Adolescent risk behaviours and psychological distress across immigrant generations.

Hamilton, Hayley A. ; Noh, Samuel ; Adlaf, Edward M. 等

Foreign-born children and children with foreign-born parents represent an estimated 20% of individuals under the age of 18 in Canada, with a projected increase to 25% by 2016. (1) These figures suggest that identifying and addressing differences in health and behaviour between foreign-born and native-born children are critical to public health. It is well known that the adolescent years are a precarious developmental stage during which psychological difficulties and health risk behaviours can influence an individual's future life success. These years may be particularly dramatic for foreign-born youth and native-born youth with foreign-born parents as many are caught between upholding the traditional values from home countries that are often stressed by parents, and engaging in behaviours that will hasten their acceptance into a particular peer group.

Research suggests that as immigrants become more acculturated, they adopt the attitudes, social norms, and behaviours of the mainstream or a social reference group within their new country. (2,3) Whether increased acculturation results in positive or negative outcomes, however, remains unknown to a large extent because of inconsistent findings. Some early studies indicated that the foreign-born were at increased risk for psychological and behavioural difficulties, (4-7) suggesting that with time and increased acculturation, such difficulties would decline to that of the mainstream. More recent studies, however, have often challenged that view with findings that foreign-born children are often at similar (8,9) or reduced risk of psychological and behavioural difficulties relative to their native-born counterparts. (10-14) A more nuanced view coincides with indications that health and behavioural differences between foreign and native-born children are complex and thus may vary, for example, across health outcomes and between multiple immigrant generations.

The main objective of this study is to examine differences in psychological distress symptoms, hazardous and harmful drinking, illicit drug use, and general delinquency between three immigrant generations. The study focuses on adolescent students in Ontario, the province of residence for 54.9% of the foreign-born population in Canada and 38% of the overall Canadian population. (15) This study also examines the moderating effects of age and sex in the association between immigrant generation and each outcome given that norms associated with adolescent behaviour are often driven by age and sex. This may be particularly so for first-generation youth who are often from countries with more traditionally defined roles. Alternatively, sex and age differences may be more evident among second and later generations because of greater acculturative influences.

METHODS

Data

Data were derived from the 2005 Ontario Student Drug Use Survey (OSDUS), a province-wide survey of 7th to 12th grade students with in regular public or Catholic schools. (16) The survey was administered in classrooms through anonymous, student-completed questionnaires. Conducted every odd year since 1977, OSDUS employs a two-stage cluster design involving a random selection of classes from within a random selection of schools (probability proportional to size) stratified by region and school type (elementary/middle or secondary). The 2005 total sample was 7,726 students from 42 school boards, 137 schools, and 445 classrooms. Completion rates were 94% and 72% for schools and students, respectively. Absenteeism (12%) and lack of parental consent (16%) were among the reasons for non-completion. Specific items important to this analysis (e.g., delinquency) were only asked of a random half sample of 4,078 students, and thus this half sample is the basis for analyses. A total of nine respondents who were younger than 12 or older than 19 years of age were excluded from analyses. The Research Ethics Board of the Centre for Addiction and Mental Health approved this study. Further details regarding the study design are available at: http://www.camh.net/research/population_ life_course.html (16) (Accessed March 17, 2008).

Measures

Health risk behaviours examined are hazardous and harmful drinking, illicit drug use, and delinquency. Hazardous and harmful drinking is based on the Alcohol Use Disorders Identification Test (AUDIT), an instrument developed by the World Health Organization. (17) AUDIT assesses drinking behaviour that increases the likelihood of current or future physical health problems (e.g., accidents, alcohol-related injuries). Psychological distress is based on the General Health Questionnaire, a general measure of emotional distress or malaise. (18)

The independent measure, immigrant generation, is represented by dummy variables contrasting three groups. First-generation immigrants are foreign-born youth. Second-generation immigrants are native-born youth (i.e., Canadian-born) with at least one foreign-born parent. Third- and later-generation immigrants are native-born youth with native-born parents.

