Estimation of data adaptive minimal clinically important difference with a nonconvex optimization procedure

Abstract

Understanding the limitation of solely relying on statistical significance, researchers have proposed methods to draw biomedical conclusions based on clinical significance. The minimal clinically important significance is one of the most fundamental concepts to study clinical significance. Based on an anchor question usually available in the patients' reported outcome, Hedayat et al. presented a method to estimate minimal clinically important significance using the classification technique. However, their method implicitly requires that the binary outcome of the anchor question is equally likely, i.e. the balanced outcome assumption. This assumption cannot be guaranteed a priori when one designs the study; hence, it cannot be satisfied in general. In this paper, we propose a data adaptive method, which can overcome this limitation. Compared to Hedayat et al., our method uses a faster gradient based algorithm and adopts a more flexible structure of the minimal clinically important significance at the individual level. We conduct comprehensive simulation studies and apply our method to the chondral lesions and meniscus procedure study to demonstrate its usefulness and also its outperformance.

Keywords

Minimal clinically important difference data adaptive surrogate loss nonconvex optimization

Get full access to this article

View all access options for this article.

References

Bhardwaj

Camacho

Derrow

, et al. Statistical significance and clinical relevance: the importance of power in clinical trials in dermatology. Arch Dermatol 2004; 140: 1520–1523.

Jaeschke

Singer

Guyatt

. Measurement of health status: ascertaining the minimal clinically important difference. Control Clin Trials 1989; 10: 407–415.

Wright

Linens

Cain

. Establishing the minimal clinical important difference and minimal detectable change for the cumberland ankle instability tool. Arch Phys Med Rehabil 2017; 98: 1806–1811.

Walenkamp

de Muinck Keizer

Goslings

, et al. The minimum clinically important difference of the patient-rated wrist evaluation score for patients with distal radius fractures. Clin Orthop Relat Res 2015; 473: 3235–3241.

Shiu

Gatsonis

. The predictive receiver operating characteristic curve for the joint assessment of the positive and negative predictive values. Philos Trans A Math Phys Eng Sci 2008; 366: 2313–2333.

Turner

Schünemann

Griffith

, et al. The minimal detectable change cannot reliably replace the minimal important difference. J Clin Epidemiol 2010; 63: 28–36.

Hedayat

Wang

. Minimum clinically important difference in medical studies. Biometrics 2015; 71: 33–41.

Vapnik VN. The nature of statistical learning theory. Springer, 1998.

Hastie T, Tibshirani R and Friedman J. The elements of statistical learning: data mining, inference, and prediction. Springer Series in Statistics. New York: Springer, 2009.

10.

Xu T. New developments of minimum clinically important difference: theory and methodology. PhD Thesis, 2014.

11.

Stratford

Binkley

Riddle

, et al. Sensitivity to change of the Roland-Morris back pain questionnaire: part 1. Phys Ther 1998; 78: 1186–1196.

12.

Crosby

Kolotkin

Williams

. Defining clinically meaningful change in health-related quality of life. J Clin Epidemiol 2003; 56(5): 395–407.

13.

Natarajan

. Sparse approximate solutions to linear systems. SIAM J Comput 1995; 24: 227–234.

14.

Thi Hoai An

Dinh Tao

. Solving a class of linearly constrained indefinite quadratic problems by dc algorithms. J Glob Optim 1997; 11: 253–285.

15.

Wahba G. Spline models for observational data, vol 59. Philadelphia, PA: SIAM, 1990.

16.

Nocedal

. Updating quasi-newton matrices with limited storage. Math Comput 1980; 35: 773–782.

17.

Wilkinson

. Results and implications of a study of fifteen years of satellite image classification experiments. IEEE Trans Geosci Remote Sens 2005; 43: 433–440.

18.

Hubbard

Benjamin-Johnson

Onega

, et al. Classification accuracy of claims-based methods for identifying providers failing to meet performance targets. Stat Med 2015; 34: 93–105.

19.

Bisson

Kluczynski

Wind

, et al. Patient outcomes after observation versus debridement of unstable chondral lesions during partial meniscectomy: the chondral lesions and meniscus procedures (champ) randomized controlled trial. J Bone Joint Surg 2017; 99: 1078–1085.

20.

Bisson

Kluczynski

Wind

, et al. How does the presence of unstable chondral lesions affect patient outcomes after partial meniscectomy? The champ randomized controlled trial. Am J Sports Med 2018; 46: 590–597.

21.

Bisson

Phillips

Matthews

, et al. Association between bone marrow lesions, chondral lesions, and pain in patients without radiographic evidence of degenerative joint disease who underwent arthroscopic partial meniscectomy. Orthop J Sports Med 2019. in press.

22.

Kluczynski

Marzo

Wind

, et al. The effect of body mass index on clinical outcomes in patients without radiographic evidence of degenerative joint disease after arthroscopic partial meniscectomy. Arthroscopy 2017; 33: 2054–2063.

23.

Zhou

Zhao

Kluczynski

. Interval estimation for minimal clinically important difference and its classification error via a bootstrap scheme. Stat Theory Relat Fields 2019. in press.