Racial/ethnic differences in the prevalence of internalizing symptoms: Do Latin-American immigrant show more symptomatology than Spanish native-born adolescents?

Abstract

This study mainly compared the prevalence of internalizing symptoms of 834 Spanish and 159 Latin-American immigrant adolescents. Participants completed self-report measures about depression, anxiety and somatic symptoms and a socio-demographic questionnaire. The results indicated that being Latin-American was associated with higher levels of depressive symptoms and being female was related to higher depressive and anxiety symptoms. Gender differences were more prevalent in Spaniards than in Latinos, with girls showing more symptoms than boys. High socio-economic status was negatively related to depressive symptoms and anxiety. The results may alert clinicians of the importance of assessing depressive symptoms in Latino adolescents in order to treat this group of youths effectively.

Keywords

adolescence anxiety depression ethnicity physical symptoms

The United States has received people from the entire world every year for many decades, especially from Latin-American countries. Nevertheless, migration from Latin-America to Spain has been increasing since approximately two decades ago. According to the National Statistical Institute (Instituto Nacional de Estadística (INE), 2010), 11.4 per cent of the Spanish population are immigrants. Only in 2010, the flow of immigration from Latin-America was 21,910 individuals (aged between 13 and 16 years) (INE, 2010). According to a report of the International Organization for Migration (IOM, 2004), in the case of other European countries, the presence of Latin-Americans is small relative to other immigrant groups. Latin-Americans speak the same language as Spaniards; perhaps, this is the reason why Spain is an attractive destination for them.

There is evidence of ethnic/racial differences in the self-report of psychopathology, stress and health conditions in different cultural groups such as Latin-Americans, African-Americans and White-Americans (Courtenay et al., 2002; Jason et al., 1999; Ostrove et al., 1999; Williams et al., 1999). In effect, Latino immigrants in the US report more psychopathologies than do US-born adolescents; above all, more internalizing symptoms (McLaughlin et al., 2007; Umaña-Taylor and Updegraff, 2007). They have shown higher levels of somatic, anxiety and depressive symptoms than have shown non-Latino White participants, and they tend to report these symptoms as distressing (Canino, 2004; McLaughlin et al., 2007; Pina and Silverman, 2004). To our knowledge, there are few studies in Europe about the prevalence of internalizing symptoms in Latino immigrants, probably because the migration from Latin-America to other countries of Europe is lower. Due to this fact, we will take into consideration the results obtained in the United States regarding racial/ethnic differences in internalizing symptomatology, between Latinos and other cultural groups.

Previous research has suggested that in some collectivistic societies where conformity and social inhibition are the norm, youths express more internalizing symptoms than do those youths in a more individualistic society such as, for instance, the United States (Valera et al., 2007). Latin-American countries are characterized by express conformity and social inhibition. It is common that parents pre-empt discussion when trouble arises (Canino, 2004). Therefore, adolescent immigrants from Latin-America may find themselves trapped between more traditional values that their parents promote to them at home and the more individualistic American values they are exposed to through school and the media (Chapman and Wooddruff-Borden, 2009). Thus, what happens with the prevalence of internalizing symptoms in the Spanish society, since it could be not as individualistic as the US society but, however, it could be less collectivistic that the Latin-American one?

Canino (2004) suggested that another possible explanation for the racial/ethnic differences in the prevalence of internalizing symptoms may be the comorbidity phenomenon. Somatic symptoms are a manifestation of depression and anxiety (Bernstein et al., 1997); thus, a cultural group with high levels of depressive symptoms may predispose individuals to high rates of somatic symptoms. Indeed, Campo (2012) indicated that the presence of one or more somatic symptoms is associated with not only concurrent anxiety and depressive symptoms, but also to an excess of anxiety and depressive disorders later in life. Prior research has found that depression and anxiety tend to co-occur (Axelson and Birmaher, 2001). Romero et al. (2010) found that 82 per cent of school children with depressive symptoms reported anxiety at the same time, and 20 per cent of these children with anxiety reported depressive symptoms. Hence, the prevalence of somatic symptoms may increase proportionally due to the prevalence of depression and depression proportionally due to anxiety. Children and adolescents with comorbid anxiety and depression have more severe symptoms than do non-comorbid anxious or depressed individuals (Axelson and Birmaher, 2001). The study of the anxiety–depression conjunction and somatic symptoms may help to clarify the nature of clinical reality and assist in improving existing nosology (Campo, 2012).

