Suitability of Middle Eastern and North African immigrant minimal reporting category for Arab immigrant health research in the National Health Interview Survey

Abstract

Objective:

To test whether the Middle East or North Africa (MENA) category reflects the Arab immigrant experience, we compare health estimates for MENA immigrants with previously published literature that included only Arab immigrants.

Methods:

Prior methodology used to examine Arab immigrant health between 2000–2014 was replicated to examine the utility of the MENA identifier. Region of birth among Whites represented the MENA immigrant category. Dependent variables included chronic conditions and health behaviors. Logistic regression was conducted to determine the odds of each outcome for foreign-born MENA Americans compared to US-born Whites. Then, adjusted 95% confidence intervals representing the more inclusive MENA immigrant categorization were compared to previous studies among Arab immigrants.

Results:

Findings did not differ whether the MENA or Arab definition was used.

Conclusions:

The MENA category reflects the Arab immigrant experience, even though it includes a wider set of origins, some of which are not Arab. Including a MENA identifier on future data collections will both represent Arab Americans, as well as identify this population as distinct from Whites to better represent and track health disparities.

INTRODUCTION

The first United States (US) Census was administered in 1790, with the following categories: free Whites; all other free persons; and slaves. Since then, the racial and ethnic landscape of the US has changed dramatically. Regardless, the US Census recognizes one ethnicity - Hispanic/Latino - and five races - American Indian/Alaska Native; Asian; Black/African American; Native Hawaiian/Other Pacific Islander and White.¹ White is defined as a “person having origins in any of the original peoples of Europe, the Middle East, or North Africa.” Conflating Middle East or North Africa (MENA) with the White category masks the presence and influence of MENA in all aspects of society, prevents fair allocation of resources, and diminishes the ability to identify health disparities.²

Recently, the announcement was made to formally review and revise the Office of Management and Budget’s (OMB) Statistical Policy Directive No. 15.³ One of the many reasons this revision was initiated was due to results of the 2015 National Content Test that indicated a separate MENA response category was necessary to better represent the MENA populations in the US.^2,4 Until the 1997 directive is revised and approved, MENA populations will continue to be undercounted or invisible by the census and the other 130 federally mandated surveys. Of these 130 surveys, there are two nationally representative surveys that can measure characteristics among MENA populations: 1) American Community Survey (ACS) and 2) National Health Interview Survey (NHIS). The ACS includes questions on ancestry and place of birth that can be used to disaggregate MENA from other non-Hispanic Whites. However, the utility of data from the ACS for health research is limited because it only includes health-related questions on disability.^5,6 the NHIS is the only national survey that allows for MENA health to be evaluated on a national scale.⁷ From 2000–2018, the NHIS included questions on place of birth (not ancestry) that can be combined with questions on race and ethnicity to disaggregate MENA individuals from other non-Hispanic Whites.⁷ Two limitations of the NHIS are that 1) the health patterns of US-born MENA individuals cannot be established; and 2) due to small sample sizes, multiple years must be combined to increase precision of health patterns among foreign-born MENA individuals. Details on how MENA nationalities and diverse ethnic and religious groups have been measured within the context of the US Census guidelines using these national surveys is provided in Table 1.

Table 1:

Details on how health disparities can be measured among MENA immigrants in alignment with US Census Guidance.

