Data Disaggregation of Asian Americans: Implications for the Physician Workforce

• Asian Americans
• Ethnicity
• Family Medicine
• Family Physicians
• Primary Health Care
• Socioeconomic Factors
• Workforce
• Workforce Diversity

Health equity advocates have long called for data disaggregation to address health disparities with more representative population data.¹ Data disaggregation involves the breakdown of large categories into specific component parts for further analysis. In racial and ethnic data, aggregated data sets have especially adversely impacted the Asian American population, a heterogenous group with roots in more than 20 countries and wide variations in socioeconomic factors across subgroups.² When reported as a whole, disparities faced by Asian Americans are often masked and misrepresented.^2,3 For instance, Asian Americans have the highest education levels compared with all other racial or ethnic groups, but >50% of people who identify as Burmese, Cambodian, Laotian, or Bhutanese only have a high school education or less as their highest level of educational attainment.⁴ Up to 31.9% of the Asian American population has limited English proficiency, but when detailed by ethnicity, Chinese, Bangladeshi, and Vietnamese populations report 43 to 49% limited English proficiency.⁴

The American Board of Family Medicine (ABFM) recently revised their race data collection standards to include the 6 largest Asian American ethnicities: Chinese, Filipino, Indian, Japanese, Korean, and Vietnamese.⁵ For those who self-report as Asian, respondents can select any of the subgroups and/or an “Other” option. In the article “Disaggregation of Asian-American Family Physicians,” Huynh et al analyzed responses from 16,348 family physicians and reported the disaggregated prevalence of each ethnic subgroup for those who identified as Asian. Although Asian Americans as a collective are well represented in family medicine (17.1 to 25.6% of all analyzed family physicians vs only 5.8% of the general US population), there are several subgroups with lower prevalence in family medicine compared with their prevalence in the general US population.⁵

To our knowledge, the ABFM’s disaggregated data on Asian American family physicians is the first disaggregated report on national Asian American physician data, as previous reports were on medical student applicants or state-based.^6,7 At nearly 19% of all active physicians in the US, Asian Americans represent a substantial proportion of the physician workforce.⁸ Disaggregating physician data are imperative to not only better understand the diversity of this population, but will also help initiatives striving to improve health equity. For instance, ethnic and language concordance between patients and clinicians is associated with improved patient outcomes and satisfaction.^9,10 Through understanding who actually comprises the Asian American physician workforce, we can begin to identify if specific subpopulations are adequately represented and cared for. Further opportunities for analysis can include how these physicians are distributed by specialty, location, or scope of practice. This data can then be compared with ethnicity population data to address health disparities such as cancer screening gaps, mental health care needs, or language access.

Data disaggregation promotes adequate representation of populations previously ignored or marginalized. The aggregated total percentage of Asian American physicians may seem high compared with the percentage of Asian Americans in the general US population, perpetuating a false notion that all Asian Americans are overrepresented in medicine. However, data disaggregation can show whether representation of some Asian ethnicities in the physician workforce are disproportionate to their representation in the general population. This aligns with the Association of American Medical Colleges’ 2004 expanded definition of “underrepresented in medicine.”¹¹ It is important that programs that support diversity and inclusion account for underrepresented Asian ethnicities—especially those with significant socioeconomic disparities compared with other subgroups. This, in turn, ensures that pathway programs, mentorship opportunities, and resources are justly distributed.

When considering Asian American representation in medicine, it is also critical to examine the significant gaps in Asian American representation in leadership roles. In 2022, the Association of American Medical Colleges found that Asian American physicians are the only group with disproportionately fewer medical school department chairs than percentage of faculty, with Asian American physicians representing 22.8% of full-time medical school faculty, but only 11.1% of department chairs.^12,13 In comparison, Black or African American physicians represented 4.3% of full-time faculty and 4.0% of department chairs and Hispanic or Latino/a physicians represented 3.7% of full-time faculty and 3.9% of department chairs.

Disproportionately low Asian American representation in leadership extends beyond academia. A 2022 study found that Asian principal investigators receiving National Science Foundation funding received the lowest rate of funding per award.¹⁴ Another study at the National Institutes of Health (NIH) found that although 22.1% of NIH employees in nonleadership positions were Asian or Asian American, only 6% were in senior leadership positions.¹⁵ In comparison, 50.5% of NIH employees in nonleadership positions were White, but represented 81.2% of senior leadership positions. These findings highlight unique barriers Asian American physicians encounter in attaining leadership positions in medicine. Although data disaggregated by ethnicity was not collected in these studies, it can be hypothesized that physicians identifying as certain Asian ethnicities represent an even lower percentage of physicians in leadership roles. As a result, although Asian Americans as a whole are not considered underrepresented in medicine, by not classifying Asian Americans as underrepresented in medical leadership and not disaggregating data by ethnicity, Asian American physicians are often not given the necessary resources, training, and support to attain leadership roles.

Although ABFM’s move toward collecting data for 6 ethnicities is a step in the right direction for data disaggregation, more work can be done. Sixteen to 20% of participating Asian American family physicians selected “Other” ethnicity, suggesting an undetermined number of hidden subgroups and the complexities of Asian American population data collection.⁵ Although there are challenges of collecting data using a more expansive list of ethnicities, such as small sample size, collecting data using a comprehensive list of ethnicities is critical. This is because Asian Americans are not a monolith, and some of the most marginalized groups that experience significant health disparities and are severely underrepresented in medicine are those not counted within the ABFM’s 6 listed ethnicities.

It is also important to recognize disaggregated Native Hawaiian and Pacific Islander (NHPI) data as an issue of health equity, as NHPI groups have historically been grouped with Asian Americans despite representing separate areas of the world and unique histories.^4,16 Especially in research, the systematic erasure of NHPI through data aggregation hides their significant disparities and perpetuates worse outcomes.¹⁶ Of note, the ABFM does collect data on physicians identifying as NHPI as a category separate from Asian Americans, who represent 0.5% of family physicians.¹⁷ It is critical to continue separately collecting and analyzing NHPI data to address ongoing inequities.

Comprehensive and accurate data are critical to eliminating health disparities. ABFM has taken an important first step, but disaggregating data on race and ethnicity across all medical specialties is important to diversifying the physician workforce and medical leadership. Diverse leadership in medicine has the power to influence research and funding priorities that center the needs of diverse populations, benefiting patients and communities and improving health outcomes. Therefore, we must make every effort to disaggregate data as we work toward achieving health equity.

• Received for publication March 1, 2024.
• Accepted for publication March 15, 2024.