Using routinely collected health data to investigate the association between ethnicity and breast cancer incidence and survival: what is the impact of missing data and multiple ethnicities?

Cancer Epidemiology Group, Centre for Epidemiology & Biostatistics, University of Leeds, Level 6 Bexley Wing, St. James' University Hospital, Leeds, UK.
Ethnicity and Health (Impact Factor: 1.28). 04/2011; 16(3):201-12. DOI: 10.1080/13557858.2011.561301
Source: PubMed

ABSTRACT The aims of this study were to: (1) investigate the relationship between ethnicity and breast cancer incidence and survival using cancer registry and Hospital Episode Statistics (HES) data; and (2) assess the impact of missing data and the recording of multiple ethnicities for some patients.
A total of 48,234 breast cancer patients diagnosed between 1997 and 2003 in two English regions were identified. Ethnicity was missing in 16% of cases. Multiple imputation (10 iterations) of missing ethnicity was undertaken using a range of predictor variables. Multiple ethnicities for a single patient were recorded in 4% of cases. Three methods of assigning ethnicity were used: 'most popular' code, 'last recorded' code, and proportions calculated using all recorded episodes for each patient. Age-standardised incidence rate ratios (IRR) and 5-year survival were calculated before and after imputation for the three methods of assigning ethnicity.
Breast cancer incidence was lower in the South Asian group (IRR=0.59, 95% confidence interval [CI] 0.51-0.69 compared to the White group). In unadjusted analyses, the South Asian group had consistently higher survival compared with the White group (hazard ratio [HR]=0.81, 95% CI 0.68-0.95). After adjustment for age and stage, there were no survival differences amongst the White, South Asian and Black groups. Survival was higher in the 'Other' ethnic group when using the 'last recorded' method to assign ethnicity (HR=0.62, 95% CI 0.45-0.85 compared with the White group). The results were similar before and after imputation, using all three methods of assigning ethnicity.
Breast cancer incidence was lower in the South Asian group than in the White group. After adjusting for casemix there were no consistent survival differences amongst the ethnic groups. Although the impact of missing data and multiple ethnicities was minimal in this study, researchers should always consider these issues, as the results may not be generalisable to other populations and datasets.

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Although international comparisons reveal large geographical differences in the incidence of breast and gynaecological cancers, incidence data for ethnic groups in England remains scarce. We compared the incidence of breast, ovarian, cervical and endometrial cancer in British Indians, Pakistanis, Bangladeshis, Black Africans, Black Caribbeans, Chinese and Whites between 2001 and 2007. We identified 357,476 cancer registrations from which incidence rates were calculated using mid-year population estimates from 2001 to 2007. Ethnicity was obtained through linkage to the Hospital Episodes Statistics database. Incidence rate ratios were calculated, comparing the 6 non-White ethnic groups to Whites, and were adjusted for age and income. We found evidence of differences in the incidence of all 4 cancers by ethnic group (p < 0.001). Relative to Whites, South Asians had much lower rates of breast, ovarian and cervical cancer (IRRs of 0.68, 0.66 and 0.33 respectively), Blacks had lower rates of breast, ovarian and cervical cancer but higher rates of endometrial cancer (IRRs of 0.85, 0.62, 0.72 and 1.16 respectively), and Chinese had lower rates of breast and cervical cancer (IRRs of 0.72 and 0.68 respectively). There were also substantial intra-ethnic differences, particularly among South Asians, with Bangladeshis experiencing the lowest rates of all 4 cancers CONCLUSIONS: Our study provides evidence that the risk of breast and gynaecological cancers varies by ethnic group and that those groups typically grouped together are not homogenous with regards to their cancer risk. Furthermore, several of our findings cannot be readily explained by known risk factors and therefore warrant further investigation.
    BMC Cancer 12/2014; 14(1):979. DOI:10.1186/1471-2407-14-979 · 3.32 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: South Asians in England have an increased risk of childhood cancer but incidence by their individual ethnicities using self-assigned ethnicity is unknown. Our objective was to compare the incidence of childhood cancer in British Indians and Whites in Leicester, which has virtually complete, self-assigned, ethnicity data and the largest population of Indians in England. We obtained data on all cancer registrations from 1996 to 2008 for Leicester with ethnicity obtained by linkage to the Hospital Episodes Statistics database. Age-standardised incidence rates were calculated for childhood cancers in Indians and Whites as well as rate ratios, adjusted for age. There were 33 cancers registered among Indian children and 39 among White children. The incidence rate for Indians was greater compared to Whites for all cancers combined (RR 1.82 (95% CI 1.14 to 2.89); p = 0.01), with some evidence of increased risk of leukaemia (RR 2.20 (0.95 to 5.07); p = 0.07), lymphoma (RR 3.96 (0.99 to 15.84); p = 0.04) and central nervous system tumours (RR 2.70 (1.00 to 7.26); p = 0.05). Rates were also higher in British Indian children compared to children in India. British Indian children in Leicester had an increased risk of developing cancer compared to White children, largely due to a higher incidence of central nervous system and haematological malignancies.
    PLoS ONE 04/2013; 8(4):e61881. DOI:10.1371/journal.pone.0061881 · 3.53 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Background:In the United Kingdom, breast cancer incidence is lower in South Asian and Black women than in White women, but the extent to which this is due to known risk factors is unknown. In a large prospective study, we describe breast cancer incidence by ethnicity, before and after adjustment for known risk factors for the disease.Methods:Women were recruited into the Million Women Study in 1996-2001, when information on reproductive and lifestyle factors known to influence the risk of breast cancer was obtained. Ethnicity was determined from study questionnaires and hospital admission data. Cox regression models were used to calculate adjusted relative risks (RR) for incident breast cancer in South Asians and Blacks compared with Whites.Results:Analyses included 5877 South Asian, 4919 Black, and 1 038 144 White women in England. The prevalence of 8 out of the 9 risk factors for breast cancer examined, differed substantially by ethnicity (P<0.001 for each), such that South Asian and Black women were at a lower risk of the disease than White women. During 12.2 years of follow-up incident breast cancer occurred in 217 South Asians, 180 Blacks, and 45 191 Whites. As expected, breast cancer incidence was lower in South Asians (RR=0.82, 95% CI 0.72-0.94) and Blacks (RR=0.85, 0.73-0.98) than in Whites when the analyses were adjusted only for age and region of residence. However, after additional adjustment for the known risk factors for the disease, breast cancer incidence was similar to that of Whites, both in South Asians (0.95, 0.83-1.09) and in Blacks (0.91, 0.78-1.05).Conclusion:South Asian and Black women in England have lower incidence rates of breast cancer than White women, but this is largely, if not wholly, because of differences in known risk factors for the disease.British Journal of Cancer advance online publication, 29 October 2013; doi:10.1038/bjc.2013.632
    British Journal of Cancer 10/2013; 110(1). DOI:10.1038/bjc.2013.632 · 4.82 Impact Factor