Analyses control for biological sex, respondents' age, family structure, parental education, and urbanicity. "Don't know" responses on parental education are included because they represent 10% of the sample. Analysis indicates that don't know respondents on this measure tend to be younger (61% are age 12-14 and 73.5% are age 12-15) than those who know their parents' education. Further descriptions of variables used in analyses are provided in the appendix.

Analyses

Given the complex sample design, Taylor series methods within Stata are used to compute unbiased variances, standard errors, and point estimates. (19) Analyses are weighted to adjust for the unequal probability of selection. (16) Ordinary least squares (OLS) regression is used to examine the relationship between immigrant generation and each of hazardous and harmful drinking, delinquency, and psychological distress. Logit regression is used to examine the relationship between immigrant generation and illicit drug use. Separate multiplicative interactions involving immigrant generation and age, and immigrant generation and sex are examined for each outcome. Only significant interactions are presented in the regression table.

Both linear and quadratic age terms are included in the OLS models to control for the possible nonlinear effect of age on the dependent variables. Age variables are centered in order to reduce the correlation between the linear, quadratic, and interaction terms. (20) OLS regressions are based on square-root transformations of drinking and delinquency scores.

RESULTS

Descriptives of the sample are provided in Table 1. First-, second-, and third-generation immigrant youth represent 16.3%, 30.9%, and 52.8% of the sample, respectively. A greater proportion of the first-generation sample resides in urban rather than rural areas, and has parents with a university degree.

Table 2 outlines the mean or percentage level of harmful drinking, drug use, delinquency, and distress by immigrant generation and control variables. There are statistically significant variations in hazardous and harmful drinking and illicit drug use across the three immigrant generations. First-generation youth report less harmful drinking on average and are less likely to use illicit drugs than second-generation youth, while second-generation youth report less drinking and drug use than their third-generation counterparts. In contrast, the main difference in delinquency and distress across immigrant generations is between first and second generations, with no significant difference between second-generation and third-generation youth. However, whereas mean delinquency is lower, symptoms of psychological distress are greater among first-generation than second-generation youth.

Results from multivariate regression analyses are outlined in Table 3. Results for harmful drinking indicate significant differences between immigrant generations in levels of drinking after adjustments for select socio-demographic factors. Such differences, however, are moderated by age as evident by the statistically significant coefficient for the age by first-generation interaction term. An illustration of this interaction (Figure 1) shows there are increasing differences in drinking between generations with increasing age, particularly between first and second generations. The curvilinear nature of the relationship is also evident as there is some tapering off in drinking in late adolescents, particularly among the first generation.

As with harmful drinking, the odds of illicit drug use increase across immigrant generations. Compared to second-generation youth, the odds of drug use are actually 1.5 times greater among third and later generations and .51 times less among first-generation youth after adjusting for socio-demographic factors. Unlike harmful drinking, however, the relationship between immigrant generation and drug use did not significantly vary by age, although age has an influence on drug use.

Results for delinquency and distress indicate that first-generation youth engage in less delinquent activities and report greater symptoms of psychological distress than second-generation youth, controlling for socio-demographic characteristics. In addition, there are no significant differences in delinquency or distress between second- and third-generation youth. Neither age nor sex is a significant moderator.

DISCUSSION

The main objective of this study was to examine differences in psychological and behavioural outcomes among adolescents of diverse immigrant generations in Ontario. A particular strength of this study is that the data represent a school-based sample of adolescent students in a region with the highest concentration of immigrants (approximately 55%) in Canada. Given the high concentration of immigrants within the country, and particularly the province of Ontario, the well-being of first- and second-generation immigrant students is of particular importance on multiple levels. Results of this study highlight the complex nature of the relationship between immigrant generation and health and behavioural outcomes. Although foreign-born adolescents, relative to their native-born counterparts, report more symptoms of psychological distress, they report fewer health risk behaviours. This suggests that the nature of differences between foreign- and native-born adolescents varies across outcomes. Also significantly, results indicate that psychological and behavioural outcomes do not consistently improve or deteriorate across immigrant generations.

