Caroline Shaw1, June Atkinson, Tony Blakely. 1. Department of Public Health, University of Otago Wellington, PO Box 7343, Wellington South, New Zealand. caroline.shaw@otago.ac.nz
Abstract
BACKGROUND: Māori and Pacific peoples are undercounted in cancer incidence statistics relative to census statistics. We use linked census and Cancer Registry data sets to determine the extent of misclassification between 1981 and 2004. METHODS: The 1981, 86, 91, 96 and 2001 censuses were anonymously and probabilistically linked to individuals with a cancer registration for the entire intercensal period, or 31 December 2004 in the 2001 cohort. We compared counts by ethnicity between census and Cancer Registry data. Correction ratios and percentage under or overcounts are presented. RESULTS: Undercounting of Māori and Pacific peoples was marked in the first cohort. For example Māori were undercounted 31% on the Cancer Registry compared to self reported ethnic origin in the 1981-86 cohort. Gradual improvements were seen and by the 2001 cohort undercounting was 15% and 10% for Māori and Pacific peoples respectively. For Asian people undercounting improved from 68% to 13% over the time studied. Reciprocally, non māori /Pacific/Asian peoples were consistently overcounted. CONCLUSION: There is undercounting of Māori, Pacific and Asian events on cancer registration data, relative to census data, throughout 1981-2004. Cancer incidence rates need revising, to correctly understand the epidemiology and to inform cancer policy. Steps to improve the quality of ethnicity information remain a priority.
BACKGROUND: Māori and Pacific peoples are undercounted in cancer incidence statistics relative to census statistics. We use linked census and Cancer Registry data sets to determine the extent of misclassification between 1981 and 2004. METHODS: The 1981, 86, 91, 96 and 2001 censuses were anonymously and probabilistically linked to individuals with a cancer registration for the entire intercensal period, or 31 December 2004 in the 2001 cohort. We compared counts by ethnicity between census and Cancer Registry data. Correction ratios and percentage under or overcounts are presented. RESULTS: Undercounting of Māori and Pacific peoples was marked in the first cohort. For example Māori were undercounted 31% on the Cancer Registry compared to self reported ethnic origin in the 1981-86 cohort. Gradual improvements were seen and by the 2001 cohort undercounting was 15% and 10% for Māori and Pacific peoples respectively. For Asian people undercounting improved from 68% to 13% over the time studied. Reciprocally, non māori /Pacific/Asian peoples were consistently overcounted. CONCLUSION: There is undercounting of Māori, Pacific and Asian events on cancer registration data, relative to census data, throughout 1981-2004. Cancer incidence rates need revising, to correctly understand the epidemiology and to inform cancer policy. Steps to improve the quality of ethnicity information remain a priority.
Authors: Hellen Geremias Dos Santos; Carla Ferreira do Nascimento; Yeda Aparecida de Oliveira Duarte; Ichiro Kawachi; Alexandre Dias Porto Chiavegatto Filho Journal: Int J Public Health Date: 2019-12-17 Impact factor: 3.380
Authors: Gabi U Dachs; Maiko Kano; Ekaterina Volkova; Helen R Morrin; Valerie C L Davey; Gavin C Harris; Michelle Cheale; Christopher Frampton; Margaret J Currie; J Elisabeth Wells; Bridget A Robinson Journal: BMC Cancer Date: 2010-10-10 Impact factor: 4.430
Authors: Tigran Avoundjian; Julia C Dombrowski; Matthew R Golden; James P Hughes; Brandon L Guthrie; Janet Baseman; Mauricio Sadinle Journal: JMIR Public Health Surveill Date: 2020-04-30
Authors: Andrea M Teng; June Atkinson; George Disney; Nick Wilson; Diana Sarfati; Melissa McLeod; Tony Blakely Journal: BMC Cancer Date: 2016-09-26 Impact factor: 4.430