
==== Front
JTCVS Open
JTCVS Open
JTCVS Open
2666-2736
Elsevier

S2666-2736(24)00113-X
10.1016/j.xjon.2024.04.010
Thoracic: Lung Cancer
Identifying Asian American lung cancer disparities: A novel analytic approach
Gu Yunna BS a
Becker Les R. PhD, MSMEdL b
Khaitan Puja G. MD, FACS c
Lazar John F. MD jflazar@me.com
d∗
a Georgetown University School of Medicine, Washington, DC
b MedStar Institute for Innovation, Simulation Training, and Education Lab, MedStar Health, Washington, DC
c Department of Thoracic Surgery, Sheikh Shakhbout Medical City, Abu Dhabi, United Arab Emirates
d Division of Thoracic Surgery, MedStar Washington Hospital Center, Washington, DC
∗ Address for reprints: John F. Lazar, MD, Division of Thoracic Surgery, MedStar Washington Hospital Center, 110 Irving St, Washington, DC 20010. jflazar@me.com
22 4 2024
8 2024
22 4 2024
20 153164
11 5 2023
19 3 2024
8 4 2024
© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by/4.0/ This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).
Objective

Asian Americans include heterogeneous subpopulations with unique burden as the only racial group with cancer as the leading cause of death. The purpose of the study was to identify differences in clinical stage and survival of patients with lung cancer between Asian Americans and its subgroups relative to other racial groups.

Methods

Patients with lung cancer from 2016 National Cancer Database were divided into East Asian, Southeast Asian, South Asian subgroups based on geographic origins, and a composite Asian American group with White non-Hispanic, Black, and Hispanic comparison groups. Columnar z score analysis with adjusted residuals was employed and the terms underrepresented and overrepresented were utilized to describe significant statistical findings.

Results

A total of 825,448 patients were analyzed. Asian Americans were underrepresented relative to White non-Hispanics in all clinical stages except IIIB and IV. In clinical stage IV, Asian Americans (51.0%), East Asians (47.2%), Southeast Asians (57.4%), and South Asians (52.2%) were overrepresented relative to White non-Hispanics (42.2%) and Southeast Asians were overrepresented relative to East Asians and South Asians. For survival across all stages, Asian Americans were overrepresented relative to White non-Hispanics and Blacks, but in clinical stage IV, Southeast Asians (17.9%) were underrepresented relative to East Asians (26.0%) and South Asians (26.6%).

Conclusions

This is the first study to address lung cancer disparity in Asian American subgroups employing a novel analytical approach. Asian American subgroups demonstrated more advanced lung cancer diagnosis yet higher survival compared with White non-Hispanics, Blacks, and/or Hispanics with differences between subgroups. Interplay of complex factors may contribute to Asian American health disparities.

Graphical Abstract

Key Words

health care disparities
lung cancer
smoking cessation
Asian Americans
columnar z-score analysis
lung cancer survivorship
Abbreviations and Acronyms

EGFR epidermal growth factor receptor(s)

NCDB National Cancer Database

NSCLC non–small cell lung cancer
==== Body
pmc East Asian, Southeast Asian, South Asian are more likely to be diagnosed at advanced stages of lung cancer yet demonstrate higher survival compared to White Non-Hispanic, Black, and/or Hispanic with differences between subgroups.

Central Message

Asian American subgroups are more likely to be diagnosed with advanced lung cancer yet show greater survival compared with White non-Hispanics, Blacks, and/or Hispanics with differences between subgroups.

Perspective

First study addressing lung cancer disparity in Asian American subgroups defined by geographic origins with a novel analytical approach. Asian Americans are more likely to be diagnosed with advanced lung cancer yet greater survival compared with White Non-Hispanics, Blacks, and/or Hispanics with differences between subgroups. Interplay of complex factors may contribute to health disparities.

See Discussion on page 165.

Asian is defined as “a person having origins in any of the original peoples of the Far East, Southeast Asia, or the Indian subcontinent” by the US Census Bureau.1 In the 2020 US Census, Asian, Native Hawaiian, or Other Pacific Islanders alone accounted for 6.2% of the total US population, which increased by 35.6% since 2010 compared with a 7.4% increase in the total US population.2, 3, 4 In addition, Asian Americans have been the fastest growing ethnic group in the past decades in the United States and are projected to increase by 79% by 2050.5 Similar to Hispanic or Latino, Asian American is an umbrella term that covers 21 different peoples of origin with distinct culture, language, history, religion, political systems, demographics, and socioeconomic backgrounds. Due to the heterogeneous nature of the Asian American population, generalizing any 1 health care issue has been a significant challenge.

For Asian Americans, lung cancer is perhaps the single greatest health burden because they are the only racial group experiencing cancer as the leading cause of death for both women and men at 25.5% and 24.8%, respectively. This is in stark contrast to all other US racial groups, besides Hispanic women, where heart disease is the leading cause of death.6 From 1990 to 2008, lung cancer consistently ranked among the top-4 cancer sites in Asian Americans with increasing trends among South Asian men, Filipinas, and Korean women. Lung cancer was the most common neoplasm in Kampuchean, Laotian, and Vietnamese men and second most common neoplasm in Indians, Pakistanis, Chinese, Filipinos, Japanese, and Koreans.7 From 1972 to 1988, Southeast Asian men had 18% higher incidence of lung cancer than Whites in Los Angeles County.8

Endemic to most immigrants in the United States is the existence of health disparities. Factors contributing to the prevalence of lung cancer and health disparities for Asian Americans are multifaceted and include the influence of the subgroup's home culture and its integration with US culture, which is extremely nuanced. This can be attributed to a complex interplay of factors such as socioeconomic status, access to health care, language barriers, level of education, cultural beliefs, and discrimination.9, 10, 11, 12, 13

In the past, a major obstacle to studying lung cancer in Asian Americans has been the small sample size. The chief bias associated with this approach is that by aggregating Asian American data, it masks valuable identifiable characteristics between ethnic subgroups, especially in evaluating health disparities.10 To overcome this limitation, a columnar z-score analysis with adjusted standardized residuals was applied to Asian American subgroups identified by geographic origins and the terms underrepresented and overrepresented were utilized to describe significant statistical findings. To our knowledge, this approach was novel to health disparity studies. The primary end point was to identify differences in clinical stage at diagnosis and survival of patients with lung cancer between Asian American subgroups and other racial groups.