Findings with regard to symptoms of psychological distress are consistent with some earlier studies that found greater psychological distress among the foreign-born. (9,21) There are other studies, however, that found less distress among the foreign-born (10,11,14,22) or no difference in distress between foreign- and native-born.8 Although studies by both Ali (22) and Beiser et al. (10) involved Canadian data on a national level, the ages of the sample were different, with the former focusing on individuals 15 and older and the latter focusing on 4-11 year olds.

Findings indicating greater delinquent activities among the native-born are generally consistent with earlier studies, (8,12,23) although this study indicates little difference in delinquency between native-born youth, specifically those with foreign-born compared to native-born parents. Results with regard to substance use are generally consistent with other studies indicating greater prevalence rates of substance use among the native-born. (24-27)

Important limitations of the study should be noted. First, data are unavailable on respondents' length of residence or age at arrival in Canada, thus an important aspect of acculturation cannot be considered in analyses. Second, data are unavailable on the race or ethnicity of the sample. The ethnic diversity of the population, especially in urban areas of the province, and the likely differences in racial and ethnic composition of the different immigrant generations suggest that unobserved differences in composition may account for some of the generational differences reported. Third, the sample was restricted to students within the regular school systems and, therefore, does not represent approximately seven percent of students. (16) Additionally, the student non-completion rate due to absenteeism and lack of parental consent may have biased the sample to some degree.

The increasing proportion of first- and second-generation immigrant youth in middle and high schools means an increase in the proportion of students adapting to a new country, struggling with identities, and struggling between old and new values and ways of life, at the same time that they are challenged by the usual barrage of issues encountered during the adolescent years. Given that a successful transition into adulthood is dependent on successfully maneuvering through adolescence, it is important to acquire knowledge about the adjustment levels of these two important population groups. Such knowledge is necessary to develop policies and programs aimed at curbing and preventing maladjustment and promoting the health of individuals, families, and the systems upon which they depend.

[FIGURE 1 OMITTED]

Appendix Description of variables used in analyses

Variable           Description                     Coding

Immigrant          First generation                Dummy variable
generation                                         (1=yes, 0=no)
                   Second generation               Reference
                                                   category
                   Third and higher                Dummy variable
                   generation                      (1=yes, 0=no)
Age                Age in years (12-19)            Mean centred
Sex                Biological sex                  Female=1, male=0
Family structure   Current living arrangements     Two biological/
                                                   adoptive parents=1,
                                                   other=0
Parental           At least 1 parent with a        Dummy variable
education                                          (1=yes, 0=no)
                   university degree               Reference category
                   No parent with a                Dummy variable
                   university degree               (1=yes, 0=no)
                   Don't know education
                   of parents
Urbanicity         Residence in urban              Rural residence=1,
                   or rural area                    urban=0
Dependent
measures
  Hazardous and    Alcohol Use Disorders           Summed responses
  harmful          Identification Test used        to 10 items (0-34,
  drinking         to identify problem drinkers    median=1)
                   with the least severe alcohol
                   use. Higher scores indicate
                   more harmful drinking
  Illicit          Use of any of 11 illicit        Use of at least 1
  drug use         drugs (excluding inhalants,     illicit drug
                   meds, and club) during          (1=yes, 0=no)
                   a 12-month period
  Delinquency      Participation in 13             Score derived by
                   delinquent acts (e.g.,          averaging responses
                   damaged property, theft,        across at least 7
                   assault, breaking and           of 13 delinquency
                   entering, carrying a             items
                   weapon). Higher scores
                   indicate greater delinquency
  Psychological    12-item General Health          Responses to at
  distress         Questionnaire. Higher scores    least 6 items
                   indicate higher distress        were averaged
                                                   to derive a score

Received: April 11, 2008

Accepted: January 9, 2009

REFERENCES

(1.) Canadian Council on Social Development. The Progress of Canada's Children and Youth. Ottawa, ON: CCSD, 2006.

