A Systematic Examination of Food Intake and Adaptation to the Food Environment by Refugees Settled in the United States 1 4

Transcription

1 REVIEW A Systematic Examination of Food Intake and Adaptation to the Food Environment by Refugees Settled in the 1 4 Youfa Wang, 5 * Jungwon Min, 5 Kisa Harris, 7 Jacob Khuri, 8 and Laura M Anderson 6 5 Systems-oriented Global Childhood Obesity Intervention Program, Department of Epidemiology and Environmental Health and 6 School of Nursing, University at Buffalo, The State University of New York, Buffalo, NY; 7 Tougaloo College, Tougaloo, MS; and 8 University of Redlands, Redlands, CA ABSTRACT The is the largest refugee resettlement country in the world. Refugees may face health-related challenges after resettlement in the, including higher rates of chronic diseases due to problems such as language barriers and difficulty adapting to new food environments. However, reported refugee diet challenges varied, and no systematic examination has been reported. This study examined refugee food intake pre- and postresettlement in the and differences in intake across various refugee groups. We systematically reviewed relevant studies that reported on refugee food intake and adaptation to the US food environment. We searched PubMed for literature published between January 1985 and April 2015, including cross-sectional and prospective studies. Eighteen studies met inclusion criteria. Limited research has been conducted, and most studies were based on small convenience samples. In general, refugees increased meat and egg consumption after resettling in the. Changes in refugee intake of vegetables, fruits, and dairy products varied by socioeconomic status, food insecurity, past food deprivation experience, length of stay in the, region of origin, and age. South Asians were more likely to maintain traditional diets, and increased age was associated with more conservative and traditional diets. Despite the abundance of food in the, postresettlement refugees reported difficulty in finding familiar or healthy foods. More research with larger samples and follow-up data are needed to study how refugees adapt to the US food environment and what factors may influence their food- and health-related outcomes. The work could inform future interventions to promote healthy eating and living among refugees and help to reduce health disparities. Adv Nutr 2016;7: Keywords: refugee, diet, food environments, health disparity, Introduction Refugees are individuals who reside outside of their home country because of suffering, feared persecution, violence, and/or war. According to the United Nations Refugee Agency, there are 59.5 million forcibly displaced people worldwide (1). The is the largest refugee resettlement country, with an allocation of 70,000 admissions for the fiscal year 2015 (2). 1 Supported in part by a research grant from the NIH (grant 1R01HD A1). 2 Author disclosures: Y Wang, J Min, K Harris, J Khuri, and LM Anderson, no conflicts of interest. 3 The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the funder. 4 Supplemental Figure 1 is available from the "Online Supporting Material" link in the online posting of the article and from the same link in the online table of contents at *To whom correspondence should be addressed. youfawan@buffalo.edu. Research indicates that US refugees have elevated rates of chronic diseases, including obesity, diabetes, hypertension, malnutrition, and anemia (3 5), compared with US-born residents or first-generation immigrants (6). Rates of chronic diseases vary within and across refugee groups by sex, region of origin, and length of stay in the (7 9). These health disparities may be attributable to refugee food intake and postresettlement as well as food-related acculturation. However, very limited research has examined dietary intake and health outcomes in postresettlement US refugees. Refugees face many health-related challenges on arriving to the. First, refugees encounter language barriers: they need time to acculturate to unfamiliar language and food environments in the (10). Second, refugee beliefs and home-country culture, in conjunction with postresettlement socioeconomic status (SES; which is 1066 ã2016 American Society for Nutrition. Adv Nutr 2016;7: ; doi: /an

2 often lower), influence what types of food can be purchased and consumed (11). Third, limited information about foods, shopping, and recipes in the creates another barrier to purchasing healthy foods (12). Fourth, high intake of processed and energy-dense foods in the contributes to chronic disease risk (13). Indeed, refugee food intake and adaptation to the US food environment is a complex phenomenon. In addition, there is limited understanding of what factors influence postresettlement food intake and how to best promote refugee health and nutrition. Given the national priority of eliminating health disparities (14), health promotion and empowerment of postresettlement refugees in the United States is necessary. This study comprehensively examined food intake including status, changes, and adaptation to the food environment among refugees living in the based on published research. Specifically, we described refugee current food groups, knowledge and perception of foods in the United States, their changes after resettlement, challenges in food insecurity, ability to locate familiar foods, adaptation to food preparation environment in the, and differences in food intake between refugees from different countries of origin. Methods Literature search strategies. We searched PubMed to identify original studies published between 1 January 1985 and 31 April 2015, evaluating food intake and adaptation to the food environment by refugees living in the. We used key words including refugee, nutrition, diet, food, and. The initial search yielded 80 articles, which were reviewed by two coauthors and resulted in 18 articles meeting inclusion criteria. Study inclusion and exclusion criteria. Studies meeting