Methods

Patients with lung cancer, identified by International Classification of Diseases for Oncology, Third Revision, code C34, were extracted from the 2016 National Cancer Database (NCDB), which included both small cell and non–small cell lung cancer (NSCLC) from 2004 to 2016. The American Joint Committee on Cancer Cancer Staging Manual, sixth edition, was used for cases diagnosed from 2004 to 2009 and the seventh edition was used for cases diagnosed from 2010 to 2016. The Medstar Health Research Institutional Review Board approved the study protocol and publication of data. The informed written consent of patient(s) for the publication of the study data was not required because the NCDB provides de-identified data (institutional review board STUDY00000285, approved March 3, 2019). Cases characterized by American Joint Committee on Cancer clinical stage IA, IB, IIA, IIB, IIIA, IIIB, and IV were retained for further analysis. A significant limitation in prior studies examining lung cancer disparities in Asian Americans is the relatively small proportion of patients who identify as Asian American. To combat this limitation as well as ensure statistical significance and generalizability, 11 Asian American ethnicities were divided into 3 Asian American subgroups based on geographic origins. In addition, we pooled the 3 subgroups to create an Asian American comparison group. A total of 6 ethnic groups were identified in this study as White non-Hispanic, Black, Hispanic, East Asian, Southeast Asian, and South Asian. East Asians were defined as Chinese, Japanese, and Korean. Southeast Asians were defined as Filipino, Vietnamese, Laotian, Hmong, and Thai. South Asians were defined as Indian, Pakistani, and Asian/Indian/Pakistani not otherwise specified. Patients were excluded if clinical stage was 0, 0A, 0is, occult, or unknown and if race was other or unknown. Cross-tabulation analyses were conducted to explore distributional differences across demographic and clinical variables (SPSS version 28; IBM-SPSS Inc). Columnar z-score analysis with Bonferroni correction was employed to identify statistically significant differences between all possible paired comparisons in a table row.14, 15, 16 The χ2 values were calculated to assess each cross-tabulation table. This cross-tabulation approach compares each row entry with every other row entry, assigning an incremented subscript to each row entry as a significant columnar difference is identified. Row entries with common subscript letters designate ethnic group categories whose column proportions; that is, the row entry, do not differ significantly from each other at the .05 level. Conversely, row entries with different subscript letters indicate row entries that do differ significantly from each other at the .05 level. The magnitude and directionality of the adjusted residual was generated for each table cell. The terms overrepresented and underrepresented were employed to describe significant z-score findings at P < .05 with an adjusted residual > +2 or < −2 indicating over- and underrepresentation, respectively. Where appropriate, the magnitude of similarly assigned residuals were compared to assign relative over- and underrepresentation. Agresti16 supports the interpretation of “sample percentages; ” that is, our column percentages and residual in this fashion.

Results

Demographics and Socioeconomic Factors

A total of 825,448 patients were included, resulting in a cohort that was 84.5% White non-Hispanic, 12.3% Black, 2.1% Hispanic, 0.6% East Asian, 0.3% Southeast Asian, and 0.2% South Asian. The proportions of male and female patients were similar within each ethnic group (Table 1). East Asians, Southeast Asians, and South Asians were underrepresented in the first median income quartile and overrepresented in the fourth median income quartile relative to White non-Hispanics, Blacks, and Hispanics (Table 2). East Asians, Southeast Asians, and South Asians were overrepresented as not insured and Medicaid primary payers and underrepresented as Medicare primary payers relative to White non-Hispanics. Southeast Asians and South Asians were more likely to be private insurance/managed care payers relative to all other ethnic subgroups (Table 3). East Asians and South Asians were overrepresented in the first quartile of noncompletion of high school relative to White non-Hispanics, Blacks, and Hispanics, whereas East Asians were also overrepresented relative to White non-Hispanics in the fourth quartile (Table 4).Table 1 Demographic breakdown

Ethnic group	Sex	Total	
Male	Female	
WNH	368,721	(84.1)	329,042	(85.1)	697,763	(84.5)	
Black	55,267	(12.6)	46,546	(12.0)	101,813	(12.3)	
Hispanic	9830	(2.2)	7233	(1.9)	17,063	(2.1)	
EA	2475	(0.6)	2222	(0.6)	4697	(0.6)	
SEA	1282	(0.3)	1082	(0.3)	2364	(0.3)	
SA	1083	(0.2)	665	(0.2)	1748	(0.2)	
Total	438,658	(100.0)	386,790	(100.0)	825,448	(100.0)	
Values are presented as n (%). WNH, White Non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

Table 2 Median income quartile × ethnic group cross-tabulation

Median income	Ethnic group	Total	
WNH	Black	Hispanic	EA	SEA	SA	
First quartile∗
<$40,227								
 Count	128,208a	53,277b	5236c	687d	255e	151e	18,7814	
 % within ethnic group	18.6	53.1	31.0	14.7	10.9	8.7	23.1	
 Adjusted residual	−222.1	241.1	24.7	−13.6	−14.1	−14.2		
Fourth quartile∗
≥$63,333								
 Count	229,561a	13,932b	4074c	2220d	1121d	1011e	251,919	
 % within ethnic group	33.3	13.9	24.1	47.5	47.7	58.3	30.9	
 Adjusted residual	110.2	−124.9	−19.4	24.6	17.7	24.7		
WNH, White non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the First Quartile row: every ethnic group is significantly different from each other execpt for SEA and SA, who are not significantly different from each other.