(2.) Gil AG, Vega WA, Dimas JM. Acculturative stress and personal adjustment among Hispanic adolescent boys. J Community Psychol 1994;22:43-55.

(3.) Vega WA, Amaro H. Latino outlook: Good health, uncertain prognosis. Annu Rev Public Health 1994;15:39-67.

(4.) Portes A, Rumbaut RG. Immigrant America: A Portrait. Berkeley, CA: University of California Press, 1996.

(5.) Malzberg B, Lee ES. Migration and Mental Disease: A Study of First Admissions to Hospitals for Mental Disease, New York, 1939-1941. New York, NY: Social Science Research Council, 1956.

(6.) Rutter M, Yule W, Berger M, Yule B, Morton J, Bagley C. Children of West Indian immigrants: Rates of behavioral deviance and psychiatric disorder. J Child Psychol Psychiatry 1974;15:241-62.

(7.) Bagley C. Deviant behavior in English and West Indian school children. Res Educ 1972;8:47-55.

(8.) Hamilton HA. Health and Well-Being Among Immigrant Youth. New York: LFB Scholarly, 2005.

(9.) Kao G. Psychological well-being and educational achievement among immigrant youth. In: Hernandez DJ (Ed.), Children of Immigrants: Health, Adjustment, and Public Assistance. Washington, DC: National Academy Press, 1999;410-77.

(10.) Beiser M, Hou F, Hyman I, Tousignant M. Poverty, family process, and the mental health of immigrant children in Canada. Am J Public Health 2002;92:220-27.

(11.) Harker K. Immigrant generation, assimilation, and adolescent psychological well-being. Social Forces 2001;79:969-1004.

(12.) Harris KM. Health status and risk behaviors of adolescents in immigrant families. In: Hernandez DJ (Ed.), Children of Immigrants: Health, Adjustment, and Public Assistance. Washington, DC: National Academy Press, 1999;286-347.

(13.) Rumbaut RG. Paradoxes (and orthodoxies) of assimilation. Sociol Perspect 1997;40:483-511.

(14.) Rumbaut RG. Ties that bind: Immigration and immigrant families in the United States. In: Booth A, Crouter AC, Landale N (Eds.), Immigration and the Family: Research and Policy on U.S. Immigrants. Mahwah, NJ: Lawrence Erlbaum Associates, 1997;3-46.

(15.) Chui T, Tran K, Maheux H. Immigration in Canada: A Portrait of the Foreign-Born Population, 2006 Census (Statistics Canada, Catalogue 97-557-XIE). Ottawa, ON: Statistics Canada, 2007.

(16.) Adlaf EM, Paglia-Boak A. Drug Use Among Ontario Students, 1977-2005: Detailed OSDUS Findings (CAMH Research Document Series No. 16). Toronto, ON: Centre for Addiction and Mental Health, 2005.

(17.) Saunders JB, Aasland OG, Babor TF, De La Fuente JR, Grant M. Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption. Addiction 1993;88:791-804.

(18.) Goldberg DP. The Detection of Psychiatric Illness by Questionnaire. Oxford, UK: Oxford University Press, 1972.

(19.) Stata Corp. Stata Statistical Software: Release 9.0. College Station, TX: Stata Corp, 2005.

(20.) Aiken LS, West SG. Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage, 1991.

(21.) Rumbaut RG. The crucible within: Ethnic identity, self-esteem, and segmented assimilation among children of immigrants. Int Migration Rev 1994;28:748-94.

(22.) Ali J. Mental health of Canada's immigrants. Supplement to Health Reports (Statistics Canada, Catalogue 82-003), 2002;13.

(23.) Gibson MA, Ogbu JU. Minority Status and Schooling: A Comparative Study of Immigrant and Involuntary Minorities. New York: Garland Publishing Inc., 1991.