the following inclusion criteria were included in the review: 1) published in English, 2) studied US refugees, and 3) reported results regarding refugee dietary intake. Our study exclusion criteria included the following: 1) not relevant to refugee dietary intake and 2) studied refugees in other countries. Data extraction. We extracted information, including study sample descriptions, settings, and the study year. Outcome data varied across studies. FFQs and focus group discussions were considered useful measures of food intake and adaptation to the food environment, respectively. Three coauthors carried out the literature search and data extraction. Results and Discussion Main characteristics of the included studies Study design. The 18 included studies consisted of 17 crosssectional studies and 1 cohort study (Table 1). Eleven studies recruited refugees by snowball (using existing subjects acquaintances) or purposive sampling, and 7 studies randomly selected participants. Only 3 studies included >200 participants; 10 included <50 participants. Studies were conducted since 1985, and most (14 studies) were reported over the past decade (Supplemental Figure 1). Research methods and assessment of dietary intake. Seven studies used qualitative approaches and included group or individual interviews. Five studies used dietary records (i.e., 24-h recall, FFQ, or food diary) to quantify the amount and/or frequency of foods consumed. Six articles used survey questionnaires without direct measurements of refugee food intake. Participant characteristics. Ten articles studied Southeast Asian refugees (Burma, Bhutan, Cambodia, Hmong, Vietnam), 4 articles studied African refugees (Sudan, Somalia, Liberia), and 4 articles had mixed samples of Southeast Asian, African, Middle Eastern, and Caribbean origin. Participants in 6 studies lived in the eastern (i.e., Massachusetts, Maine, Atlanta, and Minnesota); the remaining lived in California, Washington, Texas, and the Midwest. Ten studies included both males and females, and 8 studies included females, exclusively. Three studies included children, while the majority included adults only (83%). Most study participants had low SES. Rates of participation in governmentsupported nutrition programs (i.e., Supplemental Food Program for Women, Infants, and Children, Supplemental Nutrition Assistance Program, etc.) were 30 92% as reported in 7 studies. Among the 11 studies reporting length of stay in the or acculturation scores, most participating refugees had low acculturation or had lived in the <5 y. Most refugee participants [except 1 study (23)] were unable to speak or write in English. Interpreters assisted in data collection. Main study outcomes. The studies covered food intake related topics, including major foods consumed, food preferences, food culture, and nutritional problems (i.e., food insecurity, vitamin deficiency, unhealthy eating behaviors, unfamiliarity with available foods in the, and obesity). Additional outcomes included the level of conservativeness (i.e., about traditional meals from the country of origin) and the need for knowledge and advice pertaining to healthy nutrition. Changes in food intake between pre- and postresettlement in the Seventeen studies described food intake changes after resettlement in the. After resettling, refugees increased energy intake with more consumption of meat, eggs, high-fat food, fast food, and soft drinks. Consumption of vegetables, fruits, and dairy products, however, exhibited discrepancies (see Tables 2 and 3). Energy intake. Liberian and Hmong refugees reported increased overall food consumption in the (23, 27). In a study of Bosnian, Cuban, and Iranian refugees, 60% reported consumption of excessive energy, in general: 50%, too many sweets and 19%, too much fat (30). Cuban, Cambodian, and Hmong refugees indicated that an abundance of available food (18, 23, 30) and a variety of cuisines in the (23) compared with near-starvation and restricted food availability in regions of origin and/or refugee camps contributed to excess food consumption (23). Refugee diet adaptation review 1067

3 TABLE 1 Main characteristics of the 18 studies included in this study regarding refugees food intake 1 Author, Sampling year (ref) Study design 2 method Dharod, 2015 (15) Anderson et al., 2014 (16) Haley et al., 2014 (17) Peterman et al., 2013 (18) Dharod et al., 2013 (19) CDC, 2011 (20) and Cuffe et al., 2014 (21) Peterman et al., 2011 (22) Time of data collection II, FFQ Snowball Jun 2010 to Jul 2011 SQ via FG, FFQ Purposive 3, snowball Region of origin or race Montagnard (Vietnam) Current US living area Sample size Sex Age, y SES North Carolina 42 F Mean: 40 y No formal education: 60%; household income #$500/mo: 39%; no health insurance: 66%; SNAP: 66% Jul Oct 2002 Sudan Atlanta, GA 49 (households) M, F Child #3 y in each house; others: 68.9% from 21 to 30 y Food insecurity: 37%; reported child hunger: 12% Time lived in or acculturation to the Poor English proficiency: 88% Lived in the,4 y: 89.8% FG, II Snowball NA Burma Worcester, MA 18 M, F y NA NA FG, SQ Random Apr May 2007 (FG); Sep Nov 2007, Apr Jun 2008 (SQ) SQ via II, FFQ Snowball Oct 2006 to Dec 2007 Intervention study FG, SQ, 24-h dietary recall Random Dec 2007 to Nov 2008; Jun 2009 to Jan 2011; Sept 2010 to Jan 2011; NA Cambodia Lowell, MA 11 (FG); 160 (SQ) F y (FG); y (SQ) #1 y of education: 24%; high school: 21%; employed: 58%; above the federal poverty rate: 15%; food security rate: 24%; current FSP: 30% Somalia Lewiston, ME 195 F Mean: 33.6 y No formal education: 49%; mean household income,$1000/mo; food insecurity rate: 67%; current SNAP: 92%; current WIC: 75% Bhutan Minnesota, Utah, and Texas 99; 141; 326; 49 M, F y; y NA Random Cambodia Lowell, MA 11 (FG); 150 (SQ) F y Education #1 y: 23%; some education: 54%; high school: 23% Lived in the United States: y; acculturation score 5 : Poor English proficiency: 72% Acculturation score 5 : ; lived in the : y (Continued) 1068 Wang et al.