In addition, Canino (2004) has argued that the study of internalizing symptoms in different ethnic/racial groups should take into account the socio-economic status (SES), the type and size of the sample studied and gender differences. Given that the Latin-American immigration implies having a low SES most of the time, when a psychopathology has been associated with lower SES, researchers have to be careful with generalization concerning prevalence. Ostrove et al. (1999) found that wealth (a less traditional indicator of SES) made a significant contribution to explaining both physical and mental health in a similar fashion for African-Americans and Whites. Additionally, somatic and depressive symptoms have been associated with lower SES (Bernstein et al., 1997; Huang-Chi et al., 2008); thus, it is essential to separate the effect of SES from the effect of ethnicity (Canino, 2004).

Furthermore, to study ethnic/racial differences, it is important to analyse gender differences because prevalence of internalizing symptoms varies from girls to boys. Previous research has found that adolescent girls tend to report more internalizing symptoms than do boys (Huang-Chi et al., 2008; Romero et al., 2010; Zolog et al., 2011); therefore, a sample composed of more girls than boys may reflect an erroneous prevalence of internalizing symptoms.

There is a lack of studies on racial/ethnic differences in the prevalence of internalizing symptoms in Spain, despite the flow of immigration in the last few decades. Conducting research with Latino samples is important, as this is one of the largest minority groups in this country. It has been noted that the migration process implicitly leads to changes such as gender roles, family values, differences in lifestyles, norms, beliefs and identity, among others, which can produce high levels of stress, which in turn can produce serious mental health problems (Revollo et al., 2011). Examining racial/ethnic differences in psychopathology could help us to detect at-risk groups, to prevent the chronicity of symptoms and, as a result, to improve mental health services (McLaughlin et al., 2007; Whaley et al., 2011). It has been stated that the quality of care mediates the relationship between physician accommodation and medical adherence.

In line with the analysis made by Canino (2004) about methodological considerations, our aims are threefold (1) to compare the prevalence of anxiety, depressive and somatic symptoms in Spanish and Latin-American adolescents, (2) to contrast the prevalence of comorbidity of internalizing symptoms in Latinos and Spanish youths and (3) to examine the association of ethnicity, gender and SES with all the psychological variables studied. Based on previous research (Axelson and Birmaher, 2001; Chapman and Wooddruff-Borden, 2009; Umaña-Taylor and Updegraff, 2007), we expected (1) to find higher levels of internalizing symptoms in Latin-Americans, in comparison with Spanish adolescents, (2) that Spanish and Latin-American adolescents will show co-occurrence among the groups of symptoms studied and (3) girls from both cultural groups will report more symptomatology than will boys.

Method

Participants

Participants were in the second and third years of the nine secondary schools in Rubí – a city near Barcelona (Catalonia, Spain) – whose population was 72,987 inhabitants in 2010, according to the report of the National Statistical Institute (INE, 2010). In Rubí, there are 11,593 (16%) foreigners, of which 2071 are less than 16 years of age. Of the four secondary school levels in Spanish schools, in the city of Rubí, there were four state schools and five state-subsidized schools. In the 2009–2010 academic year, there were 1324 adolescents registered in the second and third years of secondary school classes in Rubí; of them, 1067 accepted to participate whereas 257 did not, because their parents did not allow them to take part in the study or because they did not want to. Owing to difficulties in understanding the language, six students could not answer the questionnaires. The study included adolescents with special needs but not those with a profound learning disability or pervasive developmental disorders.

A total of 1061 adolescents aged 13–16 years (M = 13.90 years, standard deviation (SD) = 0.82) participated in the study. The ethnic composition of the non-Spanish-born adolescent sample included Latin-Americans (n = 159; 15%), Africans (n = 47; 4.4%), Europeans from countries other than Spain (n = 5; 0.5%) and Asians (n =14; 1.3%). For this study, the sample was constituted only by Spanish and Latin-American adolescents; thus, the final sample was constituted by 993 adolescents (Table 1, top). The most prevalent countries of Latin-America were Ecuador (n = 88; 55.3%), Colombia (n = 18; 11.3%) and Bolivia (n = 13; 8.17%). Table 1 (top) shows the socio-demographic characteristics in more detail.

Table 1.

Description of the sample and psychological variables.