Table 3 Primary insurance payer × ethnic group cross-tabulation

Insurance status∗	Ethnic group	Total	
WNH	Black	Hispanic	EA	SEA	SA	
Not insured								
 Count	17,218a	5351b	1019c	190d	168c	119b,c	24,065	
 % within ethnic group	2.5	5.3	6.0	4.0	7.1	6.8	2.9	
 Adjusted residual	−56.5	47.4	24.0	4.6	12.1	9.7		
Private insurance/managed care								
 Count	180,868a	24,849b	4224b	1178a,b	840c	601c	212,560	
 % within ethnic group	25.9	24.4	24.8	25.1	35.5	34.4	25.8	
 Adjusted residual	8.3	−10.5	−3.0	−1.1	10.9	8.3		
Medicaid								
 Count	34,118a	14,083b	2380b	789c	291b	258b,c	51,919	
 % within ethnic group	4.9	13.8	13.9	16.8	12.3	14.8	6.3	
 Adjusted residual	−122.5	105.9	41.6	29.7	12.1	14.6		
Medicare								
 Count	446,107a	53,889b	9025b	2401b	1008c	707c	513,137	
 % within ethnic group	63.9	52.9	52.9	51.1	42.6	40.4	62.2	
 Adjusted residual	77.5	−64.9	−25.2	−15.7	−19.6	−18.7		
Other government								
 Count	9245a	1512b	101c	45a,c	21a,b,c	10a,c	10,934	
 % within ethnic group	1.3	1.5	0.6	1.0	0.9	0.6	1.3	
 Adjusted residual	0.1	4.8	−8.5	−2.2	−1.9	−2.8		
Insurance status unknown								
 Count	10,207a	2129b,c	314c	94b,c	36a,c	53b	12,833	
 % within ethnic group	1.5	2.1	1.8	2.0	1.5	3.0	1.6	
 Adjusted esidual	−15.8	14.8	3.0	2.5	−0.1	5.0		
WNH, White non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the Insurance Status Unknown row: WNH is significantly different from Black, Hispanic, EA, and SA; Black is only significantly different from WNH; Hispanic is significantly different from WNH and SA; EA is only significantly different from WNH; SEA is only significantly different from SA; SA is significantly different from WNH, Hispanic, and SEA.

Table 4 Noncompletion of high school degree × ethnic group cross-tabulation

Noncompletion of high school degree	Ethnic group	Total	
WNH	Black	Hispanic	EA	SEA	SA	
First quartile∗ <6.3%								
 Count	158,527a	6539b	1729c	1177d	538a,d	514e	169,024	
 % within ethnic group	23.0	6.5	10.2	25.2	22.9	29.7	20.7	
 Adjusted residual	118.1	−118.7	−34.0	7.6	2.6	9.2		
Fourth quartile∗
≥17.6%								
 Count	120,827a	42,309b	8131c	1522d	474e	361e	173,624	
 % within ethnic group	17.5	42.1	48.1	32.6	20.2	20.8	21.3	
 Adjusted residual	−194.2	172.2	86.1	18.9	−1.3	−0.4		
WNH, White non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the First Quartile row: WNH is significantly different from Black, Hispanic, EA, and SA; Black is significantly different from WNH, Hispanic, EA, SEA, and SA; Hispanic is significantly different from WNH, Black, EA, SEA, and SA; EA is significantly different from WNH, Black, Hispanic, and SA; SEA is significantly different from Black, Hispanic, and SA; SA is significantly different from WNH, Black, Hispanic, EA, and SEA.

Clinical Stage

Table 5 depicts clinical stage group by ethnic group cross-tabulation employing our Asian American composite group, whereas Table 6 utilizes the 3 Asian American ethnic subgroups as the basis for comparison. In all clinical stages except clinical stage IV, Asian Americans were underrepresented relative to White non-Hispanics and additionally underrepresented to Blacks and Hispanics in clinical stages IIB and IIIA. In clinical stage IV, Asian Americans were overrepresented relative to White non-Hispanics, Blacks, and Hispanics (Asian Americans 51.0%, White non-Hispanics 42.2%, Blacks 47.5%, and Hispanics 47.0%) (Table 5). Further nuances are discovered when examining Asian American ethnic groups in Table 6. For example, Southeast Asians were underrepresented relative to East Asians and South Asians in clinical stage IA while being overrepresented relative to East Asians and South Asians in clinical stage IV (clinical stage IA: East Asians 19.8%, Southeast Asians 13.3%, and South Asians 17.4%; clinical stage IV: East Asians 47.2%, Southeast Asians 57.4%, and South Asians 52.2%) (Table 6).Table 5 Clinical stage group × ethnic group cross-tabulation (combined Asian American group)

