Immigration and Stress: The Relationship Between Parents' Acculturative Stress and Young Children's Anxiety Symptoms

Specific Aims

Specific Aim 1: To examine the relationship between theoretical risk factors of mental illness surrounding immigration (i.e. acculturative stress) and parent-reported child anxiety symptoms on the Brief Infant-Toddler Social and Emotional Assessment (B-ITSEA). Specifically, the relationship between measured stress factors related to acculturation (i.e. self-reported level of English language competency, self-reported level of discrimination based on race, appearance, and language and/or ethnicity) in the parent and measured anxiety symptoms in the child will be examined. This relationship will be evaluated in children between the ages of 11 months and 5 years of age. This group will consist of immigrant children (first-generation immigrants) and children of immigrant parents (second-generation immigrants).

Hypothesis 1: Due to theoretical risk factors of mental illness within immigrant families, it is hypothesized that higher scores of stress factors related to acculturation in the parent will predict an increased level of anxiety symptoms in the child.

Specific Aim 2: To examine how the parent-reported child anxiety symptoms of first-generation and second-generation immigrant children compare to the measured anxiety symptoms of U.S. born children (non-immigrant families). The latter group will consist of children and parents both born in the U.S.

Hypothesis 2: As supported by previous research on the sources of risk for mental illness within immigrant families (i.e. acculturative stress), it is hypothesized that first-generation and second-generation immigrant children will show an increased level of anxiety symptoms, relative to children born in the U.S.

Methodology

Participants

One hundred and forty-eight families from the University of Massachusetts Boston Child and Family Development Project (CFDP) were assessed in the present study. The CFDP is a research program designed to identify children at high-risk for developing emotional problems and investigating factors that relate to parent engagement in preventive services. Participants were recruited from two Women, Infants, and Children (WIC) offices in Boston, Massachusetts. Offices are located in health centers within urban neighborhoods characterized by low socioeconomic status, high rates of community violence, and large proportions of ethnic minority populations. WIC programs provide nutritional assistance to low-income, pregnant, and postpartum women with children up to age 5. In order to be eligible for the study, WIC parents needed to meet three criteria: being a caregiver of a child within the age range of 11 months and 5 years, being at least 18 years old, and being able to respond to the survey in English or Spanish.The vast majority of respondents were biological mothers (88%). The sample was evenly divided by child sex (67 males and 76 females). The parents^[1] ranged in age from 18 - 65 (mean = 29.19; SD = 7.53) and the children from 10 – 69 months (M = 34.76). The majority of parents reported annual incomes in the $0 - $15,000/year range (52%). Only 32% of the sample reported annual incomes greater than $25,000.

Sixty-three parents were immigrants and 66 were US born. A significant percentage (13%) of parents did not indicate birthplace. Due to the small number of first-generation immigrant children (n = 8), first-generation and second-generation immigrant children (n = 55) were collapsed into one group (n = 63; 30 males and 31 females). The subgroup of non-immigrant children consisted of 30 boys and 36 females. The majority of parents were born in the US (n = 66), followed by Dominican Republic (19), Cape Verde (16), Puerto Rico (8), Jamaica (3), and Haiti (2). There was also one parent born in each of the following countries; Angola, Cameroon, Ghana, Liberia, Nigeria, Guatemala, Honduras, Venezuela, Mexico, Virgin Islands, Barbados, Spain, Portugal, and Vietnam.The majority of children were born in the US (n = 130), followed by Cape Verde (2), and Dominican Republic (2). There was also one child born in each of the following countries; Puerto Rico, Haiti, Sudan, and Finland. To explore the different ethnicities in the sample, parents were asked to report which ethnic group they belong to, in their own words. The ethnicities reported were diverse, including African American (16%), Latino/Latina (16%), Hispanic (10%), Cape Verdean (7%), Black (5%), Dominican (5%), Puerto Rican (3%), African (2%), and West Indian (2%). Parents also reported their race African American/Black (36%), Hispanic/Latino/Latina (26%), White (7%), and Cape Verdean (6%). Parents were also asked, “What ethnic group(s) does your child belong to?” The largest number of children were reported to be African American (14%); followed by Hispanic (8%), Black (8%), Latino/Latina (7%), Cape Verdean (5 %), Dominican (3%), none/other/not sure (3%), Latino/Latina American (1%), White (1%), Native American (1%), Hispanic/Black (1%), and Puerto Rican (1%). Child race was also reported by parents: African American/Black (33%), Hispanic/Latino/Latina (22%), White (6%), and Cape Verdean (4%). A third of parents indicated English (34%) as their native language, followed by Spanish (31%), Cape Verdean/Portuguese Creole (8%), English and Spanish (2%), Haitian Creole (2%), and Portuguese (2%). The survey was offered in either English or Spanish; 120 parents completed the survey in English and 27 parents completed it in Spanish. The survey was also offered in two forms: on paper or on a touch screen tablet computer; 75% of parents completed the survey online.

Procedures

Families were approached and recruited in person by bilingual research staff in the waiting area of WIC offices. Parents were briefly introduced to the project and asked if they met eligibility for the study. Eligible parents were given a brief overview of the nature of the study, the duration of the survey (20-30 minutes), and the compensation that was offered for their participation. Following the agreement of eligible parents to participate, parents provided informed consent. Once informed consent was obtained, parents were asked which format (portable tablet or paper) they preferred to take the survey on and which language version of the survey they preferred (English or Spanish). When both parents were present, it was made clear to parents that only one parent could provide responses and those responsesshould be about only one child. Parents who selected the portable tablet format were given a short tutorial of how to use the device in order to ensure competence in completing the survey. A bilingual research assistant was available throughout their participation to address any questions or issues that emerged during the course of completing the questionnaire. Following completion, each parent was given $10.00 and was offered to select one toy for his/her child. All procedures of this study were approved by the Institutional Review Board (IRB) at the University of Massachusetts Boston.