Socio-demographic characteristics (n and %)		Spanish (n = 834)		Latin-Americans(n = 159)
		n	%	n	%
Gender	Female	385	46.2	84	52.8
	Male	449	53.8	75	47.2
Socio-economic status	High-middle	213	26.2	25	16.7
	Middle	254	31.2	52	34.7
	Middle-low	346	42.6	73	48.7
Categorical measures (n and %)
SSQ: somatization		311	37.3	62	39.0
SSQ: headache		162	25.9	35	28.5
SSQ: stomach ache		111	17.7	18	14.6
SSQ: dizziness		59	9.4	18	14.6
SSQ: muscle pain		133	21.2	32	26.0
SSQ: fatigue		108	17.3	15	12.2
Anxiety + depression		114	13.7	29	18.2
Anxiety + somatic symptoms		157	18.8	29	18.2
Depression + somatic symptoms		79	9.5	24	15.1
Depression + anxiety + somatization		63	7.6	16	10.1
Quantitative measures (M and SD)		M	SD	M	SD
CDI: depression		11.2	6.3	12.6	6.2
SCARED: total anxiety		23.1	9.0	23.0	8.7
SCARED: separation anxiety		5.54	2.95	5.91	3.12
SCARED: social phobia		5.63	3.10	5.74	2.85
SCARED: generalized anxiety		8.35	3.15	7.75	3.05
SCARED: panic/somatic		3.56	2.80	3.50	3.00

SSQ: somatic symptom questionnaire; CDI: Children’s Depression Inventory; SCARED: Screen for Child Anxiety-Related Emotional Disorders; SD: standard deviation.

Instruments

Children’s Depression Inventory (Kovacs, 1992)

Children’s Depression Inventory (CDI) is one of the most widely used self-report questionnaires for depressive symptoms for 7- to 17-year-old children and adolescents. It takes 10–20 minutes to be completed and consists of 27 items scored on a 3-point Likert scale (0: absent; 1: moderate; 2: severe). A score of 17 or more is considered as having the best sensitivity (92.6%) and specificity (63.3%) in the Spanish general population (Canals et al., 1991). CDI has good internal consistency and good test–retest reliability (Canals et al., 1991; Kovacs, 1992). In our sample, internal consistency was also satisfactory (Cronbach’s alpha value of .83).

Screen for Child Anxiety-Related Emotional Disorders (Birmaher et al.,1999; Canals et al., 2012; Doval et al., 2011; Vigil-Colet et al., 2009)

The adolescents are asked about the frequency of their symptoms using a 3-point Likert-type scale (i.e. 0: not true or hardly ever true, 1: somewhat true or sometimes true and 2: very true or often true). A cut-off of 25 on the Screen for Child Anxiety-Related Emotional Disorders (SCARED) resulted in optimal sensitivity (75.9%) and specificity (68.5%) in the Spanish general population (Canals et al., 2012). Also in an American sample, Birmaher et al. (1999) found five anxiety factors: somatic/panic, generalized anxiety, separation anxiety, social phobia and school phobia. Nevertheless, in the Spanish population, specifically in Catalonia, SCARED showed only four factors (Canals et al., 2012). Internal consistency (Cronbach’s alpha) for the four scale scores was somatic/panic (.78), generalized anxiety (.69), separation anxiety (.70) and social phobia (.69). Thus, in the current study, the school phobia factor is not taken into consideration. In this study, internal consistency of the total scale was satisfactory (.85).

Somatic Symptom Questionnaire (Domènech-Llaberia et al., 2004)

It enquires about the presence and frequency (once, twice, three times and over three times) of five somatic complaints (headache, dizziness, stomach ache, fatigue and muscle pain) in the 3 months prior to assessment. Moreover, in order to assess the functional impairment due to somatic complaints, children and adolescents were asked about how many days they had missed school (1, 2, 3 or more than 4 days) owing to somatic symptoms. To rule out that somatic symptoms lacked any medical explanation, participants were asked whether, due to somatic complaints, they had been to the paediatrician and whether the doctor had found a medical explanation for this discomfort. If the explanation was given due to illness or accident, the presence of somatic symptoms was ruled out, but if, on the contrary, the subject responded that the paediatrician or doctor had found no explanation, it was considered that the symptoms were a somatic complaint. This somatic symptom questionnaire (SSQ) was based on SSQ, developed by Domènech-Llaberia et al. (2004). In the current study, internal consistency was satisfactory (α = .80). A socio-demographic questionnaire was prepared by the authors, considering the following variables: age, gender, educational level of parents (Hollingshead, 1975) and place of birth.

Procedure and design

The Commission of Ethics in human experimentation and the research committee of our university gave a favourable report of this research project. All schools agreed to participate, and they collaborated in sending the parental, written informed consent to the students’ homes. The adolescent’s oral consent was obtained in the classrooms, before the self-report questionnaires were administered.

