Skip to main content
Log in

Methods for Determining Sample Sizes for Studies Involving Health-Related Quality of Life Measures: A Tutorial

  • Published:
Health Services and Outcomes Research Methodology Aims and scope Submit manuscript

Abstract

Health Related Quality of Life (HRQoL) measures are becoming more frequently used in clinical trials and health services research, both as primary and secondary endpoints. Investigators are now asking statisticians for advice on how to plan and analyse studies using HRQoL measures, which includes questions on sample size. Sample size requirements are critically dependent on the aims and objectives of the study, the proposed summary measure and effect size and the method of calculating the test statistic.

We present a tutorial on methods of sample size calculation for HRQoL outcomes. We also briefly review the HRQoL literature to see what has been done in practice. The aim of this tutorial is provide pragmatic guidance to researchers on the methods of calculating sample sizes when using HRQoL measures as outcomes.

HRQoL measures such as the SF-36, NHP and QLQ-C30 are usually measured on an ordered categorical (ordinal) scale. We argue that it is often incorrect to treat the scales as if they were continuous and normally distributed and that the mean score may not be a good summary measure of HRQoL data. However the ordinal scaling of HRQoL measures leads to problems in determining sample size, and we suggest that the odds ratio (OR) may be a more suitable summary measure for comparing groups (rather than the mean difference) and therefore methods suitable for ordinal data be used for analysis.

The frequency distribution of HRQoL scores should be assessed to see if parametric assumptions are satisfied and whether or not the sample mean is a good summary measure of the data. Given the non-normal distribution of the majority HRQoL outcome measures, summary measures such as means and standard deviations are difficult to interpret. Thus standardised differences (effect sizes) and parametric methods may not be a suitable basis for calculation of sample size. Finally we argue, that any sample size calculation (with all its attendant assumptions) leads to better research than no sample size calculation at all.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others

References

  • D. G. Altman, Practical Statistics for Medical Research, London: Chapman & Hall, 1991.

    Google Scholar 

  • D. G. Altman, D. Machin, T. N. Bryant and M. J. Gardner, Statistics with Confidence. Confidence intervals and statistical guidelines, 2nd Ed., London: British Medical Journal, 2000.

    Google Scholar 

  • P. Armitage and G. Berry, Statistical Methods in Medical Research, 3rd Ed., Oxford: Blackwell Science, 1994.

    Google Scholar 

  • J. M. Bland and D. G. Altman, “The use of transformation when comparing two means,” British Medical Journal, 312, p. 1153, 1996.

    Google Scholar 

  • K. Bolland, M. R. Sooriyarachchi and J. Whitehead, “Sample size review in a head injury trial with ordered categorical responses,” Statistics in Medicine, 17, pp. 2835–2847, 1998.

    Google Scholar 

  • R. Brant, “Assessing proportionality in the proportional odds model for ordinal logistic regression,” Biometrics, 46, pp. 1171–1178, 1990.

    Google Scholar 

  • J. E. Brazier, R. Harper, N. M. B. Jones, A. O'Cathain, K. J. Thomas, T. Usherwood and L. Westlake, “Validating the SF-36 health survey questionnaire: new outcome measure for primary care,” British Medical Journal, 305, pp. 160–164, 1992.

    Google Scholar 

  • M. J. Campbell, S. A. Julious and D. G. Altman, “Estimating sample sizes for binary, ordered categorical, and continuous outcomes in 2 group comparisons,” British Medical Journal, 311, pp. 1145–1148, 1995.

    Google Scholar 

  • M. J. Campbell, S. A. Julious and S. L. George, “Estimating sample sizes for studies using the SF-36 health survey-Reply,” Journal of Epidemiology & Community Health, 50, pp. 473–474, 1996.

    Google Scholar 

  • J. Cohen, Statistical Power Analysis for the Behavioral Sciences, New Jersey: Lawrence Earlbaum, 1988.

    Google Scholar 

  • P. M. Fayers and D. Machin, Quality of Life Assessment, Analysis and Interpretation, Chichester: Wiley, 2000.

    Google Scholar 

  • J. F. Hilton, “The appropriateness of the Wilcoxon test in ordinal data,” Statistics in Medicine, 15, pp. 631–645, 1996.