Clinical stage∗	Ethnic group	Total	
WNH	Black	Hispanic	AsAm	
IA						
 Count	136,336a	14,518b	3008c	1550c	155,412	
 % within ethnic group	19.5	14.3	17.6	17.6	18.8	
 Adjusted residual	38.7	−39.8	−4.0	−3.0		
IB						
 Count	58,555a	6842b	1231b	586b	67,214	
 % within ethnic group	8.4	6.7	7.2	6.7	8.1	
 Adjusted residual	19.3	−17.7	−4.5	−5.1		
IIA						
 Count	23,066a	2889b	511a, b	243b	26,709	
 % within ethnic group	3.3	2.8	3.0	2.8	3.2	
 Adjusted residual	8.4	−7.7	−1.8	−2.5		
IIB						
 Count	28,720a	3904b	608b	235c	33,467	
 % within ethnic group	4.1	3.8	3.6	2.7	4.1	
 Adjusted residual	6.6	−3.8	−3.3	−6.6		
IIIA						
 Count	87,137a	13,354b	2034a	885c	103,410	
 % within ethnic group	12.5	13.1	11.9	10.0	12.5	
 Adjusted residual	−2.5	6.1	−2.4	−7.1		
IIIB						
 Count	69,659a	11,940b	1655a	821a	84,065	
 % within ethnic group	10.0	11.7	9.7	9.3	10.2	
 Adjusted residual	−14.2	17.4	−2.1	−2.7		
IV						
 Count	294,300a	48,366b	8016b	4489c	355,171	
 % within ethnic group	42.2	47.5	47.0	51.0	43.0	
 Adjusted residual	−36.5	30.8	10.5	15.1		
WNH, White non-Hispanic; AsAm, Asian American (East Asian: Chinese, Japanese, Korean; Southeast Asian: Filipino, Vietnamese, Laotian, Hmong, Thai; South Asian: Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the Clinical Stage IIA row: WNH is significantly different from Black and AsAm; Black is only significantly different from WNH; Hispanic is not significantly different from any group; AsAm is only significantly different from WNH.

Table 6 Clinical stage group × ethnic group cross-tabulation

Clinical stage∗	Ethnic group	Total	
WNH	Black	Hispanic	EA	SEA	SA	
IA								
 Count	136,336a	14,518b	3008c	931a	315b	304a,c	155,412	
 % within ethnic group	19.5	14.3	17.6	19.8	13.3	17.4	18.8	
 Adjusted residual	38.7	−39.8	−4.0	1.7	−6.9	−1.5		
IB								
 Count	58,555a	6842b,c,d,e	1231d,e	338c,e	125b	123a,b,c,d,e	67,214	
 % within ethnic group	8.4	6.7	7.2	7.2	5.3	7.0	8.1	
 Adjusted residual	19.3	−17.7	−4.5	−2.4	−5.1	−1.7		
IIA								
 Count	23,066a	2889b	511a,b	128a,b	66a,b	49a,b	26,709	
 % within ethnic group	3.3	2.8	3.0	2.7	2.8	2.8	3.2	
 Adjusted residual	8.4	−7.7	−1.8	−2.0	−1.2	−1.0		
IIB								
 Count	28,720a	3904b	608b,c	132c	58c	45b,c	33,467	
 % within ethnic group	4.1	3.8	3.6	2.8	2.5	2.6	4.1	
 Adjusted residual	6.6	−3.8	−3.3	−4.3	−4.0	−3.1		
IIIA								
 Count	87,137a	13,354b	2034a	485c	219c	181a,c	103,410	
 % within ethnic group	12.5	13.1	11.9	10.3	9.3	10.4	12.5	
 Adjusted Residual	−2.5	6.1	−2.4	−4.6	−4.8	−2.7		
IIIB								
 Count	69,659a	11,940b	1655a,c	465a,c	223a,c	133c	84,065	
 % within Ethnic Group	10.0	11.7	9.7	9.9	9.4	7.6	10.2	
 Adjusted Residual	−14.2	17.4	−2.1	−0.6	−1.2	−3.6		
IV								
 Count	294,300a	48,366b	8016b	2218b	1358c	913d	355,171	
 % within ethnic group	42.2	47.5	47.0	47.2	57.4	52.2	43.0	
 Adjusted residual	−36.5	30.8	10.5	5.8	14.2	7.8		
WNH, White non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the Clinical Stage IB row: WNH is significantly different from Black, Hispanic, EA, and SEA; Black is only significantly different from WNH; Hispanic is significantly different from WNH and SEA; EA is significantly different from WNH and SEA; SEA is significantly different from WNH, Hispanic, and EA; SA is not significantly different from any group.

Table E1 provides a visual summary of over- and underrepresentation of the Asian American composite group as well as subgroups. Asian Americans were underrepresented relative to White non-Hispanics in all stages except clinical stage IIIB (no difference) and IV (overrepresented), reflecting the trend seen in the subgroups. The converse was largely true for clinical stage IV, but East Asians were underrepresented relative to Southeast Asians and South Asians in clinical stage IV. Asian Americans were underrepresented relative to White non-Hispanics in clinical stage IIA, yet no disaggregated subgroup in isolation was identified to be over- or underrepresented.

Survival

Table 7 depicts a cross-tabulation of vital sign status alive at 30 days or date of last contact by clinical stage by ethnic group employing our Asian American composite group, whereas Table 8 utilizes the 3 Asian American ethnic subgroups as the basis for comparison. In all clinical stages, Asian American survival was overrepresented relative to White non-Hispanics, Blacks, and Hispanics except in clinical stage IIA and IIB where Asian American survival was overrepresented relative to White non-Hispanics and Blacks (Table 7). However, when comparing the subgroups in Table 8, it was found that Southeast Asian and South Asian survival were not significantly different than White non-Hispanics or Blacks in clinical stage IIA and IIB.Table 7 Vital status at date of last contact × ethnic group × clinical stage group cross-tabulation (combined Asian American group)

