Beyond Anxious Predisposition: Do Padecer de Nervios and Ataque de Nervios Add Incremental Validity to Predictions of Current Distress among Mexican Mothers?

Abstract

Background

Nervios (PNRV) and ataque de nervios (ATQ) are culture-bound syndromes with overlapping symptoms of anxiety, depression, and dissociation, shown to have inconsistent associations to psychiatric disorder. Few studies test the basic assumption that PNRV and ATQ are uniformly linked to distress outcomes across Latina/o immigrant groups. This study examined: (a) the extent to which acculturative stress, Latino/U.S. American acculturation, and anxious predisposition were associated with lifetime history of ATQ and PNRV, and (b) the extent to which ATQ and PNRV add incremental validity in explaining acculturative stress and psychological distress beyond measures of anxious predisposition.

Method

Participants (n = 82) included Mexican mothers who completed surveys on acculturation, trait anxiety, anxiety sensitivity, lifetime ATQ/PNRV, psychological distress, and acculturative stress.

Results

Lifetime PNRV, but not lifetime ATQ, was significantly predictive of psychological distress. PNRV was also linked to trait anxiety. Psychometric measures of anxious predisposition (trait anxiety, anxiety sensitivity) were more robust predictors of distress outcomes than lifetime history of ATQ/PNRV.

Conclusions

Inquiry into lifetime history of nervios may be a useful point of entry in talking to Mexican immigrant mothers about stress and distress. However, standard tools for assessing anxiety sensitivity and trait anxiety appear most useful in identifying and explaining presence of psychological distress. Further research is needed to determine the cross-cultural relevance of trait anxiety and anxiety sensitivity, and its implications for the development of anxiety treatments that are effective across cultures.

INTRODUCTION

Three decades of empirical advances in the cross-cultural study of psychopathology catalyzed inclusion of Appendix I in DSM-IV.^{[1, 2]} Appendix I provides an outline for cultural formulation and a glossary of culture-bound syndromes. Its inclusion in psychiatry’s textbook of nosology represented formal recognition of the notion that cultural context could modulate the phenomenology and assessment of psychopathology.^[2] Over 40% of the culture-bound syndromes listed in Appendix I explicitly refer to an overlap with anxiety pathology, and much research has focused on the links between these syndromes and DSM’s anxiety disorders. Proponents of a culturally-informed psychiatric nosology have called for fuller integration of experience-near concepts such as culture-bound syndromes within DSM-V and subsequent editions,^[3] but an empirical framework to guide integration of these culturally embedded psychological phenomena with information from conventional diagnostic instruments still does not exist. This study is one of the first to examine the clinical validity and utility of assessing lifetime history of nervios and ataque de nervios, (two culture-bound syndromes prevalent among Latina/os and included in the DSM-IV), among a community sample of Mexicans, relative to assessments utilizing Western measures of anxious predisposition.

The DSM-IV Glossary of Culture-Bound Syndromes defines nervios and ataque de nervios as idioms of distress with overlapping symptoms of anxiety, depression, and dissociation, used throughout Latin America and the Caribbean. ^{[1, 4]} Idioms of distress reflect a broad range of expressions of negative feeling states commonly endorsed among members of a specific community.^[5] Nervios [“nerves”] refer to a chronic, generalized sense of psychological distress;^[6–8] variants of the term are used across cultural groups.^[9] Padecer de nervios (PNRV) is the state of suffering from nervios. Between15.5% to 62.5% of Latina/os from Central America, Mexico, the Caribbean, and the U.S. endorse a lifetime history of nervios with variations in rates by geographic region. ^{[6, 10–21]} Across communities, nervios is described as an “emotional problem” ^[20] occurring in the context of an interpersonal incident or physical condition, often accompanied by distressing negative emotions, associated behavioral changes, and co-occurring depressive symptoms. ^{[6, 22–24]} Women report a higher prevalence than men, ^{[10, 13, 25]} and a history of nervios predicts higher odds for lifetime affective and anxiety disorders. ^[13]

In contrast, ataque de nervios [“attack of nerves”; ATQ] involves typical and atypical panic symptoms such as loss of control, crying, rage, aggressiveness, amnesia, and ensuing sense of relief.^{[26, 27]} In the U.S., lifetime history of ATQ is endorsed most frequently among Puerto Ricans (15%), followed by Mexicans (9.6%), and Cubans (9%), ^[28] with comparable rates observed among island Puerto Ricans. ^{[25, 29]} Among clinical samples, prevalence rates can be as high as 70%. ^{[30, 31]} Endorsement of ATQ is higher among women, and those who are highly acculturated to U.S. American life ways, widowed/separated/divorced, ^[28] of low educational attainment, or over age 45.^[25] Despite evidence that degree of acculturation to Latino ways of life may be the most significant indicator of a tendency to present and experience distress in culturally consonant terms, the association between level of acculturation and ATQ history has been inconsistent ^{[32, 33]}.