    Google Scholar 

  • J. F. Hilton and C. R. Mehta, “Power and sample size calculations for exact conditional tests with ordered categorical data,” Biometrics, 49, pp. 609–616, 1993.

    Google Scholar 

  • R. V. Hogg and E. A. Tanis, Probability and Statistical Inference, 3rd Ed. New York: McMillan, 1988.

    Google Scholar 

  • S. A. Julious and M. J. Campbell, “Sample sizes calculations for ordered categorical data,” Statistics in Medicine, 15, pp. 1065–1066, 1996.

    Google Scholar 

  • S. A. Julious and M. J. Campbell, “Sample size calculations for paired or matched ordinal data,” Statistics in Medicine, 17, pp. 1635–1642, 1998.

    Google Scholar 

  • S. A. Julious, S. George, D. Machin and R. J. Stephens, “Sample sizes for randomized trials measuring quality of life in cancer patients,” Quality of Life Research, 6, pp. 109–117, 1997.

    Google Scholar 

  • S. A. Julious, S. George and M. J. Campbell, “Sample sizes for studies using the short form 36 (SF-36),” Journal of Epidemiology & Community Health, 49, pp. 642–644, 1995.

    Google Scholar 

  • M. T. King, “The interpretation of scores from the EORTC quality of life questionnaire QLQ-C30,” Quality of Life Research, 5, pp. 555–567, 1996.

    Google Scholar 

  • J. E. Kolassa, “A comparison of size and power calculations for the Wilcoxon statistic for ordered categorical data,” Statistics in Medicine, 14, pp. 1577–1581, 1995.

    Google Scholar 

  • A. Laupacis, D. L. Sackett and R. S. Roberts, “An assessment of clinically useful measures of the consequences of treatment,” N. Engl. J. Med., 317, pp. 1728–1733, 1988.

    Google Scholar 

  • E. Lesaffre, I. Scheys, J. Frohlich and E. Bluhmki, “Calculation of power and sample size with bounded outcome scores,” Statistics in Medicine, 12, pp. 1063–1078, 1993.

    Google Scholar 

  • D. Machin, M. J. Campbell, P. M. Fayers and A. J. Y. Pinol, Sample Sizes Tables for Clinical Studies, Oxford: Blackwell Science, 1997.

    Google Scholar 

  • D. Hays and P. M. Fayers, Eds., Quality of Life Assessment in Clinical Trials: Methods and Practice, Oxford: Oxford University Press, pp. 37–50, 1998.

    Google Scholar 

  • C. J. Morrell, H. Spiby, P. Stewart, S. Walters and A. Morgan, “Costs and effectiveness of community postnatal support workers: randomised controlled trial,” British Medical Journal, 321, pp. 593–598, 2000.

    Google Scholar 

  • B. Peterson and F. E. Harrell, “Partial proportional odds models for ordinal response variables,” Applied Statistics, 39, pp. 205–217, 1990.

    Google Scholar 

  • L. Prieto, J. Alonso and J. M. Anto, “Estimating sample sizes for studies using the SF-36 health survey,” Journal of Epidemiology & Community Health, 50, p. 473, 1996.

    Google Scholar 

  • M. Roset, X. Badia and N. E. Mayo, “Sample size calculations in studies using the EuroQol 5D,” Quality of Life Research, 8, pp. 539–549, 1999.

    Google Scholar 

  • J. Whitehead, “Sample size calculations for ordered categorical data [published erratum appears in Stat Med 1994 Apr 30; 13(8): 871],” S tatistics in Medicine, 12, pp. 2257–2271, 1993.

    Google Scholar 

  • P. Williamson, J. L. Hutton, J. Bliss, J. Blunt, M. J. Campbell and R. Nicholson, “Statistical review by research ethics committees,” J. Roy. Statist. Soc. A, 163, pp. 5–13, 2000.

    Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Rights and permissions

Reprints and permissions

About this article

Cite this article

Walters, S.J., Campbell, M.J. & Paisley, S. Methods for Determining Sample Sizes for Studies Involving Health-Related Quality of Life Measures: A Tutorial. Health Services & Outcomes Research Methodology 2, 83–99 (2001). https://doi.org/10.1023/A:1020102612073

Download citation

  • Issue Date:

  • DOI: https://doi.org/10.1023/A:1020102612073

Keywords

Navigation