Information was obtained through self-reports between 20 February and 20 May 2010. Each classroom was supervised by two members of the research team and the teacher of the class. The time necessary for completing the questionnaires was about 50 minutes. The researchers explained the objectives and the importance of their participation to pupils. Before leaving the class, team members made sure pupils completed all items.

Statistical analysis

The Statistical Package for Social Science (SPSS 18) for Windows was used. In order to examine the association of ethnicity, gender and SES, with all of the psychological variables studied (depression, anxiety, somatization and comorbidities), we conducted hierarchical linear regression models for quantitative measures and binary logistic regression models for dichotomous variables. Both linear and binary logistic regression models were performed using the enter method in SPSS. First, a check was made for overlap among predictors with bivariate Pearson’s correlations. Next, the statistical significance of the totality of interaction terms between ethnicity (Latinos vs Spanish) and the other predictors was valued using a chunk test. The chunk test allows researchers to evaluate all of the interaction terms together, indicating whether at least one of the interaction terms should be retained in the model (p < .05) or suggesting that all of the terms can be eliminated (p > .05). Comorbidity was obtained by creating dichotomous variables for adolescents who equalled or exceeded the cut-off of at least two measures, for instance, SCARED ≤ 25 and CDI ≤ 17. The somatization measure was created for participants who reported one or more somatic symptoms (headache, stomach ache, dizziness, fatigue and muscle pain).

Results

Table 1 (bottom) shows the descriptions for the variables considered in each group. A higher percentage of Latinos (26.6%) reported depressive symptoms, in comparison with Spaniards (18.3%). In contrast, regarding anxiety, Spaniards (39.1%) reported a higher level of symptomatology than did Latinos (37.1%). The somatic symptom most reported by both cultural groups was headache, whereas the least reported one was dizziness for the Spanish and fatigue for Latinos.

None of the predictors showed collinearity(r < .75). Regarding interaction terms, only ethnicity × SES was statistically significant for headache (p = .004) and the panic/somatic factor scores of SCARED (p = .047); therefore, analyses within each SES level for these two measures were performed. For the rest of the measures considered, all interaction terms involving ethnicity were statistically non-significant (chunk test, p ≥ .124); thus, they were removed in the final models presented. Due to the fact that ethnicity and SES were statistically significant for one somatic symptom and one anxiety factor, we compared the prevalence of psychopathology of both ethnicity groups taking SES into account. In both Latinos and Spaniards, adolescents from low SES showed more psychopathology in comparison with the other two SES groups. Nevertheless, it should be noted that Latinos from low SES showed more psychopathology than did Spaniards from the same SES group. Figure 1 shows this information in more detail.

Figure 1.

Prevalence of psychopathology in Latinos (left) and Spaniards (right) by SES.

Being Latin-American was associated with higher depressive symptoms (β = .07) and lower generalized anxiety symptoms (β = −.85). In addition, being female was associated with higher depressive and anxiety symptoms (β: between .13 and .21). Finally, high SES was negatively related to depressive symptoms (β = −.07) and anxiety (β: between −.07 and −.14). Table 2 shows the results for the final linear regression models.

Table 2.

Linear multiple regression analyses examining the association of ethnicity, gender and SES and the psychological variables.

Criterion		Predictors	B	95% CI B	β	p-value	F (p)	R ²
CDI: depressive symptoms		Ethnicity: Latino	1.15	0.06; 2.24	.07	.039	8.54 (<.001)	.038
		Gender: female	1.68	0.89; 2.48	.13	<.001
		SES low	0.32	−0.60; 1.24	.03	.493
		SES high	−1.08	−2.14; −0.02	−.07	.045
SCARED: total anxiety symptoms		Ethnicity: Latino	−0.95	−2.48; 0.58	−.04	.225	13.08 (<.001)	.056
		Gender: female	3.62	2.51; 4.74	.20	<.001
		SES low	−0.47	−1.77; 0.82	−.03	.475
		SES high	−2.11	−3.60; −0.62	−.10	.005
SCARED: separation anxiety		Ethnicity: Latino	0.17	−5.13; 5.89	.02	.511	11.39 (<.001)	.042
		Gender: female	0.86	0.48; 1.23	.14	<.001
		SES low	−0.20	−0.64; 0.22	.03	.351
		SES high	−1.08	−1.58; −0.59	−.15	<.001
SCARED: social phobia		Ethnicity: Latino	0.11	−0.53; 0.69	.00	.969	2.10 (.079)	.009
		Gender: female	0.43	0.42; 0.83	.07	.030
		SES low	0.33	−0.12; 0.79	.05	.153
		SES high	−0.12	−0.53; 0.51	.00	.964
SCARED: generalized anxiety		Ethnicity: Latino	−0.74	−1.28; −0.19	−.85	.008	10.33 (<.001)	.042
		Gender: female	1.10	0.71; 1.50	.17	<.001
		SES low	−0.45	−0.90; −0.01	−.07	.055
		SES high	−0.59	−1.11; −0.63	−.08	.028
SCARED: panic/somatic	SES high	Ethnicity: Latino	0.62	−0.47; 1.72	.07	.265	8.09 (<.001)	.065
	SES high	Gender: female	1.34	0.65; 2.02	.24	<.001
	SES medium	Ethnicity: Latino	−0.83	−1.76; 0.09	−.10	.07	5.34 (.005)	.034
	SES medium	Gender: female	0.98	0.29; 1.67	.16	.005
	SES low	Ethnicity: Latino	−0.26	−0.94; 0.41	−.03	.438	8.99 (<.001)	.042
	SES low	Gender: female	1.10	0.43; 0.00	.20	<.001