	NHIS Public-Use Data Middle East region of birth variable	NHIS Public-Use Data Africa region of birth variable	NHIS Restricted Data (15 countries in Middle East and Arab League of Nations)	Arab League of Nations (22 countries)	US Census guidance document countries/ethnicities
Nationalities
Aden	X
Africa		X
Algeria		X		X	X
Algiers		X
Angola		X
Anojouan		X
Arabia	X
Armenia	X
Bahrain	X		X	X	X
Benin		X
Botswana		X
British East Africa		X
Burkina Faso		X
Burundi		X
Cameroon		X
Central Africa		X
Central African Republic		X
Chad		X
Comoros Islands		X		X	X
Congo		X
Cote D’ivorie		X
Cyprus	X
Djibouti		X		X	X
Demo Rep of Congo		X
Eastern Africa		X
Egypt		X		X	X
Equatorial Guinea		X
Eritrea		X
Ethiopia		X
Gabon		X
Gambia		X
Ghana		X
Guinea		X
Guinea-Bissau		X
Iran	X				X
Iraq	X		X	X	X
Israel	X				X
Jordan	X		X	X	X
Kenya		X
Kuwait	X		X	X	X
Lebanon	X		X	X	X
Lesotho		X
Liberia		X
Libya		X		X	X
Madagascar		X
Madeira Islands		X
Mauritania		X		X	X
Mauritius		X
“Middle East”	X
Morocco		X		X	X
Mozambique		X
Namibia		X
Niger		X
Nigeria		X
North Africa		X
Oman	X		X	X	X
Palestine	X		X	X	X
Arab Palestine	X		X
West Bank	X		X
Gaza Strip	X		X
Persia	X
Qatar	X		X	X	X
Rwanda		X
Sao Tome Island		X
Sao Tome & Principe		X
Saudi Arabia	X		X	X	X
Senegal		X
Seychelles		X
Sierra Leone		X
Somalia		X		X	X
South Africa		X
Republic of South Africa		X
Union of South Africa		X
Southern Africa		X
Sudan		X		X	X
Syria	X		X	X	X
Tanzania		X
Togo		X
Togoland		X
Turkey	X				X
Tunisia				X	X
United Arab Emirates	X		X	X	X
Uganda		X
West Africa		X
Yemen	X		X	X	X
Zambia		X
Zanzibar		X
Zimbabwe		X
Diverse Ethnic and Religious Groups
Assyrians					X
Chaldeans					X
Copts					X
Druze					X
Kurds					X
Shi’a					X
Sunni Muslims					X

Despite these limitations, the NHIS has been widely used to explore health disparities among Middle Eastern immigrants, using data which have included individuals from Arab and non-Arab countries. Due to the small sample sizes ranging from 56 to 140 sample adults per year, multiple years must be combined to increase precision of the estimates when compared to other non-Hispanic Whites. Studies have explored serious psychological distress,⁸ cognitive impairment,⁹ self-rated health,^10–12 hypertension,^11,13 heart conditions,¹² and diabetes.¹³ Results indicated the health of Middle Eastern immigrants differed from those of other Whites. For example, Dallo and colleagues found that Middle Eastern immigrants were more likely to report serious psychological distress compared to US-born non-Hispanic Whites.⁸ More recently, studies have included non-Hispanic White African immigrants with the Middle Eastern immigrant group to examine health outcomes among older adults while also being more inclusive of North Africans.^14,15 Sensitivity analyses from these studies revealed results were similar to those when only Middle Eastern immigrants were included.

Studies Used for Comparison in Current Analysis

Studies of Middle Eastern immigrants using NHIS data, with and without African immigrants, have provided important discoveries on the health of this group.^16–19 However, they do not directly measure health among Arab American immigrants (from hereafter referred to as Arab immigrants). To define Arab immigrants, researchers used restricted country of birth data from NHIS to create a variable that included immigrants born in 15 countries/regions located in the Middle East that are also part of the Arab League of Nations (Bahrain, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine -- Arab Palestine, West Bank, Gaza Strip, Qatar, Saudi Arabia, Syria, United Arab Emirates, Yemen) and compared their health to US-born Whites.^16–19 The difference between MENA and Arab is that MENA usually includes 22 countries belonging to the Arab League of Nations (hereafter referred to Arab Nations) plus Turkey, Iran, Israel.^4,20 Studies using NHIS data have demonstrated that Arab immigrants had lower odds of heart disease, asthma, and cancer, but no difference in diabetes or obesity than US-born Whites.^16–19 Arab immigrant women were less likely to receive flu or pneumonia vaccines, pap smears, clinical breast exams, or report current smoking than US-born White women.¹⁶ Arab immigrant men were less likely to report receiving flu or pneumonia vaccines than US-born White men.¹⁷

It is yet to be determined whether using a MENA classification can truly represent the health needs of Arab immigrants. To meet this need, the objective of this study is to assess if MENA immigrant health reflects the same patterns found in previous research focusing on as Arab immigrant health. More specifically, we will calculate 1) chronic disease, 2) women’s preventive health behaviors, 3) men’s preventive health behaviors, and 4) cigarette smoking outcomes estimates among MENA immigrants and compare findings to previous studies using Arab immigrants following the classification method for individuals born in countries located in the Middle East and are part of the Arab Nations.

