantopoulos, S, Li, W, Miller, S, van der Ploeg, A. Using quantile regression to estimate intervention effects beyond the mean. Educ Psychol Meas 2019;79:883–910. https://doi.org/10.1177/0013164419837321.Search in Google Scholar PubMed PubMed Central

14. Stasinopoulos, MD, Rigby, RA, Heller, GZ, Voudouris, V, Bastiani, FD. Flexible regression and smoothing, 1st ed. USA: CRC Press; 2017.10.1201/b21973Search in Google Scholar

15. R Core Team. R: a language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2021. Available from: https://www.r-project.org/.Search in Google Scholar

16. Koenker, R. Quantreg: quantile regression; R package version 5.86; 2021. Available from: https://CRAN.R-project.org/package=quantreg.Search in Google Scholar

17. Peterson, RA, Cavanaugh, JE. Ordered quantile normalization: a semiparametric transformation built for the cross-validation era. J Appl Stat 2020;47:2312–27. https://doi.org/10.1080/02664763.2019.1630372.Search in Google Scholar

18. Peterson, RA. Finding optimal normalizing transformations via bestNormalize. R J 2021;13:310. https://doi.org/10.32614/RJ-2021-041.Search in Google Scholar

19. Yi, X, Wang, Y, Zhang, T, Zeng, J, Zhao, H, Zhou, W, et al.. Commutability of possible external quality assessment materials for progesterone measurement. Clin Biochem 2021;87:39–45. https://doi.org/10.1016/j.clinbiochem.2020.10.012.Search in Google Scholar PubMed

20. Bondell, HD, Reich, BJ, Wang, H. Noncrossing quantile regression curve estimation. Biometrika 2010;97:825–38. https://doi.org/10.1093/biomet/asq048.Search in Google Scholar PubMed PubMed Central

21. Borghi, E, de Onis, M, Garza, C, Van den Broeck, J, Frongillo, EA, Grummer-Strawn, L, et al.. For the WHO multicentre growth reference study group, construction of the World Health Organization child growth standards: selection of methods for attained growth curves. Stat Med 2006;25:247–65. https://doi.org/10.1002/sim.2227.Search in Google Scholar PubMed

22. Nilsson, G, Budd, JR, Greenberg, N, Delatour, V, Rej, R, Panteghini, M, et al.. IFCC working group recommendations for assessing commutability part 2: using the difference in bias between a reference material and clinical samples. Clin Chem 2018;64:455–64. https://doi.org/10.1373/clinchem.2017.277541.Search in Google Scholar PubMed PubMed Central

23. Røraas, TH, Van Houcke, SK, Stöckl, D, Thienpont, LM. Statistical power for commutability testing in the presence of random, sample-related effects by use of the EP14 protocol. Clin Chim Acta 2012;413:1710–1. https://doi.org/10.1016/j.cca.2012.05.002.Search in Google Scholar PubMed

24. Stöckl, D, Stepman, HCM, Van Houcke, SK, Thienpont, LM. Importance of sample-related effects for commutability testing according to the EP14 protocol. Clin Chim Acta 2010;411:1378–9. https://doi.org/10.1016/j.cca.2010.04.024.Search in Google Scholar PubMed

25. Miller, WG, Schimmel, H, Rej, R, Greenberg, N, Ceriotti, F, Burns, C, et al.. IFCC working group recommendations for assessing commutability part 1: general experimental design. Clin Chem 2018;64:447–54. https://doi.org/10.1373/clinchem.2017.277525.Search in Google Scholar PubMed PubMed Central

26. Budd, JR, Weykamp, C, Rej, R, MacKenzie, F, Ceriotti, F, Greenberg, N, et al.. IFCC working group recommendations for assessing commutability part 3: using the calibration effectiveness of a reference material. Clin Chem 2018;64:465–74. https://doi.org/10.1373/clinchem.2017.277558.Search in Google Scholar PubMed


Supplementary Material

The online version of this article offers supplementary material (https://doi.org/10.1515/cclm-2022-0205).


Received: 2022-03-05
Accepted: 2022-05-12
Published Online: 2022-06-01
Published in Print: 2022-07-26

© 2022 Walter de Gruyter GmbH, Berlin/Boston

Scroll Up Arrow