Similar to nervios, reports of ATQ are associated with lifetime mood, anxiety, and substance use disorders, clinical severity, psychiatric hospitalizations, use of primary medical and tertiary mental health services, and non-criterion PTSD symptoms.^{[25, 28, 34]} Unlike nervios, frequent ATQs are also linked to dissociative symptoms, ^[35] childhood trauma, ^[36] dissociative predisposition, ^[37] unexplained neurological symptoms and panic disorder. ^[38]

Most of the literature on ATQ has focused on understanding its relationship to panic attacks and anxiety sensitivity. Findings indicate that ATQs are not culturally sanctioned panic attacks as originally postulated.^[30] Key defining features of ATQs include presence of an identifiable trigger, aggression, dissociative symptoms, suicidality, absence of avoidant behavior, and prolonged symptom development, which contrasts with the acute progression of symptoms present in DSM-defined panic attacks. ^{[27, 31, 32, 39, 40]} However, there is evidence of an association between ATQ and anxiety sensitivity. Latina/o community participants with history of ATQ are similar to those with elevated anxiety sensitivity and no history of ATQ in measures of anxiety and psychopathology. ^[32] Among clinical samples, anxiety sensitivity and dissociative predisposition are significant predictors of ATQ severity; but only anxiety sensitivity predicts past-month ATQ. ^{[41, 42]} Fear of negative emotions (e.g., anger and anxiety) and fear of physiological symptoms, as captured by a measure of anxiety sensitivity, are strong predictors of ATQ severity. Though there are contradictory data, ^[33] the emerging evidence supports the proposal that the fear of arousal symptoms associated with high anxiety sensitivity generates a heightened secondary fear response when an ataque occurs, escalating fear in a positive feedback loop, and creating a self-perpetuating fear cycle.^[42]

In sum, the relationship between lifetime or current ATQ/PNRV and psychiatric disorder and/or distress is not one-to-one.^{[13, 28]} ATQ is not isomorphic to panic attacks or anxiety sensitivity, and level of acculturation is not a consistent predictor of ATQ. However, most of the extant literature has focused on ATQ among Puerto Ricans. There is evidence that cultural “idioms of distress” may be used differently across cultures.^{[5, 9]} Mexicans compose the largest ethnic group within the Latina/o population in the U.S.^[43], but it is not yet known whether ATQ/PNRV carry similar clinical value in Mexican compared to Puerto Rican populations, and whether the association between acculturation, anxious predisposition, distress, ATQ and PNRV are similarly patterned in these communities. Answering calls to investigate culture-bound syndromes on their own terms, ^[44] this study examined: (a) the extent to which acculturative stress, Mexican/U.S. American acculturation, and anxious predisposition were associated with lifetime history of ATQ and PNRV, and (b) the extent to which ATQ and PNRV add incremental validity in explaining acculturative stress and psychological distress beyond anxious predisposition. These data are among the first to examine ATQ and PNRV in a Mexican sample, examining links to acculturation, acculturative stress, and anxious predisposition as measured with Western instruments. Herein, psychological acculturation refers to the behavioral, attitudinal, identity, linguistic, and value changes that occur in an individual as the result of long-term contact with people sharing initially unfamiliar cultural practices.^[45–50] The related term “acculturative stress” refers to the broad negative cognitive, behavioral, and emotional changes that are due to and sustained by immigration related stressors. ^{[51, 52]} We tested three hypotheses:

1. Mexican acculturation, acculturative stress, anxiety sensitivity, and trait anxiety would be associated with increased odds for lifetime ATQ/PNRV after adjusting for age and English proficiency.

2. Lifetime ATQ would add incremental validity in predicting past-week psychological distress beyond trait anxiety and anxiety sensitivity.

3. Lifetime ATQ/PNRV would be more robust predictors than anxiety sensitivity and trait anxiety of past three-month acculturative stress.

The theoretical foundation for these hypotheses posits that attention to cultural “idioms of distress” may be critical to a proper understanding of the clinical complaints and types of distress that Latina/os report in U.S. American mental health clinics. For example, cultural idioms may be used systematically to signal individual level psychiatric vulnerabilities (e.g., anxious predisposition) that generate distress (e.g., acculturative stress) and motivate help-seeking. Perceived stress may also increase the frequency and severity of these idiomatic experiences. Thus, a deeper understanding of these interactions may be essential to shaping more effective treatments for this large and growing segment of the U.S. population.

MATERIALS & METHODS

Participants

Ninety-nine Mexican adult mothers of infant, preschool, or elementary school age children were recruited. Eighty-two provided complete data and were included in final analyses. The mean age was approximately 30 years (Table 1). Approximately 88% had an 8^th grade educational level, and over 90% were married. The majority (98.7%)self-identified as first-generation or foreign-born immigrants. All were enrolled in a family English literacy program in a Midwestern city, and were recruited from five sites serving predominantly Mexican (migrant or immigrant) families. All signed informed consent (approved by university and literacy program IRBs).

Table 1.