METHODS

Data Source

We used multiple years of NHIS data in alignment with each former study methodology to compare our findings with four previous research studies.^16–19 The NHIS is an annual survey conducted in the US since 1957 designed to collect health information from a large representative sample of adults and children using face-to-face interviews.^7,21 The questions range from demographic, socioeconomic, healthcare access and utilization, disease prevalence, and preventive health behaviors, among many others. The complete survey details are available on the survey website.²¹ For chronic disease and preventive health behaviors (women and men), our comparison studies used 2000–2011 NHIS data.^16–18 For current smoking, our comparison study used 2000–2014 NHIS data.¹⁹ Samples were limited to adults ages 18 and older. Sample sizes for US-born non-Hispanic Whites were benchmarked with the previous studies to ensure internal validity.

Participants

2000–2011 Sample Participants

From 2000–2011, the sample included 205,763 US-born non-Hispanic Whites and 1,478 MENA immigrants to examine chronic disease prevalence. The comparison study included 463 Arab immigrants.¹⁸ To examine women’s preventive health behaviors, the sample included 113,406 US-born non-Hispanic White women and 676 MENA immigrant women. In the comparison study, there were 205 Arab immigrant women.¹⁶ To examine men’s preventive health behaviors, the sample included 88,413 US-born non-Hispanic White men and 750 MENA immigrant men. In the comparison study, there were 238 Arab immigrant men.¹⁷

2000–2014 Sample Participants

From 2000–2014, the study focused on cigarette smoking included 266,813 US-born non-Hispanic White adults and 1,942 MENA immigrants. In the comparison study, there were 668 Arab immigrants.¹⁹ When we stratified by sex, the sample included 120,102 US-born non-Hispanic White women and 1,065 MENA women. The stratified sample sizes for the comparison study were 391 men and 277 women (see table 2).

Table 2:

Summary of sample sizes by condition, nativity status and sex, 2000–2014.

Years	Topic	US-Born NHW	MENA Immigrants	Arab Immigrants
2000–2011	Chronic disease	205,763	1,478	463
	Women preventive	113,406	676	205
	Men preventive	88,413	750	238
2000–2014	Cigarette smoking	266,813	1942	668
	Men	120,102	1065	391
	Women	146,711	877	277

Independent Variable

The independent variable was region of birth among non-Hispanic Whites. The NHIS asks participants to report their country of birth. Those who were not born in the US, a military base, or US territory, were asked to report their country of birth. The NHIS combines responses by world region, including the US, Middle East, Africa, Europe, Russia/former USSR countries, among others. In the current study, two non-Hispanic White groups were created to compare 1) non-Hispanic Whites who were born in the US and 2) non-Hispanic Whites who were born in countries located in the Middle East or Africa. The Middle East category includes Aden, Arab Palestine, Arabia, Armenia, Bahrain, Cyprus, Gaza Strip, Iran, Iraq, Israel, Jordan, Kuwait, Syria, Lebanon, “Middle East,” Oman, Palestine, Persia, Qatar, Saudi Arabia, Syria, Turkey, United Arab Emirates, West Bank, Yemen. The Africa category includes countries located in the African continent (e.g. Egypt, Libya, Morocco, Somalia, Sudan, South Africa, among others), plus Canary Islands, Comoros, Madagascar, and Madeira Islands. We compare our findings to the independent variable that used country of birth to examine Arab immigrants from the Middle East. In the comparison study, data from US-born non-Hispanic Whites and Arab immigrants were extracted from the four studies. The Arab immigrant group included individuals who reported that they were born in Bahrain, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine (including Arab Palestine, West Bank, Gaza Strip) Qatar, Saudi Arabia, Syria, United Arab Emirates, or Yemen.

