## Annals of Applied Statistics

### A reference-invariant health disparity index based on Rényi divergence

Makram Talih

#### Abstract

One of four overarching goals of Healthy People 2020 (HP2020) is to achieve health equity, eliminate disparities, and improve the health of all groups. In health disparity indices (HDIs) such as the mean log deviation (MLD) and Theil index (TI), disparities are relative to the population average, whereas in the index of disparity (IDisp) the reference is the group with the least adverse health outcome. Although the latter may be preferable, identification of a reference group can be affected by statistical reliability. To address this issue, we propose a new HDI, the Rényi index (RI), which is reference-invariant. When standardized, the RI extends the Atkinson index, where a disparity aversion parameter can incorporate societal values associated with health equity. In addition, both the MLD and TI are limiting cases of the RI. Also, a symmetrized Rényi index (SRI) can be constructed, resulting in a symmetric measure in the two distributions whose relative entropy is being evaluated. We discuss alternative symmetric and reference-invariant HDIs derived from the generalized entropy (GE) class and the Bregman divergence, and argue that the SRI is more robust than its GE-based counterpart to small changes in the distribution of the adverse health outcome. We evaluate the design-based standard errors and bootstrapped sampling distributions for the SRI, and illustrate the proposed methodology using data from the National Health and Nutrition Examination Survey (NHANES) on the 2001–04 prevalence of moderate or severe periodontitis among adults aged 45–74, which track Oral Health objective OH-5 in HP2020. Such data, which use a binary individual-level outcome variable, are typical of HP2020 data.

#### Article information

Source
Ann. Appl. Stat., Volume 7, Number 2 (2013), 1217-1243.

Dates
First available in Project Euclid: 27 June 2013

https://projecteuclid.org/euclid.aoas/1372338485

Digital Object Identifier
doi:10.1214/12-AOAS621

Mathematical Reviews number (MathSciNet)
MR3113507

Zentralblatt MATH identifier
06279871

#### Citation

Talih, Makram. A reference-invariant health disparity index based on Rényi divergence. Ann. Appl. Stat. 7 (2013), no. 2, 1217--1243. doi:10.1214/12-AOAS621. https://projecteuclid.org/euclid.aoas/1372338485

