BACKGROUND: Minnesota has the second largest Hmong population in the United States. The objective of the current study was to estimate the cancer incidence among Hmong individuals in Minnesota between 2000 and 2012 to determine targets for screening and interventions. METHODS: Cancer cases in Minnesota between 2000 and 2012 were obtained from the Minnesota Cancer Surveillance System, and proportional incidence ratios (PIRs) were calculated. The 2000 and 2010 US Census reports were used to obtain total population estimates. Age-adjusted cancer incidence rates (AAR) and 95% confidence intervals (95% CIs) were calculated for Hmong individuals, Asian/Pacific Islander individuals, and all Minnesotans using direct method and Poisson regression. RESULTS: Compared with all Minnesotans, the Hmong had elevated PIRs and AARs for malignancies related to infections, including nasopharyngeal, stomach, liver, and cervical cancers. The AAR ratios in Hmong versus all Minnesotans were found to be significantly increased for nasopharyngeal (AAR, 15.90; 95% CI, 9.48-26.68), stomach (AAR, 2.99; 95% CI, 2.06-4.33), liver (AAR, 1.77; 95% CI, 1.04-3.02), and cervical (AAR, 3.88; 95% CI, 2.61-5.77) cancers. The AARs in Hmong versus all Minnesotans were significantly lower for all-cause cancer (AAR, 0.39; 95% CI, 0.35-0.44); cancers of the breast, lung, and colorectum; melanoma; and non-Hodgkin lymphoma. Compared with Asian/Pacific Islander individuals, the rates in Hmong were significantly higher for melanoma and cervical cancer, with AAR ratios of 2.23 (95% CI, 1.09-4.56) and 1.59 (95% CI, 1.01-2.49), respectively. CONCLUSIONS: Compared with all Minnesotans, the Hmong have an increased incidence of cancers related to infectious agents. These findings indicate a need for cancer prevention and screening programs in this population.
Bibliographical noteFunding Information:
The collection of the cancer incidence data used in this publication was partially supported by Cooperative Agreement Number 5 NU58DP003922 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention.
The collection of the cancer incidence data used in this publication was partially supported by Cooperative Agreement Number 5 NU58DP003922 from the Centers for Disease Control and Prevention. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the Centers for Disease Control and Prevention. The Minnesota Cancer Surveillance System (MCSS) database was used to identify cancers in the entire Minnesota population and in Hmong in Minnesota between 2000 and 2012. The MCSS is a population-based cancer registry for the state, and has been in operation since 1988. Cancers diagnosed in Minnesota residents are identified in 2 ways: 1) submission of pathology reports carrying a diagnosis of cancer by pathology laboratories; and 2) submission of standardized case abstracts by hospital-based cancer registries for each patient with cancer diagnosed or treated at the facility. An external audit of the MCSS performed in March 2008 by the National Program of Cancer Registries estimated that case completeness was approximately 99.7% and data accuracy was approximately 96.5%. The MCSS is funded by a state appropriation and a cooperative agreement with the Centers for Disease Control and Prevention through the National Program of Cancer Registries. Hmong and API cancer cases diagnosed between 2000 and 2012 were identified through the MCSS race reporting. Information regarding race in the MCSS was obtained from health care facilities that recorded patients' race and was supplemented using manual examination by MCSS staff of the records in those facilities and birth and death records when a case was reported without information regarding race/ethnicity. The denominators for analysis came from the US Census Bureau, which obtained self-reported race information in response to the 2000 or 2010 census questionnaire. Census information was obtained through a search for ?Hmong? and ?Minnesota? using the American FactFinder. We calculated age-adjusted and sex-adjusted PIRs as described by Ross et al. Observed-to-expected ratios were calculated. The observed number is the actual number of Hmong cancer cases for a particular cancer site. In generating the expected number, cancer cases were stratified into 20-year age groups (ages birth to 19 years, 20-39 years, 40-59 years, 60-79 years, and ?80 years) and by sex. The expected number in each stratum was the number of site-specific cancers in all Minnesotans divided by the number of total cancers in all Minnesotans, multiplied by the number of total cancers in the Hmong. The total number was found by summing the expected values from each stratum. The 95% CIs were calculated using the methods described by Ury and Wiggins. We also calculated average annual AARs for all-cause cancer and site-specific cancers for Hmong individuals in Minnesota, API individuals in Minnesota, and all Minnesotans using the direct method, and these were verified using Poisson regression. The numerator included all incident cancer cases for the corresponding group diagnosed between 2000 and 2012. The denominator was calculated as the average of US Census population counts in 2000 and 2012 multiplied by 13. The 2000 US population was used as the standard weight. Adjustment was performed through 5-year age groups starting from birth to 4 years and ending at age ?80 years. In addition, Poisson regression was used to calculate AAR ratios and 95% CIs for all groups: Minnesota Hmong, Minnesota API, and all Minnesotans. Microsoft Excel (version 14.4.1; Microsoft Corporation, Redmond, Washington) was used to calculate PIRs and AARs. SAS statistical software (version 9.4; SAS Institute Inc, Cary, North Carolina) was used to verify the AARs and calculate the rate ratios, 95% CIs, and P values.
- Asian/Pacific Islander (API)
- age-adjusted cancer incidence rates (AARs)