(24.) Gfroerer JC, Tan LL. Substance use among foreign-born youths in the United States: Does the length of residence matter? Am J Public Health 2003;93:189295.

(25.) Vega WA, Gil AG, Zimmerman RS. Patterns of drug use among Cuban-American, African-American, and white non-Hispanic boys. Am J Public Health 1993;83:257-59.

(26.) Blake SM, Ledsky R, Goodenow C. Recency of immigration, substance use, and sexual behavior among Massachusetts adolescents. Am J Public Health 2001;91:794-98.

(27.) Canadian Council on Social Development. Immigrant Youth in Canada, 2007. Available online at: http://www.ccsd.ca/subsites/cd/docs/iy/index.htm (Accessed May 24, 2007).

Hayley A. Hamilton, PhD, [1,2] Samuel Noh, PhD,[1,2] Edward M. Adlaf, PhD [1-3]

Author Affiliations

[1.] Centre for Addiction and Mental Health, Toronto, ON

[2.] Department of Psychiatry, University of Toronto, Toronto, ON

[3.] Dalla Lana School of Public Health, University of Toronto, Toronto, ON

Correspondence and reprint requests: Hayley Hamilton, Social Equity and Health Research, Centre for Addiction and Mental Health, 455 Spadina Ave., Suite 300, Toronto, ON M5S 2G8, E-mail: hayley_hamilton@camh.net

Table 1. Variable Means/Percentages, Overall and by
Immigrant Generation

                  First         Second        Third         Overall
                  Generation    Generation    Generation    Sample

Age               15.2 (a)      14.8 (b)      14.9          14.9
                  (14.7-15.6)   (14.6-15.1)   (14.8-15.1)   (14.8-15.1)
Female            46.4          47.0          50.7          48.9
                  (39.3-53.7)   (42.5-51.6)   (47.4-54.0)   (45.8-52.0)
Two biological    73.3          74.0          69.4          71.5
parents           (67.2-78.6)   (70.1-77.6)   (66.4-72.3)   (69.1-73.7)
Urban residence   99.3 (a)      93.8 (b)      75.5 (c)      85.0
                  (97.7-99.8)   (90.0-96.2)   (67.4-82.2)   (79.3-89.4)
Parental
education
  University      55.6 (a)      38.1 (b)      38.2 (c)      41
  degree          (48.5-62.5)   (33.9-42.5)   (34.6-42.0)   (37.6-44.5)
  Less than       33.1          47.5          52.8          48.0
  university      (27.5-39.2)   (43.5-51.5)   (49.2-56.4)   (44.8-51.1)
  Don't know      11.3          14.4          8.9           11.0
                  (8.2-15.2)    (11.9-17.4)   (7.4-10.8)    (9.6-12.6)
N                 564           1130          2322          4016

Note: 95% confidence intervals are shown in brackets below means.
Across a single row, means and percentages with different superscript
letters indicate a statistically significant group (generation)
difference at p<0.05 level.

Table 2. Variable Means/Percentages by Adolescent Outcomes

                Harmful         Illicit       Delinquency   Distress
                Drinking        Drug Use      Mean          Mean
                Mean (95% CI)   % (95% CI)    (95% CI)      (95% CI)

Immigrant
generation
  1st           1.9 (a)         17.1 (a)      .06 (a)       .20 (a)
  generation    (1.4-2.3)       (13.2-21.7)   (.04-.07)     (.18-.23)
  2nd           2.7 (b)         24.7 (b)      .08 (b)       .17 (b)
  generation    (2.3-3.1)       (20.7-29.1)   (.07-.09)     (.15-.18)
  3rd& later    4.1 (c)         33.2 (c)      .08b          .16 (b)
  generation    (3.6-4.5)       (30.5-36.0)   (.07-.09)     (.15-.17)

Age 15          3.3             30.8          0.09          0.17
                (2.8-3.8)       (27.0-34.7)   (.07-.11)     (.15-.19)