Dependent Variables

Chronic Diseases

Respondents were asked whether they had ever been told by a doctor or other health professional that they had asthma; a stomach, duodenal, or peptic ulcer; cancer; diabetes; and heart disease. Obesity was determined by body mass index (BMI) scores. (BMI < 30 kg/m² not obese; BMI > =30 kg/m² obese).¹⁸

Women’s Preventive Health Behaviors

Women were asked whether they had a flu shot in the last 12 months (yes/no), ever had a pneumonia vaccine (yes/no), and during selected years, whether they had ever had (yes/no) a Pap smear, mammogram, (2000, 2003, 2005, 2008, 2010), and clinical breast examination (2000, 2005, 2010).¹⁶

Men’s Preventive Health Behaviors

Men were asked whether they had a flu shot in the last 12 months (yes/no), ever had a pneumonia vaccine (yes/no), and during selected years, whether ever had a PSA test (2000, 2003, 2005, 2008, 2010).¹⁷

Cigarette Smoking

Respondents were asked, “Have you smoked at least 100 cigarettes in your entire life?” Those who reported “yes” were asked, “Do you now smoke cigarettes every day, some days or not at all?” Responses were combined to create a dichotomous variable which differentiated current smokers, including every day and some day, from former/never smokers based on previous research.¹⁹

Covariates

Overall, the previously published papers adjusted for demographic, socioeconomic status, health care, behavioral risk factors, and length of time living in the US. Due to low response rates for income levels, multiple imputation methods using five sets of imputed values per year were performed. Depending on the aim of the paper, other covariates were included or excluded. Please see reference for specifics.^16–19

Statistical Analysis

Weighted column percentages and standard errors were calculated among all MENA adults using the 12-year (2000–2011) and 15-year (2000–2014) samples. Weighted means and standard errors were calculated for age. Tests on the equality of proportions were conducted to compare sample characteristics of MENA immigrants to Arab immigrants reported in previous studies (p<.05). Only sample characteristics that were consistent across all studies were compared (sex, marital status, education, employment status, health insurance coverage, and length of time living in US). Logistic regression modeling procedures were conducted to determine associations between the region of birth variable and each outcome among MENA immigrants compared to US-born non-Hispanic Whites based on prior analyses.^16–19 Crude and fully adjusted multivariable models for each outcome were compared to determine whether the confidence intervals were overlapping.

RESULTS

Descriptive Results

Comparisons of selected characteristics of MENA immigrants compared to Arab immigrants are provided in Table 2. For 2000–2011, the mean age of the MENA sample was slightly older than the sample of Arab immigrants (42.9 years vs. 39.7 years). Sex, marital status, highest level of education, and health insurance characteristics were similar among MENA and Arab samples (all p’s>.05). Fewer MENA immigrants were not currently employed (41.1%) compared to 49.3% of Arab immigrants (p=.0331). Over half (54.5%) of the sample of Arab immigrants lived in the US for less than 15 years compared to 46.1% of the MENA immigrant sample (p=.0226). Similar patterns of results were found using the 2000–2014 samples. In addition, more MENA immigrants had a bachelor’s degree or higher level of education (p=.0130) and fewer reported not being employed (p=.0025).