4 TABLE 1 (Continued) Author, Sampling year (ref) Study design 2 method Vue et al., 2011 (23) Rondinelli et al., 2011 (24) Peterman et al., 2010 (25) Hadley et al., 2010 (26) Patil et al., 2009 (27) Willis and Buck, 2007 (28) Hadley and Sellen, 2006 (29) Time of data collection Region of origin or race Current US living area Sample size Sex Age, y SES FG Purposive 3 NA Hmong Northern California 15 F y Living in low-income community FG, II Random Dec 2006 to Mar 2007 Afghanistan, Ethiopia, Iran, Iraq, Russia, Somalia, Sudan, Vietnam San Diego County, CA 16 (refugees resettled in the ; other nonrefugees were included) M, F y,high school: 37%, no college degree FG, SQ Random End of 2010 Cambodia Lowell, MA 11 (FG); 133 (SQ) F y,high school: 82% FG, SQ Snowball through local resettlement agency 2006 Sierra Leone, Liberia, Ghana, Somalia, Togo, Russia (Meskhetian Turk) Midwest of the 281 M, F y $1 y of education: 79%; currently employed: 61%; household income,$500/ mo: 25%; current FSP: 53% FG, SQ Purposive 3 NA Mixed 6 NA 157 F y $High school education: 50%; household income,$1000/ mo: 48%; current FSP: 50%; current WIC: 47%; currently employed: 59% II, SQ, 3-d 24-h food diary Purposive 3 Jun 2005 to Feb 2006 Sudan (Dinka or Nuer) SQ, FG Snowball NA Liberia Northeastern NA 31 (II); 29 (SQ) M, F y Some college and above: 42%; unemployed: 35% 33 F Mean: 35.9 y $High school: 51%;,$1000/ mo: 60%; food insecurity rate: 85%; child hunger: 42%; FSP: 51%; WIC: 54%; currently employed: 48% Time lived in or acculturation to the All speak English well Lived in the United States: 3.0 y (interquartile: y) Acculturation score 5 :2 Able to read English: 56%; lived in the : y Lived in the United States,5 y: 55% All subjects had lived in the,5 y (Continued) Refugee diet adaptation review 1069

5 TABLE 1 (Continued) Author, Sampling year (ref) Study design 2 method Barnes and Almasy, 2005 (30) Rairdan and Higgs, 1992 (31) Story and Harris, 1989 (32) Story and Harris, 1988 (33) Time of data collection Region of origin or race Current US living area Sample size Sex Age, y SES II Snowball NA Bosnia, Cuba, Iran NA 31 M, F y High school: 65%; some college: 23%; employed in low-paying job: 58% Time lived in or acculturation to the Fluent English: none; lived in the United States: 2.8 y (3 mo to 5 y) II NA NA Hmong Spokane, WA 13 (households) M, F NA NA NA II (including itemized food consumption) SQ (including itemized food consumption) Randomly selected from a school list Randomly selected from Southeast Asian refugee high school students June 1984 Hmong, Cambodia NA 60 M, F y Husbands employed: 23%; wives employed: 17%; FSP: 73%; WIC: 42% Fall 1985 Vietnam, Hmong, Cambodia Minneapolis, MN 207 M, F High-school age; grades No father at home: 56.5%; no mother at home: 34.5%; youth employed part time: 8% All subjects had lived in the,5 y All subjects had lived in the,5 y 1 FFQ, Food Frequency Questionnaire; FG, focus group interview; FSP, food stamp program; II, individual interview; NA, not available; ref, reference; SES, socioeconomic status; SNAP, Supplemental Nutrition Assistance Program; SQ, survey questionnaire; WIC, Supplemental Food Program for Women, Infants, and Children. 2 Except for one intervention study (20, 21), all were cross-sectional studies. 3 Purposive sampling: a nonrepresentative subset of some larger population under very specific characteristics or purpose. 4 Mean 6 SD (all such values). 5 Acculturation was measured by using a scale that focuses on cultural orientation and values rather than behaviors or proxies for acculturation. The scale ranges from 1 to 5, with 1 representing identifying only with natives and 5 representing identifying only with Americans (10). 6 Liberia, Ivory Coast, Burundi, Ethopia, Somalia, Kenya, Russia (Meskhetian Turk) (FG); Liberia, Somalia, Russia (Meskhetian Turk) (SQ) Wang et al.

6 TABLE 2 Main research topics and findings of the 18 studies included in this study regarding refugees food intake 1 Author, year (ref) Dharod, 2015 (15) Anderson et al., 2014 (16) Haley et al., 2014 (17) Peterman et al., 2013 (18) Dharod et al., 2013 (19) CDC, 2011 (20) and Cuffe et al., 2014 (21) Region of origin or race Main research topics Main findings Montagnard (Vietnam) Sudan Burma Cambodia Somalia Bhutan Characteristics of major food groups Food insecurity and its association with food expenditure and low-cost and/or high nutrient density food consumption Needs pertaining to nutrition and exercise information Food insecurity issues in those with depression and lower income and acculturation scores Difficulties transitioning to the US food environment Food insecurity and its association with meat, fruit, and vegetable consumptions and overweight and obesity Vitamin B-12 deficiency and its related food intake Intake of meat, eggs, and dairy products increased, but plant protein, bean, and lentil intake decreased after resettlement (P # 0.05 for all) Those who experienced preresettlement hunger or received $$500/mo in SNAP were 3 times more likely to have high meat intake (P # 0.05 for all) Caregivers in food-insecure households reported significantly lessfrequent consumption of highcost foods (breakfast cereal, milk and dairy products, vegetables) but more frequent consumption of fresh meat than those in foodsecure households (P, 0.05 for all) Caregivers in households with child hun ger consumed more servings of starchy cereals (rice, maize, pasta, semolina, couscous, maize porridge, noodles) and green leafy vegetables than those in food-secure households (P, 0.05 for all) There was difficulty finding foods that were familiar or that they knew were healthy They did not eat food that they were not used to The risk of food insecurity increased with depression (OR: 3.7; 95% CI: 1.3, 11.1) and decreased with higher income (OR: 0.3; 95% CI: 0.1, 0.9), and higher acculturation scores (OR: 0.5; 95% CI: 0.2, 1.0) Those arriving in the in the 1980s had challenges with the shop ping environment and anxiety around food, but those arriving in the in the 1990s 2000s easily found many Cambodian stores in the Food insecurity was associated with 80 82% lower consumption of fruits and green leafy vegetables (P, 0.05 for all) The risk of child hunger increased with household consumption of egg (OR: 21.2; 95% CI: 7.8, 57.3) and meat (OR: 11.2; 95% CI: 1.4, 89.2) $1 time/d Insecure families had poor dietary habits and high overweight and obesity rates (41% overweight, 24% obesity; P, 0.05 for all) Past shortage of meat, egg, and dairy products in Nepal caused vitamin B-12 deficiency After the intervention, 58% showed improvement in the vitamin B-12 blood test 85% reported consuming more vitamin B-12 rich food Direction of food intake changes [: chicken, pork, beef, egg, dairy products : fruits, vegetables Y: breakfast cereal, milk, other dairy products, vegetables in those with food insecurity [: fresh meat in those with food insecurity : retained familiar food items and healthy foods : retained familiar food items and healthy foods [: meat, eggs in those with child hunger Y: fruits, vegetables in those with household food insecurity [: meat, eggs, and dairy due to CDC supplementation program (Continued) Refugee diet adaptation review 1071