Sex
  Female        3.1             27.4          .09 (a)       .22 (a)
                (2.8-3.5)       (24.6-30.3)   (.08-.11)     (.20-.23)
  Male          3.4             28.5          .06 (b)       .13 (b)
                (2.9-3.9)       (25.0-32.2)   (.05-.07)     (.12-.14)

Parental
structure
  Two           3.0 (a)         25.4 (a)      .07 (a)       .16 (a)
  biological    (2.7-3.4)       (23.0-27.9)   (.06-.08)     (.15-.17)
  parents
  Other         3.9 (b)         34.3 (b)      .10 (b)       .21 (b)
                (3.2-4.5)       (29.5-39.4)   (.09-.12)     (.19-.22)

Area of
residence
  Urban         3.2             27.6          0.08          0.17
                (2.7-3.6)       (24.5-31.0)   (.07-.09)     (.16-.18)
  Rural         3.9             29.8          0.08          0.17
                (3.3-4.5)       (24.3-36.0)   (.07-.10)     (.14-.19)
Parental
education
  University    3.0 (a)         24.5 (a)      .07 (a)       .16 (a)
  degree        (2.5-3.4)       (20.9-28.4)   (.06-.08)     (.14-.17)
  Less than     3.9 (b)         34.7 (b)      .09 (b)       .19 (b)
  university    (3.4-4.3)       (31.4-38.1)   (.08-.10)     (.17-.20)

  Don't know    1.6 (c)         11.3 (c)      .05 (c)       .15 (a)
                (1.2-2.0)       (8.5-15.6)    (.03-.06)     (.12-.17)

N               3882            4012          3938          3999

Note: Within a given column and category of predictor
variable (e.g., sex), means and percentages with different
superscript letters indicate a statistically
significant group difference at p<0.05 level.

Table 3. Psychological and Behavioural Outcomes Regressed on
Immigrant Generation and Socio-demographic Factors

                  Hazardous     Illicit     Delinquency   Psychological
                  and Harmful   Drug Use    (a)          Distress
                  Drinking (a)  (n=4012)    (n=3938)     (n=3999)
                  (n=3882)      O.R. (95%   b (s.e.)      b (s.e.)
                  b (s.e.)      CI) (b)
Immigrant
generation (c)
  1st             -.343 ***      .51          -.038 **     .035 **
  generation      (.067)         (.37-.71)    (.014)       (.012)
  3rd & later     .328 ***       1.5          .015         -.013
  generation      (.070)         (1.2-1.8)    (.009)       (.009)
Female            -.027          .93          -.061 ***    .089 ***
                  (.053)         (.78-1.1)    (.009)       (.010)
Age (d)           .287 ***       1.6          .016 ***     .018 ***
                  (.022)         (1.5-1.7)    (.002)       (.003)
Age (d)-squared   -.021 **                    -.005 ***    -.003 *
                  (.006)                      (.001)       (.001)
Two biological    -.171 **       .70          -.050 ***    -.045 ***
parents
Parental          (.059)         (.59-.83)    (.011)       (.009)
education (e)
  University      -.047          .79          -.017        -.021 *
                  (.050)         (.65-.96)    (.010)       (.010)
  Don't know      -.326 ***      .40          -.055 ***    -.016
                  (.055)         (.28-.56)    (.013)       (.015)
Interactions
  Age x 1st       -.113 **
  generation      (.035)
  Age x 3rd       .044
  generation      (.028)

Constant          1.429                       .253         .182

[R.sup.2]         .261                        .064         .073

*** p<0.001; ** p<0.01; * p<0.05.

Rural residence is non-significant and is excluded from table.
Unstandardized coefficients.

(a) Square root transformed measures.

(b) Odds ratios with 95% confidence intervals in brackets.

(c) Reference category is second generation. As such, the hazardous
and harmful drinking predicted score of 15 year old students is
.12 (.3432) less among first-generation than second-generation youth
adjusting for socio-demographic factors.

(d) Age is centered.

(e) Reference category is less than university education.