Logistic Regression Results

2000–2011 Sample

Chronic Diseases

Crude and adjusted odds ratios and 95% confidence intervals for chronic disease outcomes among MENA and Arab immigrants are reported in Table 3. For heart disease, asthma and cancer, both MENA and Arab immigrants were statistically and significantly less likely to have the condition compared to US-born non-Hispanic Whites. This pattern was consistent whether the MENA or Arab definition was used. For example, for heart disease, MENA immigrants had 0.52 times lower odds (95% CI=0.44–0.61) of having heart disease compared to US-born non-Hispanic Whites in the unadjusted model. Results remained statistically significant in the multivariable model. Similar patterns were observed for asthma and cancer. For diabetes, ulcer and obesity, and with a few exceptions, there were no statistically significant differences between MENA immigrants and US-born non-Hispanic Whites in crude or multivariable models. The few exceptions were found for ulcer and obesity when we used the MENA category. For ulcer, we found that MENA immigrants had 0.73 times lower odds (95% CI=0.58–0.91) of reporting ever having an ulcer compared to US-born non-Hispanic Whites in the unadjusted model. Results were attenuated and no longer significant in the multivariable model (OR=0.63; 95% CI=0.38–1.03). For obesity, and in the new research findings, MENA immigrants had lower odds of being obese compared to US-born non-Hispanic Whites in the unadjusted and adjusted models.

Table 3.

Selected characteristics of the NHIS samples from comparison studies.

	2000–2011 Chronic disease			2000–2014 Cigarette smoking
	MENA Weighted % (SE)	Arab Nations Weighted % (SE)	p	MENA Weighted % (SE)	Arab Nations Weighted % (SE)	p
	N=1,478	N=463		N=1,942	N=668
Mean Age (SE)	42.9 (0.59)	39.7 (1.03)	NA	43.8 (0.55)	40.2 (0.97)	NA
Male Sex	54.9 (1.62)	55.5 (2.66)	.8662	54.7 (1.45)	57.0 (2.30)	.4385
Marital status
Never married	21.0 (1.38)	21.9 (2.49)	.8476	19.8 (1.13)	24.3 (2.26)	.2392
Married/live with partner	66.7 (1.58)	66.1 (2.80)	.8459	67.6 (1.32)	63.7 (2.46)	.1379
Divorced/widowed/separated	12.3 (0.95)	12.0 (1.75)	.9522	12.6 (0.80)	11.9 (1.40)	.8697
Education
Less than HS	12.1 (1.27)	18.2 (2.62)	.1844	11.6 (1.03)	18.2 (2.11)	.0899
HS diploma/GED	18.9 (1.25)	22.0 (2.25)	.5008	19.1 (1.21)	23.0 (2.02)	.3113
Some college/Associate’s	21.2 (1.13)	19.7 (2.01)	.7566	20.9 (1.07)	19.1 (1.73)	.6599
Bachelor’s degree or higher	47.8 (2.17)	40.1 (3.85)	.0609	48.4 (2.10)	39.8 (2.98)	.0130
Not currently employed	41.1 (1.64)	49.3 (2.99)	.0331	42.8 (1.54)	52.4 (2.74)	.0025
No health insurance	22.3 (1.50)	23.0 (2.74)	.8806	20.6 (1.30)	22.0 (2.41)	.7215
<15 years living in US	46.1 (1.96)	54.5 (3.71)	.0226	45.4 (1.82)	53.1 (3.20)	.0142

^aMiddle Eastern and North African (MENA) variable includes non-Hispanic White adults born in countries located in the Middle East and Africa.

^bArab Nations variable includes foreign-born adults born in Arab Palestine, Bahrain, Gaza Strip, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine, Qatar, Saudi Arabia, Syria, United Arab Emirates, West Bank, and Yemen.

Abbreviations: NA=Not available, unable to compare because standard deviation not reported; GED=General educational development test; HS=High school; US=United States.

Women’s Preventive Health Behaviors

Crude and adjusted odds ratios and 95% confidence intervals for both MENA and Arab immigrant (i.e. Arab Nations) women’s preventive health behaviors are reported in Table 4. All confidence intervals were overlapping, indicating no differences in the odds of women’s preventive health behaviors for MENA immigrants compared to US-born non-Hispanic Whites when using the more inclusive classification method. Notably, for mammogram, MENA women had 0.57 times lower odds (95% CI=0.42–0.77) of receiving a mammogram compared to US-born non-Hispanic Whites in the unadjusted model. Results were attenuated and no longer significant in the multivariable model (OR=1.37; 95% CI=0.61, 3.07).

Table 4.