7 TABLE 2 (Continued ) Author, year (ref) Peterman et al., 2011 (22) Region of origin or race Main research topics Main findings Cambodia Fruit, vegetable, and whole grain consumption and its association with levels of education and acculturation Vue et al., 2011 (23) Hmong Perceptions about Hmong and American food culture Rondinelli et al., 2011 (24) Peterman et al., 2010 (25) Hadley et al., 2010 (26) Afghanistan, Ethiopia, Iran, Iraq, Russia, Somalia, Sudan, Vietnam Cambodia Sierra Leone, Liberia, Ghana, Somalia, Togo, Russia (Meskhetian Turk) Factors influencing nutritional problems: past food deprivation, low acculturation or SES, unfamiliarity with available foods The association between past food deprivation or insecurity and the risk of unhealthy eating practices and being overweight or obese Difficulties in navigating the food environment Those with a higher education or acculturation or receiving nutrition advice were more likely to consume more fruits, vegetables, brown rice, and whole grains (P, 0.05 for all) High-acculturation group seldom ate Asian sauce (P, 0.01) Adolescent dietary preferences were influenced by American peers Fast food eaten $2 times/mo: 44% of families with children vs. 13% of those without children (P, 0.01) Rice was the main item for satiety and fresh ingredients of the Hmong diet Fast food and convenience foods changed their dietary habits because of convenience and low price Limited dietary changes among the elderly, while children s diets were affected by television; much less consumption of traditional food among young generations They had many food options and an abundance of food available, and individuals could eat by desire (vs. need or hunger-only) in the Most of them were aware of increased rates of overweight and obesity among refugees The past food shortage, low acculturation or SES, and lack of knowledge contributed to poor eating habits and overeating after resettlement in the United States as food became more widely available They were concerned about weight gain after resettlement in the A few felt uncomfortable with fruits and vegetables not typically in season and a large variety of food American culture and neighborhood environment influenced children s food choices and physical exercise 93.6% of the respondents did not consume whole grains on the reference day Those with higher past food deprivation were more likely to eat high-fat meat (OR: 1.1; 95% CI: 1.0, 1.3) and to be overweight or obese (OR: 1.2; 95% CI:1.0,1.4) Difficulties in shopping (46%), cooking (63%), knowing store locations (40%), locating preferred foods (40%) Direction of food intake changes [: fruits, vegetables, whole grains in those with higher acculturation scores in the and higher education level Y: Asian sauce in those with higher acculturation scores : rice, fresh ingredients, especially in older generations [: overeating in those with previous food insecurity Y: whole grain [: high-fat meat in those with previous food insecurity (Continued) 1072 Wang et al.

8 TABLE 2 (Continued ) Author, year (ref) Patil et al., 2009 (27) Willis and Buck, 2007 (28) Hadley and Sellen, 2006 (29) Region of origin or race Main research topics Main findings Mixed 2 Sudan (Dinka or Nuer) Liberia The association between food insecurity and levels of income and education Mechanisms of diets change in newly arrived refugees and immigrants by amount of time spent in the Refugees limited understanding of healthy US food and its impact on nutrient deficiency and related diseases Household food insecurity and child hunger and associations with foods stamps, income, and education levels The probability of food insecurity was inversely associated with household income (β: 20.28; SE: 0.13; P = 0.04) and having $1 y education (β: 20.98; SE: 0.49; P = 0.05) Those who had difficulties locating food stores were more likely to have food insecurity (OR: 2.5; P, 0.05) The Liberians increased consumption in almost all food categories, especially in meat (86.1%), milk (84.2%), vegetables (78.2%), fruits (76%), and soda (72.3%) The length of stay in was associated with more consumption of seasonings, hot drinks, vegetables, added sugar and sweets, oils, and milk The length of stay in the by Liberian caretakers was associated with their children s fruit consumption Children of caretakers with difficulty speaking English were more likely to consume sodas and snacks and less likely to consume fruits Fruit and vegetable intake did not meet the dietary guidelines and also decreased after resettlement in the They had an unbalanced diet: No fruit (63% Nuer F, 56% Nuer M, 46% Dinka M) No raw or steamed vegetables (50% Dinka F, 46% Dinka M) No milk (38% Nuer F, 33% Nuer M, 54% Dinka M) Easy American and convenience foods (75% of energy resource among Dinka) All refugees shopped at American supermarkets, but they did not know the food items well They were unfamiliar with the US foods and their recipes but still consumed convenience foods and sweetened beverages in large quantities They found an abundance of food in the and were concerned about weight gain after resettlement in the United States Children did not prefer eating traditionally prepared food Child hunger was significantly higher in a family with low income, low education, and food stamps (P, 0.01 for all) Daily intake of milk (P = 0.10) and fruits (P = 0.07) were marginally lower among households with child hunger Direction of food intake changes [: meat, milk, vegetables, fruits, soda in all groups [: seasoning, hot drinks, vegetables, sugar and sweets, oils, milk, and soda in those with higher acculturation scores Y: fruits, vegetables, milk [: high-protein foods (meat, egg, fish), high-starch foods, easy American foods, fast food, sweetened beverages Y: fruits, milk in children with hunger (Continued) Refugee diet adaptation review 1073

