A Nonparametric Method for Combining Multilaboratory Data


In multilaboratory clinical studies, laboratory differences cannot be ignored when analyzing data from different laboratories together. In this paper, we propose a new method to transform data across laboratories. The method uses established empirical distributions of assay results from different laboratories for the transformation. This transformation does not rely on either the distributional assumption of the assay reading data or the reference ranges that are required by commonly-used standardization methods. A simulation study shows that the mean and standard deviation of the transformed data are close to the true values. A clinical example is given to compare the usual standardization method to our method. In general, our method is shown to be useful in combining data from multiple laboratories, especially when the data are nonnormal and when the reference ranges are not reliable. Finally, the pros and cons of the method are discussed.

This is a preview of subscription content, access via your institution.


  1. 1.

    International Conference on Harmonization. E9. Statistical Principles for Clinical Trials. February 5, 1998. www.ifpma.org/pdfifpma/e9.pdf.

    Google Scholar 

  2. 2.

    Chuang-Stein C. Summarizing laboratory data with different reference ranges in multi-center clinical trials. Drug Inf J. 1992;26(1):77–84.

    Article  Google Scholar 

  3. 3.

    Gilbert G, Mosher K, Zubkof L. A procedure for the simplification and assessment of lab parameters in clinical trials. Drug Inf J. 1986;20:279–296.

    Article  Google Scholar 

  4. 4.

    Gilbert G, Ting N. Statistical comparison of drug safety in controlled clinical trials: the Gene score as an objective measure of lab abnormalities. Drug Inf J. 1991;25:81–96.

    Article  Google Scholar 

  5. 5.

    Royston P, Mattews JNS. Estimation of reference ranges from normal samples. Stat Med. 1991;10:691–695.

    CAS  Article  Google Scholar 

  6. 6.

    Reed AH, Henry RJ, Mason WB. Influence of statistical method used on the resulting estimate of normal range. Clinical Chemistry. 1971;17(4):275–284.

    CAS  Article  Google Scholar 

  7. 7.

    Brunelle RL, Huang J, Fineberg SE, Anderson JH. Insulin antibody responses with results from insulin lispro in long term studies in patients with type 1 and type 2 diabetes. Diabetologia. 1999;42 (suppl 1): A98.

    Google Scholar 

  8. 8.

    Brunelle R, Huang J, Zhao Q, Anderson J. Combining insulin antibody data from different laboratories using an empirical cumulative distribution function (CDF) transformation. Diabetes. 1999;48(suppl 1):A430.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Jie Huang PhD.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Huang, J., Brunelle, R. A Nonparametric Method for Combining Multilaboratory Data. Ther Innov Regul Sci 36, 395–406 (2002). https://doi.org/10.1177/009286150203600219

Download citation

Key Words

  • Cumulative density function
  • Standardization
  • Transformation
  • Reference ranges
  • Empirical distribution