MENA and Arab Immigrant comparisons of crude and adjusted odds ratios (95% confidence intervals) for chronic disease outcomes, 2000–2011 NHIS.

	New Research Findings MENA Variablea		Previous Research Findings Arab Nations Variableb
	Unadjusted Model OR (95% CI)	Multivariable ModelcOR (95% CI)	Unadjusted Model OR (95% CI)	Multivariable ModelcOR (95% CI)
Heart disease
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.52 (0.44, 0.61)	0.44 (0,.31, 0.62)	0.46 (0.35, 0.62)	0.46 (0.33, 0.65)
Asthma
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.54 (0.43, 0.69)	0.44 (0.27, 0.70)	0.47 (0.31, 0.73)	0.43 (0.27, 0.69)
Cancer
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.24 (0.16, 0.34)	0.12 (0.03, 0.41)	0.07 (0.02, 0.23)	0.05 (0.01, 0.16)
Diabetes
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.87 (0.67, 1.12)	1.07 (0.58, 1.96)	0.95 (0.60, 1.50)	1.15 (0.66, 1.99)
Ulcer
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.73 (0.58, 0.91)	0.63 (0.38, 1.03)	0.84 (0.55, 1.29)	0.63 (0.36, 1.11)
Obesity
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.70 (0.60, 0.82)	0.47 (0.32, 0.68)	0.81 (0.62, 1.05)	0.86 (0.62, 1.18)

^aMiddle Eastern and North African (MENA) variable includes non-Hispanic White adults born in countries located in the Middle East and Africa.

^bArab Nations variable includes foreign-born adults born in Arab Palestine, Bahrain, Gaza Strip, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine, Qatar, Saudi Arabia, Syria, United Arab Emirates, West Bank, and Yemen.

^cMultivariable models adjusts for age (<45 years as referent), sex (female as referent), marital status (married as referent), education (bachelor’s degree or higher as referent), employment (employed as referent), imputed poverty ratio (≥200 % as referent), health insurance coverage (yes as referent), place most often received care (doctor’s office/HMO as referent), smoking history (no as referent) and years in the US (US-born as referent).

Men’s Preventive Health Behaviors

Crude and adjusted odds ratios and 95% confidence intervals for MENA and Arab immigrant men’s preventive health behaviors are reported in Table 5. All confidence intervals were overlapping, indicating no differences in the odds of men’s preventive health behaviors for MENA immigrant men compared to US-born non-Hispanic White men when using the more inclusive classification method. For PSA tests, MENA immigrant men had 0.60 times lower odds (95% CI=0.40–0.88) of receiving a PSA test compared to US-born non-Hispanic White men in the unadjusted model. In the multivariable model, MENA immigrant men had 0.21 times lower odds (95% CI=0.07–0.60) of receiving a PSA test compared to US-born non-Hispanic White men. For men from the Arab Nations there were no statistically significant differences between foreign born MENA/Arab and US-born non-Hispanic whites in either the unadjusted or adjusted model.

Table 5.

MENA and Arab immigrant comparisons of crude and adjusted odds ratios (95% confidence intervals) for women’s preventive health outcomes, 2000–2011 NHIS.

	New Research Findings MENA Variablea		Previous Research Findings Arab Nations Variableb
	Crude Model OR (95% CI)	Multivariable Modelc OR (95% CI)	Crude Model OR (95% CI)	Multivariable Modelc OR (95% CI)
Flu Vaccine
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.50 (0.40, 0.63)	0.42 (0.26, 0.70)	0.48 (0.32, 0.72)	0.34 (0.21, 0.58)
Pneumonia Vaccine
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.39 (0.28, 0.53)	0.38 (0.18, 0.82)	0.29 (0.15, 0.56)	0.14 (0.06, 0.32)
Mammogram
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.57 (0.42, 0.77)	1.37 (0.61, 3.07)	0.47 (0.27, 0.84)	0.50 (0.25, 1.00)
Pap Smear
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.21 (0.14, 0.33)	0.07 (0.03, 0.17)	0.21 (0.11, 0.39)	0.13 (0.05, 0.31)
Clinical Breast Exam
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.28 (0.18, 0.44)	0.27 (0.09, 0.81)	0.27 (0.11, 0.61)	0.16 (0.07, 0.37)

^aMiddle Eastern and North African (MENA) variable includes non-Hispanic White adults born in countries located in the Middle East and Africa.