Clinical stage∗	Vital status at date of last contact	Ethnic group	Total	
WNH	Black	Hispanic	AsAm	
IA	Alive	Count	62,673a	6834b	1743c	1044d	72,294	
		% within ethnic group	52.6	54.0	68.7	79.2	53.3	
		Adjusted residual	−13.7	1.7	15.7	18.9		
IB	Alive	Count	19,814a	2401a	582b	320c	23,117	
		% within ethnic group	37.0	38.4	53.1	61.2	37.6	
		Adjusted residual	−8.5	1.4	10.7	11.2		
IIA	Alive	Count	7851a	1019a	234b	114b	9218	
		% within ethnic group	38.6	40.2	53.1	57.0	39.2	
		Adjusted residual	−4.8	1.1	6.0	5.2		
IIB	Alive	Count	7149a	1022a	242b	97b	8510	
		% within ethnic group	27.3	28.8	42.5	46.2	27.9	
		Adjusted residual	−5.6	1.3	7.8	5.9		
IIIA	Alive	Count	16,737a	2864b	604c	315d	20,520	
		% within ethnic group	21.2	23.8	33.2	39.7	21.9	
		Adjusted residual	−12.6	5.5	11.8	12.1		
IIIB	Alive	Count	7505a	1553b	339c	214d	9611	
		% within ethnic group	11.4	13.8	21.6	28.2	12.1	
		Adjusted residual	−13.3	6.0	11.6	13.7		
IV	Alive	Count	19,560a	3884b	1367c	937d	25,748	
		% within ethnic group	7.3	8.8	19.0	23.6	8.0	
		Adjusted residual	−31.2	7.3	35.1	36.7		
WNH, White non-Hispanic; AsAm, Asian American (East Asian: Chinese, Japanese, Korean; Southeast Asian: Filipino, Vietnamese, Laotian, Hmong, Thai; South Asian: Indian Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the Clinical Stage IB row: WNH is significantly different from Hispanic and AsAm; Black is significantly different from Hispanic and AsAm; Hispanic is significiantly different from WNH, Black, and AsAm; AsAm is significantly different from WNH, Black, and Hispanic.

Table 8 Vital status at date of last contact × ethnic group × clinical stage group cross-tabulation

Clinical stage∗	Vital status at date of last contact	WNH	Ethnic group	Total	
Black	Hispanic	EA	SEA	SA	
IA	Alive	Count	62,673a	6834b	1743c	639d	204c,d	201c,d	72,294	
		% within ethnic group	52.6	54.0	68.7	81.2	76.1	76.1	53.3	
		Adjusted residual	−13.7	1.7	15.7	15.7	7.5	7.5		
IB	Alive	Count	19,814a	2401a,b	582c	199d	63c,d	58b,c,d	23,117	
		% within ethnic group	37.0	38.4	53.1	65.7	58.3	51.8	37.6	
		Adjusted residual	−8.5	1.4	10.7	10.1	4.4	3.1		
IIA	Alive	Count	7851a	1019a	234b	68b	25a,b	21a,b	9218	
		% within ethnic group	38.6	40.2	53.1	63.6	44.6	56.8	39.2	
		Adjusted residual	−4.8	1.1	6.0	5.2	0.8	2.2		
IIB	Alive	Count	7149a	1022a	242b	62b	19a, b	16a,b	8510	
		% within ethnic group	27.3	28.8	42.5	52.1	37.3	40.0	27.9	
		Adjusted residual	−5.6	1.3	7.8	5.9	1.5	1.7		
IIIA	Alive	Count	16,737a	2864b	604c	177c	76c	62c	20,520	
		% within ethnic group	21.2	23.8	33.2	40.3	39.8	37.8	21.9	
		Adjusted residual	−12.6	5.5	11.8	9.3	6.0	4.9		
IIIB	Alive	Count	7505a	1553b	339c	132d	42b,c,d	40d	9611	
		% within ethnic group	11.4	13.8	21.6	30.6	19.9	34.2	12.1	
		Adjusted residual	−13.3	6.0	11.6	11.8	3.5	7.3		
IV	Alive	Count	19,560a	3884b	1367c	503d	219c	215d	25,748	
		% within ethnic group	7.3	8.8	19.0	26.0	17.9	26.6	8.0	
		Adjusted residual	−31.2	7.3	35.1	29.4	12.9	19.6		
WNH, White non-Hispanic; EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified).

∗ Within a given table row, cells annotated with the same subscript letter indicate that the column proportions do not differ significantly from each other at the .05 level. Similarly, within a given table row, cells annotated with different subscript letters indicate that the column proportions do differ significantly from each other at the .05 level. For example, for the Clinical Stage IB row: WNH is significantly different form Hispanic, EA, SEA, and SA; Black is significantly different from Hispanic, EA, and SEA; Hispanic is significantly different from WNH, Black, and EA; EA is significantly different from WNH, Black, and Hispanic; SEA is significantly different from WNH and Black; SA is only significantly different from WNH.

Table E2 provides a visual summary of over- and underrepresentation of the Asian American composite group as well as subgroups regarding survival. Asian American survival was consistently overrepresented in all stages relative to White non-Hispanics and Blacks, but within the subgroups, several nuances became apparent. For example, all subgroups’ survival was equivalent in clinical stage IA or IIIB, but Southeast Asian survival in clinical stage IV was significantly lower than in East Asians and South Asians. Such differences were obscured when the comparison group is broad (ie, the pooled Asian American group).

Discussion

The NCDB collects data on patients with any cancer diagnosis from more than 1500 cancer-accredited facilities. Despite this large source of data, analysis of Asian Americans with lung cancer has been difficult in the past when compared with other major US ethnic groups due to low counts within ethnic subgroups. By applying columnar z-score analysis with adjusted standardized residuals, we attempted to overcome this issue, which to our knowledge was the first time this analytical approach has been applied to this type of study. We employed the terms underrepresented and overrepresented when cross-tabulation percentages were accompanied by a z-score proportion difference of P < .05 (Figure 1). In addition, our analysis went a step further than prior Asian American lung cancer studies by looking at clinical stage at diagnosis and survival in East Asian, Southeast Asian, and South Asian ethnic subgroups (see Figure 2 for a graphical abstract of the study). The results were consistent with other studies that examined lung cancer stage and survival in Asian Americans,17, 18, 19 except that we also examined Asian American subgroups.Figure 1 Comparison of advanced stage lung cancer and survival between ethnic groups. East Asian, Southeast Asian, and South Asian individuals are more likely to be diagnosed at advanced stages of lung cancer yet demonstrate higher survival compared with White Non-Hispanic, Black, and/or Hispanic with differences between subgroups.

