Sample size calculations based on a difference in medians for positively skewed outcomes in health care studies

BackgroundIn healthcare research, outcomes with skewed probability distributions are common. Sample size calculations for such outcomes are typically based on estimates on a transformed scale (e.g. log) which may sometimes be difficult to obtain. In contrast, estimates of median and variance on the untransformed scale are generally easier to pre-specify. The aim of this paper is to describe how to calculate a sample size for a two group comparison of interest based on median and untransformed variance estimates for log-normal outcome data.MethodsA log-normal distribution for outcome data is assumed and a sample size calculation approach for a two-sample t-test that compares log-transformed outcome data is demonstrated where the change of interest is specified as difference in median values on the untransformed scale. A simulation study is used to compare the method with a non-parametric alternative (Mann-Whitney U test) in a variety of scenarios and the method is applied to a real example in neurosurgery.ResultsThe method attained a nominal power value in simulation studies and was favourable in comparison to a Mann-Whitney U test and a two-sample t-test of untransformed outcomes. In addition, the method can be adjusted and used in some situations where the outcome distribution is not strictly log-normal.ConclusionsWe recommend the use of this sample size calculation approach for outcome data that are expected to be positively skewed and where a two group comparison on a log-transformed scale is planned. An advantage of this method over usual calculations based on estimates on the log-transformed scale is that it allows clinical efficacy to be specified as a difference in medians and requires a variance estimate on the untransformed scale. Such estimates are often easier to obtain and more interpretable than those for log-transformed outcomes.

[1]  Shein-Chung Chow,et al.  Sample Size Calculations in Clinical Research, Second Edition , 2003 .

[2]  D. Glueck.  Sample Size Calculations in Clinical Research 2nd edition by CHOW, S.‐C., SHAO, J., and WANG, H. , 2008 .

[3]  T C Chalmers,et al.  The importance of beta, the type II error and sample size in the design and interpretation of the randomized control trial. Survey of 71 "negative" trials. , 1978, The New England journal of medicine.

[4]  S. Senn,et al.  Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design. , 1994, Statistics in medicine.

[5]  Bal Sanghera,et al.  Assessment of tumor heterogeneity: an emerging imaging tool for clinical practice? , 2012, Insights into Imaging.

[6]  John S Duncan,et al.  The long-term outcome of adult epilepsy surgery, patterns of seizure remission, and relapse: a cohort study , 2011, The Lancet.

[7]  David Machin,et al.  Sample Size Tables for Clinical Studies , 1997 .

[8]  D. Spiegelhalter,et al.  Quality of life measures in health care. I: Applications and issues in assessment. , 1992, BMJ.

[9]  Andrew J Vickers,et al.  Parametric versus non-parametric statistics in the analysis of randomized trials with non-normally distributed data , 2005, BMC medical research methodology.

[10]  Thomas Czech,et al.  A novel miniature robotic device for frameless implantation of depth electrodes in refractory epilepsy. , 2017, Journal of neurosurgery.

[11]  J B Carlin,et al.  Sample-size calculation for a log-transformed outcome measure. , 1999, Controlled clinical trials.

[12]  K. Schulz,et al.  Sample size calculations in randomised trials: mandatory and mystical , 2005, The Lancet.

[13]  S. Chow,et al.  Sample Size Calculations In Clinical Research , 2007 .

[14]  Douglas A. Wolfe,et al.  Introduction to the Theory of Nonparametric Statistics. , 1980 .

[15]  C. Heij,et al.  Modeling Procedure and Surgical Times for Current Procedural Terminology-Anesthesia-Surgeon Combinations and Evaluation in Terms of Case-Duration Prediction and Operating Room Efficiency: A Multicenter Study , 2009, Anesthesia and analgesia.

[16]  R. Randles,et al.  Introduction to the Theory of Nonparametric Statistics , 1991 .

[17]  P. Lee,et al.  Interpretation and Uses of Medical Statistics. , 1969 .

[18]  N. Mikuni,et al.  Invasive Evaluations for Epilepsy Surgery: A Review of the Literature , 2016, Neurologia medico-chirurgica.

[19]  B. Cundill,et al.  Sample size calculations for skewed distributions , 2015, BMC Medical Research Methodology.

[20]  P. Hougaard,et al.  Fundamentals of Survival Data , 1999, Biometrics.

[21]  F. O’Sullivan,et al.  Statistical assessment of treatment response in a cancer patient based on pre‐therapy and post‐therapy FDG‐PET scans , 2017, Statistics in medicine.

[22]  C. Erhardt,et al.  INTERPRETATION AND USES OF MEDICAL STATISTICS , 1970 .

[23]  Sebastien Ourselin,et al.  A Novel Method for Implementation of Frameless StereoEEG in Epilepsy Surgery , 2014, Neurosurgery.