Aggregating Data Hides Cancer Disparities Affecting Asian and NHPI Patients in the US

There are steps clinicians and researchers can take to overcome the problems stemming from aggregating data across Asian and NHPI populations in oncology and other fields. 

Breast Cancer Findings Contradict Prior Narrative

In a 2021 study, Dr Yu and colleagues found that outcomes of breast cancer vary across subgroups of Asian-American patients, a finding that contradicts the previously accepted narrative.3  

“The traditional teaching I received in medical school is simply that Asians — in aggregate — have better mortality rates than NHW women,” Dr Yu said. 

What she and her colleagues found in their study was that patients with Southeast Asian ancestry had worse outcomes than other Asian subgroups, and Japanese-American patients had better outcomes. The Southeast Asian patients had the lowest unadjusted cancer-specific survival rate at 10 years, and Japanese patients had the highest. In adjusted analyses, all subgroups except Southeast Asian patients had a lower risk of cancer-specific death or any death, compared with NHW patients.

“Japanese women, who historically have had a different immigration pattern than many other ethnicities, were more similar to NHW women than other Asian subgroups along many demographic factors,” Dr Yu noted.

These factors include older median age at diagnosis, earlier stage of disease and presentation, higher rates of health insurance coverage, less urban geographic location, greater likelihood of hormone receptor-positive tumors, lower rates of mastectomy and chemotherapy, and higher rates of adjuvant radiation therapy.

Differences in screening for breast cancer may also play a role in the observed disparities. A 2019 study showed differences in breast cancer screening among Vietnamese- Chinese-, and Korean-American patients.8 The Korean patients had lower rates of mammography than the Vietnamese or Chinese patients. Among Chinese patients, the odds of never having a mammogram decreased as the number of physician visits increased. 

“Perhaps encouraging visits in general would be more effective in increasing mammography rates in Chinese women, whereas the study suggests structural barriers such as lack of health insurance were more likely the root of the lower Korean mammography rates,” Dr Yu noted. “Therefore, to increase mammography rates among ‘Asians,’ these groups would be best served with ethnic-tailored, distinct outreach initiatives — an important distinction that would have been lost in aggregation.” 

Ongoing Research Needs and Using Data in Practice

There are steps clinicians and researchers can take to overcome the problems stemming from aggregating data across Asian and NHPI populations in oncology and other fields. 

Dr Yu recommends that clinicians and researchers consider whether the data at hand is relevant to patients before citing statistics or developing new studies or public health initiatives. 

“It’s easy to fall back on big-picture racial differences — and that’s not always wrong — but it’s worth considering if there is data more specific to the goal you are trying to achieve,” Dr Yu said. 

Failing to account for such data, if available, could lead to ineffective and potentially harmful results, she added. “That oversight could lead to undue psychological stress, such as for a patient with a new diagnosis, or a public health initiative that fails to be effective,” Dr Yu said. 

Dr Taparra outlined 3 key steps that “those in power should consider in order to combat systemic racism through data disaggregation.” First, the data must be collected. If researchers do not start with granular data, there is no way of evaluating the subgroups within the broader Asian population. 

Secondly, Dr Taparra said, the data must be disaggregated appropriately. Disaggregation on the most granular level may not be possible if there are too few patients of a given ethnicity. However, this may hide disparities, as researchers will be unequipped to assess true differences. Researchers should consider a middle level of grouping, such as grouping data by region of ancestry (ie, South Asia, Southeast Asia, and East Asia).

Lastly, the disaggregated data should be reported, Dr Taparra said. Collecting the granular data without disaggregating the groups continues to hide the fact that Asian patients are not a monolithic group. 

Dr Taparra noted that ongoing research is exploring the extent of disaggregation that should occur when evaluating health disparities in various Asian subgroups. 

“The problem with disaggregating all nearly 50 country or regional ethnicities in Asia is that we then lose power to make clear distinctions between the groups,” he explained. “An intermediate level of disaggregation into South Asia, Southeast Asia, and East Asian groupings should be strongly considered, as individuals with these groupings often do share the same trends when disaggregated further into individual ethnicities.” 

Dr Taparra also emphasized that data from NHPI patients should be collected, disaggregated, and reported separately from those of Asian patients. 

Dr Yu added that more funding should be directed toward research investigating key differences between groups. “There is no shortage of researchers who want to put in the work, and it’s our responsibility as a community to support these endeavors,” she said.

Disclosures: Dr Yu and Dr Taparra reported having no relevant disclosures.

This article originally appeared on Cancer Therapy Advisor


1. Taparra K, Pellegrin K. Data aggregation hides Pacific Islander health disparities. Lancet. 2022;400(10345):2-3. doi:10.1016/S0140-6736(22)01100-X

2. Pham C, Fong TL, Zhang J, Liu L. Striking racial/ethnic disparities in liver cancer incidence rates and temporal trends in California, 1988-2012. J Natl Cancer Inst. 2018;110(11):1259-1269. doi:10.1093/jnci/djy051

3. Yu AYL, Thomas SM, DiLalla GD, et al. Disease characteristics and mortality among Asian women with breast cancer. Cancer. 2022;128(5):1024-1037. doi:10.1002/cncr.34015

4. Lee AW, Sou A, Liu L. Abstract PO-198: Disparities in nasopharyngeal cancer incidence among Asian American ethnic subgroups. Cancer Epidemiol Biomarkers Prev. 2022;31(1_Supplement):PO-198. doi:10.1158/1538-7755.DISP21-PO-198

5. Li DY, VoPham T, Tang MC, Li CI. Disparities in risk of advanced-stage liver cancer and mortality by race and ethnicity. J Natl Cancer Inst. 2022;114(9):1238-1245. doi:10.1093/jnci/djac097

6. Taparra K, Qu V, Pollom E. Disparities in survival and comorbidity burden between Asian and Native Hawaiian and other Pacific Islander patients with cancer. JAMA Netw Open. 2022;5(8):e2226327. doi:10.1001/jamanetworkopen.2022.26327

7. Taparra K, Dee EC, Dao D, Patel R, Santos P, Chino F. Disaggregation of Asian American and Pacific Islander women with stage 0-II breast cancer unmasks disparities in survival and surgery-to-radiation intervals: A National Cancer Database analysis from 2004 to 2017. JCO Oncol Pract. 2022;18(8):e1255-e1264. doi:10.1200/OP.22.00001

8. Shon EJ, Townsend AL. Predictors of never having a mammogram among Chinese, Vietnamese, and Korean immigrant women in the U.S. PLoS One. 2019;14(11):e0224505. doi:10.1371/journal.pone.0224505