Figure 2 Graphical abstract of the study. Graphic outlines Asian American health disparities in lung cancer with a novel analytic approach.

Among the biggest obstacles in studying Asian American health disparities is the issue of aggregate data. The generalization of Asian Americans as 1 group can mask the unique disease risk of different subgroups and assume similar disease risk as the White non-Hispanic majority.10 Tables E1 and E2 demonstrate the underlying issues and limitation of aggregating Asian Americans as 1 group. Asian Americans demonstrated significantly higher proportion of clinical stage IV lung cancer diagnosis compared with White non-Hispanics, Blacks, and Hispanics. However, when examining the Asian American subgroups, clinical stage IV diagnosis was significantly higher in Southeast Asians than East Asians and South Asians and significantly higher in South Asians than East Asians (Table E1). With disaggregated data, physicians may assume that all Asian American patients have a similar degree of being diagnosed at advanced stages of lung cancer, which can in turn negatively influence the care of Asian American patients. Additionally, Asian American survival was significantly higher in all stages of lung cancer compared with White non-Hispanics and Blacks, but Southeast Asians’ survival in clinical stage IV was significantly lower than East Asians and South Asians (Table E2). Without the knowledge of this difference between subgroups, Southeast Asian patients may be treated similarly as East Asian and South Asian patients, which could potentially result in unexpected outcomes.

The interplay of factors associated with the complex findings of advanced diagnosis yet greater survival with differences between subgroups includes cultural beliefs, environmental exposures, genetics, immigration, diet, and barriers to care, just to name a few, which may begin to unlock clues to understanding Asian American health disparities.

Because lung cancer is the leading cause of death for Asian Americans, smoking is a significant risk factor and the number-1 underlying cause of lung cancer worldwide.20 Smoking prevalence ranged from 7.6% among Chinese and Indians to 20.0% among Koreans.21 When smoking cessation in Asian Americans was examined, only 34.2% of Asian Americans who smoke were advised to quit compared with 60.2% White non-Hispanics and 42.2% Hispanics who smoke. Additionally, Asian Americans who smoke received less counseling and/or medication for smoking cessation compared with White non-Hispanics who smoke (20.5% vs 34.3%).22 The discrepancy in smoking prevalence and cessation in Asian Americans could be a direct result of aggregating Asian Americans as 1 group, which demonstrates the importance of examining Asian American subgroups. Given the high prevalence of smoking in Asian cultures, environmental tobacco smoke is a factor to consider as tobacco use by spouse was associated with a 30% excess risk of lung cancer.23 Other environmental risk factors include air pollution and cooking oil fumes.24,25 Indoor air pollutants from Chinese-style cooking has been linked to increased risk of lung cancer.25 Lastly, dietary factors may be protective, such as high consumption of vegetables, fruits, and soy products in Asian cultures.26

The potential of lung cancer screening has also poorly penetrated Asian Americans because they had a lower rate of cancer screening than White non-Hispanics and was the only racial group where screening disparity was not well explained by socioeconomic factors.27 This is possibly due to lack of knowledge of the health care system, language barriers, and culture. Many Asian cultures believe illness as one's fate, so preventative health care is not a widely embraced practice.11 Another contributor to potentially explain delayed diagnosis could stem from the labeling of Asian Americans as the so-called model minority. By accepting this stereotype in attempt to assimilate, Asian Americans may be reluctant to disclose and advocate for their health.28 These stereotypes may lead to the perception that Asian Americans do not need help. In fact, Asian Americans were the least likely among all racial groups to have seen a physician in the past 12 months.12

Further complicating these issues is that not every aforementioned factor can or should be generalized toward all Asian Americans because some Asian American ethnic groups immigrated to the United States more recently than others. Many Japanese immigrants arrived in the 19th century with 63% of Japanese Americans having been in the United States for more than 10 years.13 Depending on the timing of immigration, Asian Americans may carry different risk factors and retain different amounts of cultural beliefs. Because of the heterogeneity of Asian Americans, they should not be regarded or treated similarly. Thus, culturally competent care is crucial for Asian American patients.

Studies have shown improved survival in Asian Americans with NSCLC compared with White non-Hispanics, which is consistent with our findings.29, 30, 31 This perhaps is due to the role of epidermal growth factor receptor (EGFR) mutations and polymorphism.29,32, 33, 34 Persons who do not smoke have more EGFR mutations and improved survival.29 It has been well established that there is a predominance of Asian American patients who do not smoke with NSCLC, especially Asian American women.29,35 A high level of EGFR protein was also linked to poorer outcome in NSCLC.32 Compared with other ethnic groups, Asian Americans had lower EGFR expression due to polymorphism.33 Along with the higher frequency of amplification in a closely related HER2 gene in East Asians, one can speculate the influence of genetic diversity on clinical outcomes such as survival.34

Limitations

There are several limitations to this study. First, a total of 11 out of 18 Asian ethnicities were divided based on geographic origins into 3 Asian American subgroups in this study due to limited number of patients in each Asian ethnicity. Our creation of cohorts resulted in the exclusion of 5 mixed Asian ethnicities within the NCDB. This was done mostly due to American cultural influence within the area (eg, American Samoa and Philippines). Secondly, histological subtypes of lung cancer were not explored in this study due to the uncertainty of its clinical significance given the relatively small number of patients in each Asian American subgroup. With consistent and comprehensive identity collection nationwide, further investigation is possible and warranted to aid in understanding of these results. Lastly, because screening and smoking status are crucial components of lung cancer risk reduction, collaboration with medical centers in densely Asian American-populated regions may allow the exploration of intersectionality between risk reduction and diagnosis, which we were unable to do with the NCDB data. This may allow future studies to distinguish more specific characteristics within the Asian American subpopulations.