Measures

Sociodemographic information

Parents provided responses to a range of questions relevant to providing a detailed description of the sample. Each participant was asked to indicate where they were born, where their child was born, number of years lived in the US, number of years their child has lived in the US, the racial and ethnic group they belong to, the racial and ethnic group their child belongs to, and their native language. Other information was obtained on family structure, income, and education level.  

Immigrant status

To identify the immigrant status of the parent and his/her child, each parent was asked, “Where was your child born?” and “Where were you born?” Parents indicating birth outside of the US were identified as first-generation immigrants. Parents indicating that their child was born outside of the US determined first-generation immigrant status in the child. Second-generation immigrant status in the child was identified by a parent response of being born outside of the US and their child being born in the US. For a parent or child to be identified as non-immigrant, the parent must have indicated that both s/he and her/his child were born in the US.

Parent acculturative stress

Acculturative stress was assessed using a composite score consisting of level of competency in English, the amount of discrimination experienced, and exposure to immigration-related stress. Participants were asked, “How well do you speak English?” and rated their level of competence on a 5-point scale (4 = not at all, 3 = very little, 2 = moderately, 1 = pretty well, 0 = very well). Participants were also asked, “How much discrimination have you experienced due to race?” and “How much discrimination have you experienced due to your language and/or ethnicity?” In response to both questions, participants rated the amount of discrimination experienced on a 4-point scale (0 = none, 1 = very little, 2 = some, 3 = a lot). Finally, participants were asked if they or their child was exposed to a “stressful event related to immigration.” They could select one or more of the following; does not apply, it happened to me, I saw it or heard it happening to someone else, it happened to my child, my child saw or heard it happening to someone else. Responses were weighted to capture differences in personal influence for different types of exposures (parent exposure = 2; child exposure = 1; parent or child witnessing = 0.5). Responses were then summed; total responses on this item could range from 0-4. Each selection was included in the sum, except does not apply, and the sum of these selections determined the participant’s score on this question. The composite score was derived from the sum of the four sub-scores in this scale: level of competency in English, the amount of discrimination that they have experienced due to language/ethnicity and race, and their exposure to a stressful event related to immigration. The range of the composite score was from 0 to 14 points. As expected, this scale demonstrated low internal reliability in the overall sample (α = .39). The internal reliability of this scale was higher for non-immigrants (α = .49) than for immigrants (α = .46). While the concept of acculturative stress is typically applied to immigrant populations, the way it is operationalized here is largely relevant to both immigrants and non-immigrants, and it is examined in both groups.

Parent anxiety symptoms

Although parent anxiety was not included in hypotheses, it was included in analysis to further investigate the effects of acculturative stress. Symptoms of anxiety in the parent were assessed using the Beck Anxiety Inventory; a widely-used self-report measure of physiological and cognitive symptoms of anxiety (Beck, Epstein, Brown, & Steer, 1988). Each parent was asked to rate the degree to which he or she has “been bothered by these feelings” (e.g., nervous) in the past week. The BAI has 21 items with answers given on a 4-point scale of 0 (not at all) to 3 (severely). Internal consistency is excellent (α = .92) and 1-week test-retest reliability is good (r = .75). A cut-score of 10 was used (Leyfer, Ruberg, & Woodruff-Borden, 2006).

Child anxiety symptoms

The Brief Infant-Toddler Social and Emotional Assessment (BITSEA) is a 42-item parent-report, developmentally appropriate screening questionnaire that measures social-emotional/behavioral problems and competencies in 12-36 month-old children (Briggs-Gowan & Carter, 2006). Items are rated on a 3-point scale; from 0 = Not true/Rarely to 2 = Very True/Often. The BITSEA has demonstrated good test-retest (r = .92) and inter-rater (r = .74) reliability and convergent validity with other measures and observation. Anxiety symptoms in the child were assessed on a 15-item Anxiety Risk Subscale. The CFDP devised this subscale, which consists of 8 items from the BITSEA and 7 items from the full ITSEA, in order to more fully assess anxiety symptoms. Parents rated each item and the sum of each rated item determined the child’s anxiety symptoms score. The Anxiety Risk Scale demonstrated good internal reliability in the larger sample used in the CFDP (Cronbach’s α = .65).

Protection of Subjects

Before participating in the study, each parent was informed about risks or discomforts s/he may experience, confidentiality, voluntary participation, and compensation as part of the consent form. Each parent was informed that participating in the study poses minimal risk and should they feel uncomfortable with any of the questions, they are free not to answer. Each parent was informed that their participation is confidential and explained that the information gathered for this project will not be stored, published, or presented in a way that would allow anyone to identify them or their child. Each parent was informed their answers will be kept separately from their name and any contact information that they provided using an identification number. All electronic information was temporarily stored in a secure online database (www.psychdata.com) and all paper information was stored in a locked file cabinet which only members of the research team had access to. Each parent was informed that their participation in the project is voluntary and that they may withdraw their participation at any time. The CFDP is approved by the Institutional Review Board (IRB) at the University of Massachusetts Boston.

Data Analysis

All of the data in the study was organized and analyzed using Statistical Package for the Social Sciences (SPSS). A Pearson correlation was used to test the first hypothesis of the study that higher scores of acculturative stress in the parent will predict an increased level of anxiety symptoms in the child. A T-test was used to test the second hypothesis that first-generation and second-generation children will show an increased level of anxiety symptoms, relative to children born in the US. A hierarchal regression analysis was used to test for acculturative stress as a moderator for parent immigrant status and symptoms of anxiety in the child.Continued on Next Page »