9 TABLE 2 (Continued ) Author, year (ref) Barnes and Almasy, 2005 (30) Rairdan and Higgs, 1992 (31) Story and Harris, 1989 (32) Story and Harris, 1988 (33) Region of origin or race Main research topics Main findings Bosnia, Cuba, Iran Hmong Hmong, Cambodia Vietnam, Hmong, Cambodia Knowledge and perceptions of nutrition and healthy behaviors Diet preference of natural and fresh foods without fertilizer and processed with preservatives Strong ties to native foods and traditional diets and their variation by age Strong ties to native foods and traditional meals They reported eating better since their arrival in the (F: 60%, M: 38%) All Cubans mentioned the abundance of available food in They perceived themselves to be consuming too much energy (60%), sweets (50%), or fats (19%), which increased after arriving in ; 55% were overweight Hmong still ate their traditional foods of rice and vegetables accented with chicken, pork, fish, and beef Hmong preferred to eat natural, fresh foods that had not been grown with or processed with preservatives Some drank carbonated beverages or fruit juices as a substitute for water Increased consumption of a variety of foods (see next column) except rice, oil, fresh vegetables, and tea (P, 0.05 for all) Most-preferred foods among adults: rice (100%), steak (97%), oranges (92%), bananas (90%), and apples (82%) Least-preferred foods: cheese (53%), chocolate milk (48%), milk (38%), coffee (32%), and pizza (20%) 42% of adults believed their children need to eat native food 83% of adults preferred Hmong or Cambodian food 75% of teenagers preferred both native and American food 88% felt their diets in the United States were healthier than their diets in Southeast Asia The mean weight gain was 4.54 kg among 63% of respondents since resettling in the Best-liked foods: all Asians, rice (88.8%); Cambodians, orange juice (57.4%); Hmong, chicken (77.6%), apples (75%), and bananas (64.7%); Vietnamese, soda (60.3%) Disliked food: cheese (40.8%) Fruits: consumed regularly (i.e., oranges or orange juice daily: 50%) Pizza and hot dogs weekly: 40% Sweets weekly: 50% Fast food: 30.6% Hmong, 18.9% Cambodians, and 13.3% Vietnamese Preferred their native foods: 38% Hmong, 91% Vietnamese, and 85% Cambodians Desired more American foods: 80% Hmong, 35% Vietnamese, and 43% Cambodian Direction of food intake changes [: energy, fat (hamburgers and French fries, etc.), sugar, and sweets : rice, vegetables (retained traditional diet) [: carbonated beverages, fruit juices : rice, oil, fresh vegetables, and tea [: sweets (soft drinks, candy, etc.), fats, dairy products, proteins, grains and starches, fruits, canned fruits and vegetables, frozen vegetables, coffee Y: fish Like: rice, orange juice, chicken, apples, bananas, soda; dislike: cheese 1 ref, reference; SES, socioeconomic status; SNAP, Supplemental Nutrition Assistance Program; [, increase;, no change; Y, decrease. 2 Liberia, Ivory Coast, Burundi, Ethiopia, Somalia, Kenya, Russia (Meskhetian Turk) from focus group interview; Liberia, Somalia, Russia (Meskhetian Turk) from survey questionnaire Wang et al.

10 TABLE 3 Patterns of changes in dietary intakes of the refugees between post- and preresettlement in the 1 Increased, post- vs. preresettlement Increased consumption of food items. The most common foods with increased consumption were meat, eggs, and highfat meat or fat, reported in 9 studies (15, 16, 19 21, 25, 27, 28, 30, 32). There were several reasons for elevated meat consumption: meat was more accessible, refugees worried about the high cost and reduced satiety of vegetables, and there were concerns about timely use of food stamps and the ability to keep foods fresh (15, 18, 29). For example, Sudanese caregivers in severely food-insecure households consumed affordable fresh meat more frequently but higher-cost and nutritious foods, such as milk and other dairy products, less frequently (P < 0.05) than those in food-secure households (16). This was additional evidence that government-supported nutrition programs, including food stamp provision, may have side effects on refugee health, diet, and food intake (34, 35). Somalian refugee intake of eggs (OR: 21.20; 95% CI: 7.83, 57.34) and meat $1 time/d (OR: 11.21; 95% CI: 1.41, 89.19) was higher among households with child hunger (19). In addition, one study showed that a supplementary program for B-12 deficiency resulted in increased refugee consumption of meat, eggs, and dairy products (20, 21). Finally, a fairly common finding was that the consumption of sweets, sweetened beverages, fruit juices, and fast food increased in refugees after US resettlement (27, 28, 30, 31 33). The longer refugees lived in the, the more likely they were to consume added sugars, oils, seasonings, hot drinks, and vegetables (P < 0.05) (27). Studies reporting increases, n Decreased, post- vs. preresettlement Studies reporting decreases, n Energy Hmong (23), Liberian (27), mixed (30), mixed 2 (24) 4 NR 0 Nutrients Fat South Asian (32), mixed (30) 2 NR 0 Food groups Dairy products Bhutanese (20), Montagnard (15), mixed (27) 3 Liberian 3 (29), 4 Sudanese 2 (16), South Asian (32), Sudanese (28) Eggs Somalian 3 (19), Montagnard (15), Bhutanese (20), Sudanese (28) 4 NR 0 Fast foods Sudanese (28), mixed (30) 2 NR 0 Fruits Cambodian 4 (22), South Asian (32), Liberian (27) 3 Liberian 3 (29), 3 Somalian 2 (19), Sudanese (28) High-starch foods 5 Sudanese (28), Sudanese 3 (16) 2 NR 0 Meats Bhutanese (20), Cambodian 2 (25), Liberian (27), Montagnard 7 NR 0 (15), Somalian 3 (19), Sudanese 2 (16), Sudanese (28) Soft drinks 6 Hmong (31), Liberian 4 (27), South Asian (33), Sudanese (28), 5 NR 0 mixed (32) Sweets and candies Liberian 4 (27), South Asian (32), mixed (30) 3 NR 0 Vegetables Cambodian 4 (22), Liberian (27), Sudanese 3 (16; green leafy vegetables only) 3 Somalian 2 (19), Sudanese (28), Sudanese 2 (16) Whole grains Cambodian 4 (22) 1 Cambodian (25) 1 1 Results are based on reviewing all 18 studies, and only some studies reported on some of the food groups and nutrients. NR, not reported by the identified studies. 2 Subject to past or current food insecurity. 3 Subject to child hunger. 4 Subject to acculturation or length of stay in or education level. 5 Rice, maize, pasta, semolina, couscous, maize, porridge, noodle. 