^bArab Nations variable includes foreign-born adults born in Arab Palestine, Bahrain, Gaza Strip, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine, Qatar, Saudi Arabia, Syria, United Arab Emirates, West Bank, and Yemen.

^cMultivariable models adjusts for age (<45 years as referent), marital status (married as referent), education (bachelor’s degree or higher as referent), employment (employed as referent), imputed poverty ratio (≥200 % as referent), health insurance coverage (yes as referent), place most often received care (doctor’s office/HMO as referent), smoking history (no as referent), BMI (normal weight as referent) and years in the US (US-born as referent).

2000–2014 Sample

Cigarette Smoking

Crude and adjusted odds ratios and 95% confidence intervals for cigarette smoking among MENA and Arab immigrants are reported in Table 6. Like the 2000–2011 sample, all confidence intervals were overlapping, indicating no differences in the odds of cigarette smoking among MENA immigrants compared to US-born non-Hispanic Whites when using the more inclusive classification method. MENA immigrants had 0.70 times lower odds (95% CI=0.60–0.83) of being a current smoker compared to US-born non-Hispanic Whites in the unadjusted model. Results remained statistically significant in the multivariable model. MENA immigrants had 0.66 times lower odds (95% CI=0.54–0.81) of being a current smoker compared to US-born non-Hispanic Whites after adjusting for age, sex, marital status, education, employment, income based on the poverty ratio, health insurance coverage, diagnosis of lung cancer, COPD, stroke or coronary artery disease, citizenship and years living in the US. When stratified by sex, there were no statistically significant differences in current smoking between MENA immigrant men and US-born non-Hispanic White men in crude or multivariable models. However, among women, whether defined as MENA or using the Arab Nations definition, they were less likely to report being a current smoker compared to US-born non-Hispanic White women.

Table 6.

MENA and Arab immigrant comparisons of crude and adjusted odds ratios (95% confidence intervals) for men’s preventive health behaviors, 2000–2011 NHIS.

	New Research Findings MENA Variablea		Previous Research Findings Arab Nations Variableb
	Unadjusted Model OR (95% CI)	Multivariable Modelc OR (95% CI)	Unadjusted Model OR (95% CI)	Multivariable Modelc OR (95% CI)
Flu Vaccine
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.64 (0.51, 0.81)	0.46 (0.25, 0.82)	0.42 (0.27, 0.64)	0.38 (0.21, 0.67)
Pneumonia Vaccine
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.83 (0.75, 0.93)	0.22 (0.10, 0.49)	0.39 (0.24, 0.63)	0.33 (0.16, 0.70)
PSA Test
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.60 (0.40, 0.88)	0.21 (0.07, 0.60)	0.57 (0.26, 1.27)	0.91 (0.17, 4.74)

^aMiddle Eastern and North African (MENA) variable includes non-Hispanic White adults born in countries located in the Middle East and Africa.

^bArab Nations variable includes foreign-born adults born in Arab Palestine, Bahrain, Gaza Strip, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine, Qatar, Saudi Arabia, Syria, United Arab Emirates, West Bank, and Yemen.

^cMultivariable models adjusts for age (<45 years as referent), marital status (married as referent), education (bachelor’s degree or higher as referent), employment (employed as referent), imputed poverty ratio (≥200 % as referent), health insurance coverage (yes as referent), place most often received care (doctor’s office/HMO as referent), smoking history (no as referent), BMI (normal weight as referent) and years in the US (US-born as referent).