#### References

• Ali, S. M. and Silvey, S. D. (1966). A general class of coefficients of divergence of one distribution from another. J. Roy. Statist. Soc. Ser. B 28 131–142.
• Atkinson, A. B. (1970). On the measurement of inequality. J. Econom. Theory 2 244–263.
• Biewen, M. and Jenkins, S. P. (2006). Variance estimation for generalized entropy and Atkinson inequality indices: The complex survey data case. Oxford Bulletin of Economics and Statistics 68 371–383.
• Borrell, L. N. and Talih, M. (2012). Examining periodontal disease disparities among U.S. adults 20 years of age and older: NHANES III (1988–1994) and NHANES 1999–2004. Public Health Rep. 127 497–506.
• Borrell, L. N. and Talih, M. (2011). A symmetrized Theil index measure of health disparities: An example using dental caries in U.S. children and adolescents. Stat. Med. 30 277–290.
• Bourguignon, F. (1979). Decomposable income inequality measures. Econometrica 47 901–920.
• Braveman, P. (2006). Health disparities and health equity: Concepts and measurement. Annu. Rev. Public Health 27 167–194.
• Cheng, N. F., Han, P. Z. and Gansky, S. A. (2008). Methods and software for estimating health disparities: The case of children’s oral health. Am. J. Epidemiol. 168 906–914.
• Chernoff, H. (1952). A measure of asymptotic efficiency for tests of a hypothesis based on the sum of observations. Ann. Math. Statistics 23 493–507.
• Cichocki, A. and Amari, S.-i. (2010). Families of alpha- beta- and gamma-divergences: Flexible and robust measures of similarities. Entropy 12 1532–1568.
• Cowell, F. A., Davidson, R. and Flachaire, E. (2011). Goodness of fit: An axiomatic approach. Groupement de Recherche en Economie Quantitative D’Aix-Marseille (GREQAM) DT 2011-50. Available at http://halshs.archives-ouvertes.fr/docs/00/63/90/75/PDF/DTGREQAM2011_50.pdf.
• Cowell, F. A. and Kuga, K. (1981). Additivity and the entropy concept: An axiomatic approach to inequality measurement. J. Econom. Theory 25 131–143.
• Cressie, N. and Read, T. R. C. (1984). Multinomial goodness-of-fit tests. J. Roy. Statist. Soc. Ser. B 46 440–464.
• Elbers, C., Lanjouw, P., Mistiaen, J. A. and Özler, B. (2008). Reinterpreting between-group inequality. Journal of Economic Inequality 6 231–245.
• Fay, R. E. (1989). Theoretical application of weighting for variance calculation. In Proceedings of the Section on Survey Research Methods 212–217. Amer. Statist. Assoc., Alexandria, VA.
• Firebaugh, G. (1999). Empirics of world income inequality. American Journal of Sociology 104 1597–1630.
• Fleurbaey, M. and Schokkaert, E. (2009). Unfair inequalities in health and health care. J. Health Econ. 28 73–90.
• Frohlich, K. L. and Potvin, L. (2008). Transcending the known in public health practice: The inequality paradox: The population approach and vulnerable populations. Am. J. Public Health 98 216–221.
• Fujisawa, H. and Eguchi, S. (2008). Robust parameter estimation with a small bias against heavy contamination. J. Multivariate Anal. 99 2053–2081.
• Green, L. W. and Fielding, J. (2011). The U.S. healthy people initiative: Its genesis and its sustainability. Annu. Rev. Public Health 32 451–470.
• Harper, S., Lynch, J., Meersman, S. C., Breen, N., Davis, W. W. and Reichman, M. E. (2008). An overview of methods for monitoring social disparities in cancer with an example using trends in lung cancer incidence by area-socioeconomic position and race-ethnicity, 1992–2004. Am. J. Epidemiol. 167 889–899.
• Harper, S., King, N. B., Meersman, S. C., Reichman, M. E., Breen, N. and Lynch, J. (2010). Implicit value judgments in the measurement of health inequalities. Milbank Quaterly 88 4–29.
• Haughton, J. and Khander, S. R. (2009). Handbook on Poverty and Inequality. The World Bank, Washington, DC.
• Judkins, D. R. (1990). Fay’s method for variance estimation. Journal of Official Statistics 6 223–239.
• Keppel, K. G., Pearcy, J. N. and Klein, R. J. (2004). Measuring progress in Healthy People 2010. Healthy People 2010 Stat. Notes 25 1–16.
• Keppel, K., Pamuk, E., Lynch, J., Carter-Pokras, O., Kim, I., Mays, V., Pearcy, J., Schoenbach, V. and Weissman, J. S. (2005). Methodological Issues in Measuring Health Disparities. Vital and Health Statistics, Series 2 141. National Center for Health Statistics, Hyattsville, MD.
• Kullback, S. and Leibler, R. A. (1951). On information and sufficiency. Ann. Math. Statistics 22 79–86.