Conclusions

This study demonstrated the importance of disaggregating Asian Americans as 1 group while also comparing them to the dominant US populations allowing us to both quantify and analyze these differences. Although Asian Americans were more likely to be diagnosed at advanced stages of lung cancer, they demonstrated higher survival compared with White non-Hispanics, Blacks, and/or Hispanics, whereas differences between Asian American subgroups were also apparent. Disaggregating the Asian American umbrella group can reveal unique characteristics of individual subgroups otherwise missed. Further investigation to explore differences between the Asian American subgroups is necessary to make definitive conclusions. Understanding the interplaying factors comprising Asian American health disparities is recommended for culturally competent care, community-based interventions, and personalized medicine.

Webcast

You can watch a Webcast of this AATS meeting presentation by going to: https://www.aats.org/resources/identifying-asian-american-lung-cancer-disparities-a-novel-analytic-approach.

Conflict of Interest Statement

The authors reported no conflicts of interest.

The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.

Appendix E1

Table E1 Summary of under- and overrepresentation of Asian subgroups in lung cancer clinical stages

Clinical stage	Asian group	
Comparison group	EA	SEA	SA	AsAm	
WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	
Stage IA	–	–	–		–	–	↓	–	↓	↓		↓	–	–	–	–	–		↓	↑	–	
Stage IB	↓	–	–		↑	–	↓	–	↓	↓		–	–	–	–	–	–		↓	–	–	
Stage IIA	–	–	–		–	–	–	–	–	–		–	–	–	–	–	–		↓	–	–	
Stage IIB	↓	↓	–		–	–	↓	↓	–	–		–	↓	–	–	–	–		↓	↓	↓	
Stage IIIA	↓	↓	↓		–	–	↓	↓	↓	–		–	–	↓	–	–	–		↓	↓	↓	
Stage IIIB	–	–	–		–	–	–	–	–	–		–	↓	↓	–	–	–		–	↓	–	
Stage IV	↑	–	–		↓	↓	↑	↑	↑	↑		↑	↑	↑	↑	↑	↓		↑	↑	↑	
Upward ↑ and downward ↓ facing arrows indicate over- and underrepresentation relative to comparison groups at P < .05, respectively. A dash indicates no significant difference in representation. EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified); AsAm, Asian American (East Asian, Southeast Asian, South Asian); WNH, White non-Hispanic.

Table E2 Summary of under- and overrepresentation of Asian subgroups in vital status = alive at date of last contact

Vital status at date of last contact = Alive	Asian group	
	EA	SEA	SA	AsAm	
Comparison group	WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	EA	SEA	SA	WNH	Black	Hispanic	
Stage IA	↑	↑	↑		–	–	↑	↑	–	–		–	↑	↑	–	–	–		↑	↑	↑	
Stage IB	↑	↑	↑		–	–	↑	↑	–	–		–	↑	–	–	–	–		↑	↑	↑	
Stage IIA	↑	↑	–		–	–	–	–	–	–		–	–	–	–	–	–		↑	↑	–	
Stage IIB	↑	↑	–		–	–	–	–	–	–		–	–	–	–	–	–		↑	↑	–	
Stage IIIA	↑	↑	–		–	–	↑	↑	–	–		–	↑	↑	–	–	–		↑	↑	↑	
Stage IIIB	↑	↑	↑		–	–	↑	–	–	–		–	↑	↑	↑	–	–		↑	↑	↑	
Stage IV	↑	↑	↑		↑	-	↑	↑	-	↓		↓	↑	↑	↑	-	↑		↑	↑	↑	
Upward ↑ and downward ↓ facing arrows indicate over- and underrepresentation relative to comparison groups at P < .05, respectively. A dash indicates no significant difference in representation. EA, East Asian (Chinese, Japanese, Korean); SEA, Southeast Asian (Filipino, Vietnamese, Laotian, Hmong, Thai); SA, South Asian (Indian, Pakistani, Asian Indian/Pakistani not otherwise specified); AsAm, Asian American (East Asian, Southeast Asian, South Asian); WNH, White non-Hispanic.

IRB Number: STUDY0000028; approval date: March 3, 2019.

The Medstar Health Research Institutional Review Board (IRB) approved the study protocol and publication of data. The informed written consent of patient(s) for the publication of the study data was not required because the National Cancer Database provides de-identified data.
==== Refs
References