6 Mainly sugar-sweetened beverages. Discrepancies in food items consumed. Changes in refugee consumption of vegetables, fruits, and dairy products varied across studies. Specifically, Liberian, Somalian, and Cambodian refugees increased vegetable, fruit, whole-grain (22, 27, 32), and dairy consumption (15, 20, 21, 27), and they reported believing that vegetables and fruits were good for health (32). In addition, refugee caretakers reported increased fruit consumption in children with increased time in the (27). Refugees with household food insecurity reported lower intakes of vegetables, fruits, milk, dairy, and whole grains (16, 19, 29). For example, household food insecurity among Somali refugees was significantly associated with 80 82% lower consumption of fruits (OR: 0.18; 95% CI: 0.05, 0.67) and green leafy vegetables (OR: 0.20; 95% CI: 0.08, 0.51) (19). Four studies reported no change in consumption of fruits or vegetables after US resettlement because of a preference for traditional diets (15, 23, 31, 32). Refugees also reported beliefs that children should consume native food (27, 32) to preserve culture (16, 23). In particular, South Asian refugees from Hmong, Burma, Cambodia, and Vietnam had strong ties to native foods (17, 32, 33). However, Cambodian adolescent dietary preferences were influenced by American peers (22). One study found that most adults preferred a traditional diet, whereas teenagers preferred a combination of native and US foods (32). In summary, the majority of refugees experienced food intake changes after US resettlement. The amount and 3 Refugee diet adaptation review 1075

11 TABLE 4 Challenges in dietary intakes of the refugees after they resettled in the 1 Food insecurity Difficulties locating familiar foods in the Unfamiliar with food items and food preparation in the 1 Results are based on reviewing all 18 studies; only 8 studies reported about food insecurity. direction of changes varied based on refugee characteristics such as age, region of origin, past food deprivation experience, current SES, food insecurity status, and length of stay in the. Challenges adapting to the US food environment Food insecurity. Eight studies addressed food insecurity, and 7 investigated its association with sociodemographic factors. Food insecurity rates varied across the 4 studies, ranging from 24% (18) to 85% (29) (see Table 4). These studies indicated low income, low education, low acculturation, shorter length of time in the, and language barriers as major factors leading to food insecurity among refugees. Further, refugees lacking knowledge about food outlet locations were 2.5 times more likely to report food insecurity (26). Depression was also associated with food insecurity among refugees (18). Difficulties in locating familiar foods. Although the US food environment provides a myriad of food choices, Burmese refugees, especially during the initial US resettlement period, had trouble locating familiar or healthy foods (17). In a study of Liberian refugees, child hunger was most likely to be present in households in which mothers reported difficulty navigating and understanding the American food environment (29). In addition, 63% of refugees reported not knowing how to cook American foods, and 40% reported difficulties finding stores with desired foods (26). Nevertheless, the level of difficulty in adapting to the US food environment was inconsistent across studies and refugee groups. In one study, although 12% of refugees reported Varied food insecurity rates Cambodian: 24% (18), Sudanese: 37% (16), Somalian: 67% (19), Liberian: 85% (29) Major factors associated with food insecurity in the Depression: Cambodian (18) Low income: Cambodian (18), Liberian (29), Somalian (19), Sudanese (16), mixed (26) Low education: Liberian (29), mixed (26), Somalian (19), Sudanese (16) Limited access to employment: Liberian (29), Sudanese (16) Lack of knowledge in locating food outlets or in navigating food environments: Liberian (29), mixed (26) Short length of time spent in or low acculturation to the : Cambodian (18), Liberian (29), Somalian (19) Language barrier: mixed (26), Somalian (19), Sudanese (16) Consequences of past (in home country or refugee camp) or current food insecurity Overeating: mixed refugee groups (24) Consumed more starchy cereals and green leafy vegetables: Sudanese (16) More-frequent consumption of fresh meat: Cambodian (25), Somalian (19), Sudanese (16); of eggs: Somalian (19) Less-frequent consumption of dairy products: Liberian (29), Sudanese (16); of vegetables: Burmese (17), Somalian (19), Sudanese (16); of fruits: Burmese (17), Liberian (29), Somalian (19) Difficulties finding stores with desired food items: Cambodian (18), Liberian (29), mixed (26) Difficulties finding foods that were familiar or that they knew were healthy: Burmese (17), Liberian (29), Sudanese (16) Missed their own fruit or vegetable garden: Bosnian (30), Burmese (17), Cambodian and Laotian (32); own rice paddy: Cambodian and Laotian (32); own livestock: Cambodian and Laotian (32) Unfamiliar with recipes of American foods: Liberian (29), mixed (26) Unfamiliar with ingredients of food items in packages: Sudanese (28) Uncomfortable with the fruits and vegetables not typically in season: mixed (24) Uncomfortable with pasteurized foods or preservatives: Hmong (31) difficulty locating desired foods, the majority reported that locating preferred foods was unproblematic (27). Difficulties navigating the food environment varied by arrival date in the. For example, Cambodian refugees arriving to the in the 1980s faced challenges providing acceptable meals for themselves in the new shopping environment: this unfamiliar US food led to anxiety around food. However, Cambodians arriving in the 1990s and 2000s found familiar foods in Cambodian stores that had emerged in the (18). Uncomfortable within the US food preparation environment. Despite the abundance of food, not all refugees reported comfort with the food preparation environment in the. For instance, while all Sudanese respondents shopped at American supermarkets, most indicated discomfort or unfamiliarity with ingredients in packaged foods (e.g., bread) (28). South Asian refugees reported discomfort with foods that were pasteurized or contained preservatives (31). Refugees were uncomfortable with the availability of fruits and vegetables not typically in season; most had never seen such a variety of food in their lifetime (24). Because most refugee families in Cambodia, Laos, Bosnia, and Cuba owned rice paddy lots, vegetable gardens, and/or livestock, they were accustomed to conveniently and readily accessing seasonal fruits, vegetables, grains, and fresh protein (17, 30, 32). These families reported challenges finding desired foods in the. Public health implications Overconsumption and obesity. Several researchers reported issues pertaining to refugee food overconsumption Wang et al.