SES: socio-economic status; CI: confidence interval; CDI: Children’s Depression Inventory; SCARED: Screen for Child Anxiety-Related Emotional Disorders.

Category reference for SES was medium.

Being a girl was related to the presence of all somatic symptoms (odds ratio (OR) ranging from 1.40 to 4.09) except muscle pain and headache in low SES (OR = 3.69) and was also related to comorbidity (OR: between 1.77 and 2.27). In both cultural groups, girls reported more symptoms than boys, but these differences were more pronounced for Spaniards than for Latinos. Table 3 shows gender differences of Latinos and Spaniards in the prevalence of anxiety, depression, somatic symptoms and different types of comorbidities. For headache, simple effects for ethnicity × SES (p = .004) revealed that being Latin-American was positively associated (OR = 3.55) in high SES and negatively (OR = 0.38) in medium SES. Table 4 shows the results for the final binary logistic regression models.

Table 3.

Gender differences of both cultural groups (Latinos vs Spaniards) in the prevalence of anxiety, depression, somatic symptoms and comorbidities.

	Latino boys		Latino girls		χ ²	Spaniard boys		Spaniard girls		χ ²
	n	%	n	%		n	%	n	%
Anxiety	20	26.7	39	46.4	.010	140	31.2	186	48.3	<.001
Depression	17	23.0	25	29.8	.335	67	14.9	86	22.3	.006
Headache	8	13.6	27	42.2	<.001	60	17.4	102	35.3	<.001
Stomach ache	3	5.1	15	23.4	.004	35	10.4	76	26.3	<.001
Dizziness	6	8.0	21	25.0	.004	20	4.5	63	16.4	<.001
Muscle pain	13	22.0	19	29.7	.334	65	19.3	68	23.5	.196
Fatigue	7	9.3	19	22.6	.024	73	16.3	77	20.0	.161
Anxiety + depression	11	14.7	18	21.4	.270	40	8.9	74	19.2	<.001
Anxiety + somatic symptoms	35	46.7	62	73.8	.038	231	51.4	270	70.1	<.001
Depression + somatic symptoms	35	46.7	53	63.1	<.001	196	43.7	222	57.7	<.001
Depression + anxiety + somatization	7	9.3	16	19.0	.082	25	5.6	57	14.8	<.001

Table 4.

Binary logistic regression analyses examining the association of ethnicity, gender and SES and the psychological variables.