1 About the topic of race. United States Census Bureau https://www.census.gov/topics/population/race/about.html
2 Monte L.M. Shin H.B. 20.6 Million people in the U.S. identify as Asian, Native Hawaiian or Pacific Islander United States Census Bureau. 2022. 2022 https://www.census.gov/library/stories/2002/05/aanhpi-population-diverse-geographically-dispersed.html
3 Ogunwole S.U. Rabe M.A. Roberts A.W. Caplan Z. Population under age 18 declined last decade United States Census Bureau. 2021. 2021 https://www.census.gov/library/stories/2021/08/united-states-adult-population-grew-faster-than-nations-total-population-from-2010-to-2020.html
4 Race and ethnicity in the United States: 2010 census and 2020 census United States Census Bureau. 2021 2021 https://www.census.gov/library/visualizations/interactive/race-and-ethnicity-in-the-united-state-2010-and-2020-census.html
5 Ortman J. Guarneri C. United states population projections: 2000 to 2050 United States Census Bureau. 2009. 2009 https://www.census.gov/content/dam/Census/library/working-papers/2009/demo/us-pop-proj-2000-2050/analytical-document09.pdf
6 Heron M. Deaths: leading causes for 2019 Natl Vital Stat Rep 70 2021 1 114
7 Gomez S.L. Noone A. Lichtensztajn D.Y. Cancer incidence trends among Asian American populations in the United States, 1990-2008 J Natl Cancer Inst 105 2013 1096 1110 23878350
8 Coultas D.B. Gong H. Grad R. Respiratory diseases in minorities of the United States Am J Respir Crit Care Med 149 3 Pt 2 1994 S93 S131 8118656
9 Lee R.J. Madan R.A. Kim J. Posadas E.M. Yu E.Y. Disparities in cancer care and the Asian American population Oncologist 26 2021 453 460 33683795
10 Grineske S.E. Collins T.W. Morales D.X. Asian Americans and disproportionate exposure to carcinogenic hazardous air pollutants: a national study Soc Sci Med 185 2017 71 80 28554161
11 Sin M. Taylor V. Tu S. Issues in lung cancer screening among Asian American Immigrants J Immigr Minor Health 18 2016 495 496 26048357
12 Chen M.S. Cancer health disparities among Asian Americans: what we do and what we need to do Cancer 104 Suppl 12 2005 2895 2902 16270313
13 Farley R. Haaga J. The American People: Census 2000 2005 Russell Sage Foundation 424 432
14 Garson G.D. Crosstabulation (Statistical Associates Blue Book Series 13) 2013 Statistical Associates Publisher 3 32
15 Bland J.M. Altman D.G. Multiple significance tests: the Bonferroni method BMJ 310 6973 1995 170 7833759
16 Agresti A. An Introduction to Categorical Data Analysis 2007 Jonn Wiley & Sons 34 40
17 Finlay G.A. Joseph B. Rodrigues C.R. Griffith J. White A.C. Advanced presentation of lung cancer in asian immigrants: a case-control study Chest 122 2002 1938 1943 12475830
18 Underwood J.M. Townsend J.S. Tai E. Racial and regional disparities in lung cancer incidence Cancer 118 2012 1910 1918 21918961
19 Dalwadi S.M. Lewis G.D. Bernicker E.H. Butler E.B. Teh B.S. Farach A.M. Disparities in the treatment and outcome of stage I non–small-cell lung cancer in the 21st century Clin Lung Cancer 20 2019 194 200 30655194
20 Tobacco use among U.S. Racial/Ethnic Minority Groups--African Americans, American Indians and Alaska Natives, Asian Americans and Pacific Islander, and Hispanics: A Report of the Surgeon General 1998 Center for Disease Control and Prevention 18-27, 56-65
21 Martell B.N. Garrett B.E. Caraballo R.S. Disparities in adult cigarette smoking – United States, 2002-2005 and 2010-2013 MMWR Morbid Mortal Weekly Rep 65 2016 753 758
22 Lei F. Zheng Y. Li C. Health disparities across lung cancer continuum among Asian Americans: a systematic review J Immigr Minor Health 24 2022 526 545 34156594
23 Fontham E.T.H. Correa P. Reynolds P. Environmental tobacco smoke and lung cancer in nonsmoking women: a multicenter study JAMA 271 22 1994 1752 1759 8196118
24 Pope C.A. III Burnett R.T. Thun M.J. Lung cancer, cardiopulmonary mortality, and long-term exposure to fine particulate air pollution JAMA 287 9 2002 1132 1141 11879110
25 Zhong L. Goldberg M.S. Gao Y. Jin F. Lung cancer and indoor air pollution arising from Chinese-style cooking among nonsmoking women living in Shanghai, China Epidemiology 10 5 1999 488 494 10468420
26 Seow A. Koh W. Wang R. Lee H. Yu M.C. Reproductive variables, soy intake, and lung cancer risk among nonsmoking women in the Singapore Chinese Health Study Cancer Epidemiol Biomarkers Prev 18 3 2009 821 827 19240237
27 Jun J. Cancer/health communication and breast/cervical cancer screening among Asian Americans and five Asian ethnic groups Ethn Health 25 2020 960 981 29792075
28 Ibaraki A.Y. Hall N.G.C. Sabin J.A. Asian American cancer disparities: the potential effects of model minority health stereotypes Asian Am J Psychol 5 2014 75 81
29 Ou S.I. Ziogas A. Zell J.A. Asian ethnicity is a favorable prognostic factor for overall survival in non-small cell lung cancer (NSCLC) and is independent of smoking status J Thorac Oncol 4 2009 1083 1093 19652625
30 Florescu M. Hasan B. Seymour L. Ding K. Shepherd F.A. A clinical prognostic index for patients treated with erlotinib in National Cancer Institute of Canada Clinical Trials Group study BR.21 J Thorac Oncol 3 2008 590 598 18520796
31 Pirker R. Szczesna A. von Pawel J. FLEX: a randomized, multicenter, phase III study of cetuximab in combination with cisplatin/vinorelbine (CV) versus CV alone in the first-line treatment of patients with advanced non-small cell lung cancer (NSCLC) J Clin Oncol 26 2008 3
32 Meert A. Martin B. Delmotte P. The role of EGF-R expression on patient survival in lung cancer: a systematic review with meta-analysis Eur Respir J 20 2002 975 981 12412692
33 Nomura M. Shigematsu H. Li L. Polymorphisms, mutations, and amplification of the EGFR gene in non-small cell lung cancers PLoS Med 4 2007 e125 17455987
34 Calvo E. Baselga J. Ethnic differences in response to epidermal growth factor receptor tyrosine kinase inhibitors J Clin Oncol 24 2006 2158 2163 16682734
35 Subramanian J. Govindan R. Lung cancer in never smokers: a review J Clin Oncol 25 2007 561 570 17290066
