The AAPS Journal

, 21:102 | Cite as

On the Comparison of Methods in Analyzing Bounded Outcome Score Data

  • Chuanpu HuEmail author


Clinical trial endpoints often take the form of bounded outcome scores (BOS) which report a discrete set of values on a finite range. Conceptually such endpoints are ordered categorical in nature, but in practice they are often analyzed as continuous variables, which may result in data range violations and difficulties to handle data skewness. Analysis methods dedicated for BOS data have been proposed; however, much confusion exists among pharmacometricians on how to compare the possible methods. This commentary reviews the main methods used in pharmacometrics applications and discusses their theoretical and practical comparisons. The expected performance of some conceptually appealing methods in different situations is discussed, and a guideline is provided on selecting analysis methods in practice.


categorical data likelihood model selection nonlinear mixed-effects modeling transformation 



  1. 1.
    Lesaffre E, Rizopoulos D, Tsonaka R. The logistic transform for bounded outcome scores. Biostatistics. 2007;8(1):72–85.CrossRefPubMedGoogle Scholar
  2. 2.
    Hu C, Randazzo B, Sharma A, Zhou H. Improvement in latent variable indirect response modeling of multiple categorical clinical endpoints: application to modeling of guselkumab treatment effects in psoriatic patients. J Pharmacokinet Pharmacodyn. 2017;44(5):437–48.CrossRefPubMedGoogle Scholar
  3. 3.
    Smithson M, Verkuilen J. A better lemon squeezer? Maximum-likelihood regression with beta-distributed dependent variables. Psychol Methods. 2006;11(1):54–71.CrossRefPubMedGoogle Scholar
  4. 4.
    Hu C, Yeilding N, Davis HM, Zhou H. Bounded outcome score modeling: application to treating psoriasis with ustekinumab. J Pharmacokinet Pharmacodyn. 2011;38(4):497–517.CrossRefPubMedGoogle Scholar
  5. 5.
    Hutmacher MM, French JL, Krishnaswami S, Menon S. Estimating transformations for repeated measures modeling of continuous bounded outcome data. Stat Med. 2011;30(9):935–49.CrossRefPubMedGoogle Scholar
  6. 6.
    Wellhagen GJ, Kjellsson MC, Karlsson MO. A bounded integer model for rating and composite scale data. AAPS J. 2019;21(4):74.CrossRefPubMedPubMedCentralGoogle Scholar
  7. 7.
    Karlsson MO, Holford NHG. A tutorial on visual predictive checks 2008 [updated 2008. Available from: Accessed 9 Aug 2019.
  8. 8.
    Piccolo D, Simone R, Iannario M. Cumulative and CUB models for rating data: a comparative analysis. Int Stat Rev. 2018;0(0):1–30.Google Scholar
  9. 9.
    Hu C, Adedokun OJ, Zhang L, Sharma A, Zhou H. Modeling near-continuous clinical endpoint as categorical: application to longitudinal exposure-response modeling of Mayo scores for golimumab in patients with ulcerative colitis. J Pharmacokinet Pharmacodyn. 2018;45(6):803–16.CrossRefPubMedGoogle Scholar
  10. 10.
    Hu C. Exposure-response modeling of clinical end points using latent variable indirect response models. CPT Pharmacometrics Syst Pharmacol. 2014;3:e117.CrossRefPubMedPubMedCentralGoogle Scholar
  11. 11.
    Liu Q, Shepherd BE, Li C, Harrell FE Jr. Modeling continuous response variables using ordinal regression. Stat Med. 2017;36(27):4316–35.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© American Association of Pharmaceutical Scientists 2019

Authors and Affiliations

  1. 1.Clinical Pharmacology and PharmacometricsJanssen Research & Development, LLCSpring HouseUSA

Personalised recommendations