Criterion		Predictors	OR	95% CI OR	p-value	H–L	R ²
SSQ: total somatization		Ethnicity: Latino	0.18	0.82; 1.68	.359	.123	.039
		Gender: female	1.92	1.48; 2.49	<.001
		SES low	0.12	0.79; 1.59	.515
		SES high	1.12	0.90; 1.65	.196
SSQ: headache	SES high	Ethnicity: Latino	3.55	1.29; 9.77	.014	.784	.072
		Gender: female	1.92	0.90; 4.09	.090
	SES medium	Ethnicity: Latino	0.38	0.15; 0.97	.043	.996	.064
		Gender: female	2.08	1.15; 3.78	.016
	SES low	Ethnicity: Latino	1.38	0.74; 2.60	.313	.107	.115
		Gender: female	3.69	2.15; 6.35	<.001
SSQ: stomach ache		Ethnicity: Latino	0.70	0.40; 1.24	.221	.404	.079
		Gender: female	2.99	1.97; 4.55	<.001
		SES low	0.94	0.53; 1.67	.825
		SES high	1.43	0.90; 2.28	.127
SSQ: dizziness		Ethnicity: Latino	1.89	1.16; 3.10	.011	>.999	098
		Gender: female	4.09	2.57; 6.53	<.001
		SES low	1.15	0.67; 1.99	.596
		SES high	0.78	0.48; 1.27	.335
SSQ: muscle pain		Ethnicity: Latino	1.13	0.70; 1.81	.618	.985	.009
		Gender: female	1.36	0.95; 1.94	.088
		SES low	1.07	0.66; 1.73	.788
		SES high	1.16	0.77; 1.76	.481
SSQ: fatigue		Ethnicity: Latino	0.94	0.59; 1.49	.794	.869	.016
		Gender: female	1.40	1.00; 1.96	.046
		SES low	1.62	1.02; 2.55	.038
		SES high	1.52	1.01; 2.28	.043
Anxiety + depression		Ethnicity: Latino	1.25	0.78; 2.00	.349	.533	.046
		Gender: female	2.17	1.48; 3.16	<.001
		SES low	0.57	0.33; 0.97	.039
		SES high	0.94	0.63; 1.42	.784
Anxiety + somatic symptoms		Ethnicity: Latino	0.92	0.64; 1.33	.684	.578	.059
		Gender: female	2.27	1.73; 296	<.001
		SES low	0.89	0.63; 1.27	.182
		SES high	1.20	0.70; 1.49	.536
Depression + somatic symptoms		Ethnicity: Latino	1.17	0.82; 1.67	.382	.009	.032
		Gender: female	1.77	1.36; 2.28	<.001
		SES low	1.09	0.77; 1.54	.592
		SES high	1.23	0.91; 1.66	.167
Depression + anxiety + somatization		Ethnicity: Latino	1.43	0.85; 2.40	.175	.487	.058
		Gender: female	2.26	1.70; 4.16	<.001
		SES low	5.67	0.30; 1.05	.073
		SES high	8.75	0.55; 1.39	.573

SES: socio-economic status; OR: odds ratio; CI: confidence interval; H–L: Hosmer–Lemeshow test; SSQ: somatic symptom questionnaire.

Category reference for SES was medium.

Discussion

The main purpose of this study is to compare the prevalence of anxiety, depression, somatic symptoms and comorbidity in Spanish and Latin-American adolescents. Similar to previous research, we found differences in the self-report of psychopathology between the two cultural groups studied (Courtenay et al., 2002; Jason et al., 1997; Ostrove et al., 1999; Williams et al., 1997). Owing to the lack of studies in Spain, we compared results obtained in other cultures as the United States. Hence, in line with other studies, Latin-Americans reported more depressive symptoms than did the Spanish (Umaña-Taylor and Updegraff, 2007; Valera et al., 2007; Van Voorhees et al., 2008); however, we did not find differences in the prevalence of anxiety and somatic symptoms. This result is not consistent with prior studies, which indicated that in the United States, anxiety and somatic symptoms are more prevalent in Latin-American and African-American than in White-American adolescents (Valera et al., 2007; Van Voorhees et al., 2008). Our findings could be explained by cultural differences between Spain and the United States. Indeed, the American culture is more individualistic in comparison with Spain’s culture; thus, Spanish adolescents may express more social inhibition than do youth in the United States. Furthermore, there is evidence that in Spain, children report high levels of anxiety symptoms (Romero et al., 2010). Thus, results of the present study suggest that Spaniards are as anxious as Latino adolescents. Nevertheless, more research is needed to confirm these findings. Moreover, Spanish youths reported more generalized anxiety than did Latinos, although there was no difference in the total score of anxiety symptoms. Perhaps, Spanish adolescents only present high levels of some anxiety disorders and not all.

As we anticipated, Spanish and Latin-American adolescents showed co-occurrence among the groups of symptoms studied, although differences were not statistically significant. The highest prevalence values corresponded to somatic symptoms and depression, which is aligned with Bernstein et al. (1997), who indicated the high comorbidity between both groups of symptoms. Latinos could report more comorbidity depressive–somatic symptoms because they also reported high levels of depressive and somatic symptoms. It can be noted that comorbidity should be taken into account when studying the prevalence of symptom presentation (Canino, 2004).