CONCLUSION

The objective of this study was to assess if MENA immigrant health reflects the same patterns found in previous research focusing on as Arab immigrant health. More specifically, we calculated 1) chronic disease, 2) women’s preventive health behaviors, 3) men’s preventive health behaviors, and 4) cigarette smoking outcomes estimates among MENA immigrants and compared findings to previous studies using Arab immigrants following the classification method for individuals born in countries located in the Middle East and are part of the Arab Nations. The findings did not differ whether the MENA or Arab definition was used. We did find statistically significant differences between MENA and Arab Nations regarding bachelor’s degree or higher, not employed and years in US. Given the MENA category captures more countries, these statistically significant differences speak to the heterogeneity of the MENA category versus restricting it to just Arab Nations.

From the research, it appears that the MENA category reflects the Arab immigrant experience despite variations in SES and years in the US, even though it includes a wider set of origins, some of which are not Arab. Further research should be conducted among the MENA category to understand the complexity of sociodemographic characteristics such as SES and years in the US, which might showcase important heterogeneity depending on the individual’s place of origin. We provide several examples for additional clarity as to why the broader MENA category should be used compared to the Arab category. First, the MENA category allows for a larger sample size and representation of MENA individuals. Second, many of the confidence intervals in our results are narrower (and thus more stable) for MENA immigrants compared to Arab immigrants. This suggests our results may be more reliable and valid when using the MENA definition. Third, if we use the broader MENA category, we can better explore the heterogeneity of MENA individuals. That is, it is easier to translate results from a larger group to a smaller group versus the other way around. Lastly, the study that compared Arab Americans to Iranians in California suggested that Iranian Americans had a lower prevalence of obesity, current smoking and ever smoking but a higher prevalence of hyperlipidemia, prediabetes, depression, and anxiety.²² This illustrates the importance of including MENA groups in the discourse and how the results vary when comparing one MENA group to all Arab Americans, for example.

The path to inclusion and recognition of MENA/Arab individuals by the federal government has been circuitous, extensive, and ongoing. In 1944, a legal ruling deemed all persons from the MENA region, “White by law”.²³ Approximately 35 years later (in the early 1980s), Arab Americans began lobbying for a distinct racial, or ancestral, classification.²⁴ During the same time, the US Census began collecting data on Arab Americans using the ancestry question.²⁴ This is approximately 90 years after the first US Census was instituted and 100 years after the first Arabs began immigrating to the US. In effect, the health of MENA/Arab American individuals has been invisible and understudied. Including a MENA identifier on future data collections will both represent Arab Americans as well as better represent health disparities by distinguishing this population from the White racial category.

Table 7.

MENA and Arab immigrant comparisons of crude and adjusted odds ratios (95% confidence intervals) for current smoking, 2000–2014 NHIS.

	New Research Findings MENA Variablea		Previous Research Findings Arab Nations Variableb
	Unadjusted Model OR (95% CI)	Multivariable Modelc OR (95% CI)	Unadjusted Model OR (95% CI)	Multivariable Modelc OR (95% CI)
All
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.70 (0.60, 0.83)	0.66 (0.54, 0.81)	0.92 (0.73, 1.16)	0.64 (0.46, 0.88)
Men
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.93 (0.77, 1.11)	0.90 (0.72, 1.12)	1.33 (1.02, 1.74)	0.96 (0.65, 1.43)
Women
US-Born NH Whites	1.00	1.00	1.00	1.00
Foreign-born MENA/Arab	0.43 (0.33, 0.57)	0.42 (0.30, 0.59)	0.41 (0.24, 0.71)	0.28 (0.15, 0.53)

^aMiddle Eastern and North African (MENA) variable includes non-Hispanic White adults born in countries located in the Middle East and Africa.

^bArab Nations variable includes foreign-born adults born in Arab Palestine, Bahrain, Gaza Strip, Iraq, Jordan, Kuwait, Lebanon, Oman, Palestine, Qatar, Saudi Arabia, Syria, United Arab Emirates, West Bank, and Yemen.

^cMultivariable models adjusts for age, sex, marital status (married as referent), education (bachelor’s degree or higher as referent), employment (employed as referent), imputed poverty ratio (≥200 % as referent), health insurance coverage (yes as referent), comorbidity, citizenship status (citizen as referent) and years in the US (US-born as referent).