12 TABLE 5 Future research needs for better food adaption among refugees resettled in the 1 Concerning overconsumption and obesity Perceptions and knowledge of healthy diet What refugees need in the United States and from intervention programs for healthy eating 1 Suggestions are based on our review and interpretation of findings from all 18 studies. Liberian refugees reported increased food consumption after resettling in the (27). Although unfamiliar with American foods, Sudanese refugees reported 75% of total energy from high-protein, high-starch, or convenience foods. They also consumed sugar-sweetened beverages in large quantities and fewer than recommended fruits and vegetables (28). Hmong respondents reported that access to fast and convenience foods changed their dietary habits (23). Patil et al. (16) found that meat and dairy comprised a greater portion of total daily energy in the (12% and 10.4%, respectively) than in Liberia and Somalia (#3% for both). One study found that, given increased availability and accessibility to food, multiple refugee groups consistently overconsumed (including high-fat meats) in the, which was likely associated with past food deprivation in the country of origin (24). For example, Cambodian refugees with higher past food-deprivation scores were more likely to report eating fatty meats (OR: 1.14 for every point increase on the 9- to-27-point food-deprivation measure) (25). This type of unhealthy diet (e.g., convenience and fast foods, fatty meats, and sugar-sweetened beverages) was associated with refugee overweight and obesity (25). Several studies reported refugee concerns about excessive energy intake (30), overweight, and obesity (19, 23 25, 28, 30). Indeed, they were aware of the expanding waistlines in their communities. Only 1 study addressed underweight and vitamin B-12 deficiencies among refugees (20, 21). Perceptions and knowledge of healthy diet. Several studies investigated whether refugees perceived foods in the United Increased food consumption in almost all food categories: Liberian (27) Increased consumption of high-protein and high-starch foods, sugar-sweetened beverages: Sudanese (28) Consumption of convenience and fast foods in large quantities: Sudanese (28), mixed (30) Increased overweight and obesity: Cambodian (25), Hmong (23), Somalian (19), Sudanese (28), mixed (24, 30) Not aware of or wanted to learn the healthy food in the American grocery stores or how to prepare American meals: Burmese (17), Liberian (29), Sudanese (25), mixed (26) Considered that foods or diets improved since arrival in the : mixed (30, 32) Considered their diets less healthy after resettlement in the : Hmong (23), Sudanese (16), mixed (30) Foods perceived as healthy fruits: mixed (30); vegetables: mixed (30, 33); oranges: mixed (33); food items with less fat and/or sugar: mixed (30); small food portion size: mixed (30); boiled water: Hmong (31); whole grains: Cambodian (25); beef: mixed (33); chicken: mixed (33); traditional or native diets: Hmong (23), Sudanese (16) [including Asian fruits, Asian vegetables, Asian herbs: Cambodian (22); including rice: mixed (33)]; fresh, natural, and wholesome ingredients: Hmong (23), mixed (30); minimal use of fat and meat: Hmong (23); without use of preservatives: Hmong (31) Foods perceived as unhealthy high energy food: mixed (24); fat: Cambodian (25); sugar: mixed (30); dairy products: mixed (33); soda: Burmese (17); fast food: Cambodian (25); smoking: Burmese (17); tap water: Hmong (31) Cooking classes, or knowledge of the healthy food items and their recipes in the : Burmese (17), South Asian (32), Liberian (29), Sudanese (28), mixed (26) Understanding of their neighborhood environment: Liberian (29), mixed (24, 26, 27) Knowledge of how to best use resources including food stamps: Cambodian (18), Liberian (29), Montagnard (15) Information about how to participate in exercise: Burmese (17), mixed (30) Health care providers help with their diet: Cambodian (25) Support with the cost of fruits and vegetables: Hmong (23), Somalian (19), Sudanese (16), mixed (30) States compared with home country as healthy and whether they had accurate knowledge about healthy diets (see Table 5). This information may inform future interventions. Healthy foods. Burmese refugees reported neither knowing which foods were healthy in US stores nor how to prepare meals with those foods (17). Multiple refugee groups wanted to learn more about healthy foods in the United States (17, 26, 28, 29). In 2 studies, refugees reported improvements in diets since US resettlement (30, 32); however, more refugees considered US diets less healthy than preresettlement diets (16, 23, 30, 31). Perceived healthy foods included fruits (30), oranges (33), vegetables (30, 33), whole grains (25), beef (33), chicken (33), boiled water (31), and foods with less fat and sugar (30) or small portion size (30). South Asian refugees believed that traditional diets (16, 23) were healthy because they included less meat and more vegetables (23), Asian fruits and vegetables, Asian herbs (18), and rice (33). More than two-thirds of Hmong refugees reported Hmong food to be healthier than American food because of the frequent use of fresh ingredients, minimal use of fat and meat (23), and no preservatives (31). Unhealthy foods. Refugees considered high energy, fat (25), sugar, dairy products (33), tap water (31), soda (17), and fast food (25) to be unhealthy. Further, they believed American food choices were unhealthy because they were less natural or wholesome (30); however, they acknowledged US foods as more cost beneficial (27, 30, 32). One article reported refugees believing dairy products were unhealthy because they can gain fat from dairy products (33). Refugee diet adaptation review 1077