Similar to previous studies, in comparison with boys, girls reported higher rates of depression, anxiety and somatic symptoms (Axelson and Birmaher, 2001; Huang-Chi et al., 2008). Additionally, females from medium and low SES reported higher levels of generalized anxiety than did boys in the same SES. Previous research has suggested that gender differences in the prevalence of internalizing symptoms in adolescence could be attributed to endocrine and/or neurobiological changes associated with puberty and sociocultural factors (Axelson and Birmaher, 2001; Canino, 2004). We found that girls showed more anxiety and somatic symptoms than did boys. However, regarding depression and comorbidity, Spanish girls showed more symptomatology than did boys. This is not in line with prior research, which found that Hispanic/Latino females in the United States reported higher levels of comorbidity, depressive and anxiety symptoms in comparison with other cultural groups such as Whites and Blacks (McLaughlin et al., 2007). More research on racial/ethnic differences is needed in Spain in order to identify gender differences in the prevalence of adolescent psychopathology. It would be useful to treat and prevent internalizing psychopathology.

Results of this study support findings of previous work, showing that SES was related to the prevalence of symptoms (Canino, 2004). Adolescents with high SES showed less anxiety, depression, panic/somatic symptoms and generalized anxiety than did participants from medium SES. In contrast, youths with low SES reported more comorbidity depression and anxiety, in comparison with medium SES. Thus, high SES was a protecting factor for reporting internalizing symptoms. In the total sample, Latin-Americans were mostly of low SES and a minority of the upper-middle group; this should make us reflect on the fact that in this sense, Latinos are more likely to report internalizing symptoms not for being Latinos, but for belonging mainly to middle and middle-low SES.

Conclusion

Results yield a description of racial/ethnic differences of internalizing symptoms in Spanish and Latin-American adolescents, living in Spain; however, we should support more future research about racial/ethnic differences because there is no definitive conclusion stating that Latin-American adolescents have higher rates of symptoms than do other cultural groups. This new information could help clinicians for assessing depressive, somatic symptoms and distress, particularly when they attend Latin-American adolescents in Spain, especially girls. In addition, more studies about psychopathology and ethnicity are necessary in order to plan effective programs not only to treat, but also to prevent the increase of internalizing psychopathologies. It is incumbent to address more research concerning the relation between ethnicity, psychopathology and some psychosocial variables such as acculturation level, resilience and stressful life events. For the time being, this study has contributed to the methodological considerations raised by Canino (2004), who recommended the study of racial/ethnic differences to take into consideration methodological issues such as SES, type and size of the sample studied (if possible, community sample), gender differences and the comorbidity phenomenon.

Limitations

Any generalization of these results should be made in the light of several considerations. First, this study was cross-sectional; therefore, causal inferences among variables cannot be derived. Furthermore, data were obtained from the information given by adolescents using self-report questionnaires and not from a clinical interview. In addition, regarding results of the regressions, it is worth mentioning that the SCARED social phobia and the depression plus somatyc symptom models showed unsatisfactory results. Despite these limitations, to date, this is the first study in Spain comparing the prevalence of anxiety, depression, somatic symptoms and comorbidity in a community sample of Spanish and Latin-American adolescents. We hope to have contributed to the lack of information about racial/ethnic differences in Europe, particularly in Spain. Undoubtedly, more research is needed in the area of internalizing symptoms among children and adolescents from different ethnic groups.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Axelson

Birmaher

(2001) Relation between anxiety and depressive disorders in childhood and adolescence. Depression and Anxiety 14(2): 67–78.

Bernstein

Massie

Thuras

. (1997) Somatic symptoms in anxious-depressed school refusers. Journal of the American Academy of Child and Adolescent Psychiatry 36(5): 661–668.

Birmaher

Brent

Chiapetta

. (1999) Psychometric properties of the Screen for Child Anxiety Related Emotional Disorders (SCARED): A replication study. Journal of the American Academy of Child and Adolescent Psychiatry 38(10): 1230–1236.

Campo

(2012) Annual research review – Functional somatic symptoms and associated anxiety and depression: Developmental psychopathology in pediatric practice. Journal of Child Psychology and Psychiatry, and Allied Disciplines 53(5): 575–592.

Canals

Domènech-Llaberia

Clivillé

. (1991) Síntomas de depresión en la etapa puberal: primeros resultados de un estudio epidemiológico longitudinal. Actas Luso-Españolas de Neurología, Psiquiatría y Ciencias Afines 19(3): 155–161.

Canals

Hernández-Martínez

Cosí

. (2012) Examinations of a cut-off score for the Screen for Child Anxiety Related Emotional Disorders (SCARED) in a non-clinical Spanish population. Journal of Anxiety Disorders 26: 785–791.

Canino

(2004) Are somatic symptoms and related distress more prevalent in Hispanic/Latino Youth? Some methodological considerations. Journal of Clinical Child and Adolescent Psychology 33(2): 272–275.

Chapman

Wooddruff-Borden

(2009) The impact of family functioning on anxiety symptoms in African American and European American young adults. Personality and Individual Differences 47(6): 583–589.

