Correction

Correction: what potential has tobacco control for reducing health inequalities? The New Zealand situation

Nick Wilson^1*, Tony Blakely¹ and Martin Tobias²

• * Corresponding author: Nick Wilson nick.wilson@otago.ac.nz

Author Affiliations

¹ Department of Public Health, Wellington School of Medicine & Health Sciences, University of Otago, PO Box 7343, Wellington South, New Zealand

² Ministry of Health, PO Box 5013, Wellington, New Zealand

For all author emails, please log on.

International Journal for Equity in Health 2006, 5:16 doi:10.1186/1475-9276-5-16

The electronic version of this article is the complete one and can be found online at: http://www.equityhealthj.com/content/5/1/16

Received:	5 December 2006
Accepted:	18 December 2006
Published:	18 December 2006

© 2006 Wilson et al; licensee BioMed Central Ltd.

This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

This is a correction article.

Text

In our article [1] there was an error in the calculation of population attributable risk percents (PAR%) for 1996–99 shown in Table 1 (bottom row). The corrected Table is in this correction article (see Table 1). This correction has also required revising Figure 2 (see Figure 1 in this correction article). In the process of making these corrections, we have extended our presentation of the contribution of smoking to mortality gaps by ethnicity and education to include a more explicit acknowledgement of the choice of counterfactual assumption. Figure 1 now shows the estimated 'never smoker rate' plus the 'smoking attributable rate' for each ethnic and educational group. Note that the 'smoking attributable rate' as a percentage of the total rate is equivalent to the relevant PAR% shown in Table 1. 'A' and 'B' signify two alternative counterfactual scenarios that can be used to estimate the contribution of smoking to ethnic or socioeconomic gaps in mortality. Scenario A for ethnic gaps is whereby the non-Māori non-Pacific (nMnP) population adopt the smoking rates of Māori, calculated using direct standardisation as given elsewhere [ref 27 of the original paper]. Scenario A for education gaps is whereby each educational group is given an 'average' smoking rate, calculated using Poisson regression as given elsewhere [ref 75 of the original paper]. Scenario B is more extreme (and arguably somewhat unrealistic) whereby we assume there had never been smoking in New Zealand, with the area labelled 'B' in Figure 1 being that for Scenario B over and above that for Scenario A. The contribution of smoking to gaps under Scenario B is calculated using standard population attributable rate methods, that is the difference in "attributable smoking rates" between Māori and nMnP or between nil and post-school qualifications. Thus estimating the contribution of smoking to mortality gaps depends on how extreme the counterfactual assumptions are [2]. Halving total population smoking rates, and making smoking rates for all ethnic and socioeconomic groups the same, might (allowing for time lags) close mortality gaps by an amount mid-way between Scenarios A and B shown in Figure 1.

References