13 Health knowledge-behavior gap. Even when refugees had knowledge about healthy diets and physical activity, they did not necessarily translate knowledge into healthy behavior. One study reported that only 13% of refugees studied felt they ate generally healthy diets in the (30). They also reported difficulties locating preferred foods (24, 27, 29). Lack of healthy food options in the past shaped their dietary habits and food choices poorly after resettlement (24). For example, 32% of Bhutanese refugees had vitamin B-12 deficiencies, likely due to a scarcity of meat, eggs, and dairy in their Nepal diets (20). Participants reported wanting to change diets with health care providers help, but they also indicated they would not voluntarily limit high-fat meat intake (25). In addition, several researchers found that refugees were unfamiliar with preventive health care, although they expressed a desire to improve dietary knowledge (36 38). Insights for future study and intervention. Several studies collected self-reported refugee needs. Data suggested that refugee health and nutrition could be improved by learning: 1) what are healthy foods in the (17, 26, 28, 29), 2) recipes in the (17, 26, 28, 29), 3)foodmarket locations (24, 26, 27, 29), 4) how to participate in exercise (17, 30), and 5) how to utilize resources including food stamps (15, 18, 29). More than half of the Hmong and Cambodian interviewees expressed interest in taking American food cooking classes (32). Future intervention efforts must help educate and empower refugee adaptation to the US food environment. There are limitations in the current literature regarding refugee food intake after US resettlement. First, available studies are limited by small sample size (<300), variable methodological designs (i.e., type and quality), sampling approaches, data collection, and analyses. Most studies obtained limited data, which is likely because of language challenges and funding constraints. Some studies included multiple ethnic groups yet neither reported specific characteristics of the region of origin nor reported results by regional characteristics. Finally, most studies used convenience sampling; thus, findings may not be representative or generalizable. Nevertheless, studies of refugee dietary intake provided important information regarding food intake, dietary changes postresettlement, and beliefs and barriers to healthy nutrition among resettled US refugees. Variables influencing between-group and longitudinal differences in postresettlement food intake and needs included SES, age, sex, food insecurity, past food deprivation, length of time in the United States, and region of origin. Future research with strong methodological designs will help us better understand these unique, diverse, and underserved refugee populations. Our findings have several public health implications. First, there are large between-group differences in dietary intake and needs. Second, changes in food intake and adaptation to US food environments after resettlement vary across age and SES groups. Third, effective and sustainable nutrition and health education programs are needed to help refugees maintain healthy diets and good health. Resettled refugees and their children are likely to become part of the US population. Therefore, promoting health in this diverse, underserved population is a critical component of our national priority to reduce health disparities. Conclusion Our comprehensive examination revealed considerable variation in refugee food intake and postresettlement dietary changes in the. The patterns varied considerably across refugee groups. Future research is needed to investigate associations between refugee diet changes and related risk factors and health consequences. There is a special need for effective nutrition and public health intervention programs for refugees. To effectively target a critical health disparity in the, health promotion programs will need support and coordination from multiple sectors of American society. Acknowledgments We thank Hong Xue for his important contribution in the early stage of the study. Kisa Harris and Jacob Khuri are summer students who received training and conducted research under Youfa Wang s guidance at University at Buffalo, The State University of New York. All authors read and approved the final version of the manuscript. References 1. The UN Refugee Agency. Facts and figures about refugees [Internet]. [cited 2015 Jul 31]. Available from: 2. The White House. Presidential memorandum: fiscal year 2015 refugee admissions [Internet]. [cited 2015 Sept 8.] Available from: whitehouse.gov/the-press-office/2014/09/30/presidential-memorandum-fy refugee-admissions. 3. Bhatta MP, Assad L, Shakya S. Socio-demographic and dietary factors associated with excess body weight and abdominal obesity among resettled Bhutanese refugee women in Northeast Ohio,. Int J Environ Res Public Health 2014;11: Gordon-Larsen P, Harris KM, Ward DS, Popkin BM. Acculturation and overweight-related behaviors among Hispanic immigrants to the US: the National Longitudinal Study of Adolescent Health. Soc Sci Med 2003;57: Dookeran NM, Battaglia T, Cochran J, Geltman PL. Chronic disease and its risk factors among refugees and asylees in Massachusetts, Prev Chronic Dis 2010;7:A Yun K, Hebrank K, Graber LK, Sullivan MC, Chen I, Gupta J. High prevalence of chronic non-communicable conditions among adult refugees: implications for practice and policy. J Community Health 2012; 37: Gordon AM Jr. Nutritional status of Cuban refugees: a field study on the health and nutriture of refugees processed at Opa Locka, Florida. Am J Clin Nutr 1982;35: Geltman PL, Radin M, Zhang Z, Cochran J, Meyers AF. Growth status and related medical conditions among refugee children in Massachusetts, Am J Public Health 2001;91: Culhane-Pera KA, Her C, Her B. We are out of balance here : a Hmong cultural model of diabetes. J Immigr Minor Health 2007;9: Tropp LR, Erkut S, Coll CG, Alarcon O, Vazquez Garcia HA. Psychological acculturation: development of a new measure for Puerto Ricans on the U.S. mainland. Educ Psychol Meas 1999;59: Hadley C, Zodhiates A, Sellen DW. Acculturation, economics and food insecurity among refugees resettled in the USA: a case study of West African refugees. Public Health Nutr 2007;10: Wang et al.