Courtenay

Mccreary

Merighi

(2002) Gender and ethnic differences in health beliefs and behaviours. Journal of Health Psychology 7(3): 219–231.

10.

Domènech-Llaberia

Jané

Canals

. (2004) Parental reports of somatic symptoms in preschool children: Prevalence and associations in a Spanish sample. Journal of the American Academy of Child and Adolescent Psychiatry 43(5): 598–604.

11.

Doval

Martínez

Domènech-Llaberia

(2011) Evidencias de calidad psicométrica de la versión en castellano del Screen for child anxiety emotional disorders (SCARED) de 41 ítems. Ansiedad y Estrés 17(2–3): 199–210.

12.

Hollingshead

(1975) Four factor index of social status. Unpublished manuscript, Department of Sociology, Yale University, New Haven, CT Retrieved 2012, November 20, from http://www.yale.edu/sociology/faculty/docs/hollingshead_socStat4factor.pdf.

13.

Huang-Chi

Tze-Chun

Ju-Yu

. (2008) Depression and its association with self-esteem, family, peer and school factors in a population of 9586 adolescents in southern Taiwan. Psychiatry and Clinical Neuroscience 62(4): 412–420.

14.

Instituto Nacional de Estadística (INE) (2010) Available at: http://www.ine.es/ (accessed 16 December 2011).

15.

International Organization for Migration (IOM) (2004) Migration from Latin America to Europe: Trends and policy changes. Available at: http://publications.iom.int/bookstore/index.php?main_page=product_info&;cPath=17_20&products_id=199 (accessed 29 March 2012).

16.

Jason

Jordan

Richman

. (1999) A community-based study of prolonged fatigue and chronic fatigue. Journal of Health Psychology 4(1): 9–26.

17.

Kovacs

(1992) Children Depression Inventory CDI (Manual). Toronto, ON, Canada: Multi-Health Systems.

18.

McLaughlin

Hilt

Nolen-Hoeksema

(2007) Racial/ethnic differences in internalizing and externalizing symptoms in adolescents. Journal of Abnormal Child Psychology 35(5): 801–816.

19.

Ostrove

Feldman

Adler

(1999) Relations among socioeconomic status indicators and health for African-Americans and Whites. Journal of Health Psychology 4(4): 451–463.

20.

Pina

Silverman

(2004) Clinical phenomenology, somatic symptoms, and distress in Hispanic/Latino and European American youths with anxiety disorders. Journal of Clinical Child and Adolescent Psychology 33(2): 227–236.

21.

Revollo

Qureshi

Collazos

. (2011) Acculturative stress as a risk factor of depression and anxiety in the Latin America immigrant population. International Review of Psychiatry 23(1): 84–92.

22.

Romero

Canals

Hernández-Martínez

.(2010) Comorbilidad entre los factores del SCARED y la sintomatología depresiva en niños de 8-12 años. Psicothema 22(3): 613–618.

23.

Umaña-Taylor

Updegraff

(2007) Latino adolescents’ mental health: Exploring the interrelations among discrimination, ethnic identity, cultural orientation, self-esteem, and depressive symptoms. Journal of Adolescence 30(4): 549–567.

24.

Valera

Weems

Berman

. (2007) Internalizing symptoms in Latinos: The role of anxiety sensitivity. Journal of Youth and Adolescence 36(4): 429–440.

25.

Van Voorhees

Paunesku

Kuwabara

. (2008) Protective and vulnerability factors predicting new-onset depressive episode in a representative of U.S. adolescents. The Journal of Adolescent Health 42(6): 605–616.

26.

Vigil-Colet

Canals

Cosí

. (2009) The factorial structure of the 41-item version of the Screen for Child Anxiety Related Emotional Disorders (SCARED) in a Spanish population of 8 to 12 years-old. International Journal of Clinical and Health Psychology 9(2): 313–327.

27.

Whaley

Smith

Hancock

(2011) Ethnic/racial differences in the self-reported physical and mental health correlated of adolescent obesity. Journal of Health Psychology 16(7): 1048–1057.

28.

Williams

Jackson

. (1997) Racial differences in physical and mental health: Socio-economic status, stress and discrimination. Journal of Health Psychology 2(3): 335–351.

29.

Zolog

Jané-Balladriga

Bonillo-Martín

. (2011) Somatic complaints and symptoms of anxiety and depression in a school-based simple of preadolescents and early adolescents. Functional impairment and implications for treatment. Journal of Cognitive and Behavioral Psychotherapies 11(2): 191–208.