• Levy, J. I., Chemerynski, S. M. and Tuchmann, J. L. (2006). Incorporating concepts of inequality and inequity into health benefits analysis. International Journal of Equity in Health 5. Available at DOI:10.1186/1475-9276-5-2.
• Lumley, T. (2004). Analysis of complex survey samples. Journal of Statistical Software 9 1–19.
• Lumley, T. (2011). “Survey”: Analysis of complex survey samples. R package version 3.26.
• Mackenbach, J. P. and Kunst, A. E. (1997). Measuring the magnitude of socio-economic inequalities in health: An overview of available measures illustrated with two examples from Europe. Soc. Sci. Med. 44 757–771.
• Magdalou, B. and Nock, R. (2011). Income distributions and decomposable divergence measures. J. Econom. Theory 146 2440–2454.
• Martínez-Camblor, P. (2007). Central limit theorems for $S$-Gini and Theil inequality coefficients. Rev. Colombiana Estadíst. 30 287–300.
• McCarthy, P. J. (1969). Pseudo-replication: Half samples. Revue de l’Institut International de Statistique—Review of the International Statistical Institute 37 239–264.
• National Center for Health Statistics. (2011). Healthy People 2010 Final Review. National Center for Health Statistics, Hyattsville, MD.
• Page, R. C. and Eke, P. I. (2007). Case definitions for use in population-based surveillance of periodontitis. J. Periodontol. 78 1387–1399.
• Pearcy, J. N. and Keppel, K. G. (2002). A summary measure of health disparity. Public Health Reports 117 273–280.
• Pollard, D. E. (2002). A User’s Guide to Measure Theoretic Probability. Cambridge Univ. Press, Cambridge, UK.
• R Development Core Team. (2011). R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0. Available at http://www.R-project.org.
• Rao, J. N. K. and Wu, C. F. J. (1988). Resampling inference with complex survey data. J. Amer. Statist. Assoc. 83 231–241.
• Rao, J. N. K., Wu, C. F. J. and Yue, K. (1992). Some recent work in resampling methods. Survey Methodology 18 209–217.
• Rényi, A. (1960). On measures of entropy and information. In Proc. 4th Berkeley Sympos. Math. Statist. and Prob. 547–561. Univ. California Press, Berkeley, CA.
• Rose, G. (1985). Sick individuals and sick populations. International Journal of Epidemiology 14 32–38.
• Shorrocks, A. F. (1980). The class of additively decomposable inequality measures. Econometrica 48 613–625.
• Talih, M. (2013a). Supplement to “A reference-invariant health disparity index based on Rényi divergence—technical appendix.” DOI:10.1214/12-AOAS621SUPPA.
• Talih, M. (2013b). Supplement to “A reference-invariant health disparity index based on Rényi divergence—additional case study from NHANES.” DOI:10.1214/12-AOAS621SUPPB.
• Talih, M. (2013c). Supplement to “A reference-invariant health disparity index based on Rényi divergence—R syntax and output files.” DOI:10.1214/12-AOAS621SUPPC.
• Theil, H. (1967). Economics and Information Theory. North Holland, Amsterdam, Netherlands.
• U.S. Department of Health and Human Services. (2000). Healthy People 2010, 2nd ed: With Understanding and Improving Health and Objectives for Improving Health, Vol. 2. U.S. Government Printing Office, Washington, DC.
• U.S. Department of Health and Human Services. (2006). Healthy People 2010 Midcourse Review. U.S. Government Printing Office, Washington, DC.
• van Erven, T. A. L. (2010). When data compression and statistics disagree: Two frequentist challenges for the minimum description length principle. Ph.D. thesis, Leiden University—CWI, the Netherlands. ISBN 978-90-9025673-3. Available at http://hdl.handle.net/1887/15879.
• Wagstaff, A., Paci, P. andvan Doorslaer, E. (1991). On the measurement of inequalities in health. Soc. Sci. Med. 33 545–557.

#### Supplemental materials

• Supplementary material A: Technical appendix: Decomposability. Expressions and variance calculations for the total or aggregate RI and SRI and their within-group components when individual-level data are continuous.
• Supplementary material B: Additional case study from NHANES. Disparities in mean total blood cholesterol levels ($\mu\mathrm{g/dL}$) in U.S. adults aged 20 and over, 2005–08.
• Supplementary material C: R syntax and output files. Syntax and output from case studies comparing the equally-weighted and population-weighted RI and SRI; their group-specific, between-, and within-group components; and their design-based standard errors and sampling distributions, obtained via Taylor series linearization, balanced repeated replication, and rescaled bootstrap. Syntax is reverse-compatible with that in Borrell and Talih (2011, 2012).