Background Variables for Entire Sample and Data Cohorts

	Overall Sample	Cohort 1	Cohort 2
	N =82	n =48	n = 34
Variable	M (SD) or N (%)	M (SD) or N (%)	M (SD) or N (%)	t(df) or χ2(df)
Demographics
Age	29.88 (5.59)	28.50 (5.21)	31.82 (5.60)	−2.76 (80)**
English Proficiency (BEST Score)	59.08 (12.65)	56.44 (11.94)	62.81(12.85)	−2.30 (80)*
History of Ataques
No	48 (58.5)	28 (58.3)	20 (58.8)	.002 (1)
Yes	34 (41.5)	20 (41.7)	14 (41.2)
History of Padecer de Nervios
No	34 (41.5)	17 (35.4)	17 (50.0)	1.74 (1)
Yes	48 (58.5)	31 (64.6)	17 (50.0)
Distress/Stress
Psychological Distress (BSI)	.80 (.57)	.82 (.58)	.79 (.55)	.29 (80)
Acculturative Stress (HSI)	11.56 (2.57)	11.33 (2.48)	11.89 (2.70)	−.96 (80)
Anxiety
Anxiety Sensitivity (ASI)	17.20 (11.07)	15.77 (10.93)	19.21 (11.12)	−1.39 (80)
Trait Anxiety (STAIY-2)	41.44 (10.07)	41 (9.46)	42.06 (11.00)	−.47 (80)
Acculturation
US American Acc. (ARSMA AOS)	2.08 (.48)	2.13(.52)	2.01 (.41)	1.13 (80)
Mexican Acc. (ARSMA MOS)	4.26 (.41)	4.34 (.32)	4.15(.49)	2.16 (80)*

^*p < 0.05,

^**p < 0.01,

^***p < 0.001

Measures

A dichotomous survey assessed whether participants endorsed lifetime history of ATQ or PNRV. Respondents were asked: 1) have you ever had an ataque de nervios? and 2) have you ever suffered from nervios? There are no official diagnostic criteria for these constructs, and our goal was to capture self-identified use of the cultural idioms, as a potential expression of mental distress or disorder. Data from a small subset of participants (n = 22) endorsing lifetime history of ATQ or PNRV verified that these survey items captured participants with significant typical and atypical panic symptom burdens (M = 8.46; SD = 3.58) and significant rates of lifetime (54.54%) and past-month (31.82%) mood or anxiety disorders as diagnosed by the SCID.^[53]

The Acculturation Rating Scale for Mexican Americans-Revised (ARSMA-II) was used to assess acculturation status.^[45] This multidimensional scale assesses extent of involvement in Mexican and “Anglo” culture, providing independent subscales for Mexican Orientation (MOS), and Anglo Orientation (AOS). In this sample, Cronbach’s alpha coefficients indicated good internal consistency for the AOS subscale (α = .81) and adequate consistency for MOS subscale (α = .67). The Basic English Skills Test (BEST) was used to measure English proficiency.^[54] Scaled scores yield English as a Second Language (ESL) education and functioning levels. Testing occurred within two months of study participation. The Spanish Version of the Brief Symptom Inventory (BSI)^{[55, 56]} was administered to assess psychological distress. The BSI includes 53 items about past week distress.^{[57, 58]} Its Global Severity Index provides a composite measure of psychological distress. The BSI displayed good internal consistency in this sample (α = .97). The Immigrant Version of the Hispanic Stress Inventory (HSI)^[59] was used to measure acculturative stress. This instrument measures distress/worry associated with interpersonal, economic and immigration conflict in the past three-months. Our Cronbach’s alpha coefficient (α = .92) indicated good internal consistency. The acculturative stress composite variable was divided by ten so as to facilitate meaningful interpretation of unit increment changes in the outcome variable. Anxiety sensitivity was measured with the Anxiety Sensitivity Index (ASI),^{[60, 61]} which taps the extent to which a person finds anxiety-related sensations to be frightening or catastrophic.^[60–62] Good internal consistency was observed (α =.82). Trait anxiety was measured with the trait scale of the State/Trait Anxiety Inventory (STAI T-Anxiety),^{[63, 64]} which assessed proneness to anxiety and tendency to appraise stressful situations as threatening. The STAI-T exhibited good internal consistency (α = .79).

Procedure

Participants were recruited through flyers and in-person efforts. Participation was voluntary and confidential. After consenting, interested mothers completed the survey battery in Spanish. All instruments were previously translated and validated in Spanish-speaking Latina/o communities.^[65–68] Participants received $10 for their time and were given a list of local mental health resources. Data for 53 participants were collected between November 2007 and May 2008 from two sites serving immigrant and migrant mothers of preschool children (Cohort 1). An additional 46 participants were recruited in May and June 2009 from three additional sites serving immigrant mothers with elementary school-age children (Cohort 2).

RESULTS

Descriptive Statistics

Thirty four participants endorsed lifetime ATQ, whereas 48 endorsed lifetime PNRV (Table 1). Participants in Cohort 1 were younger, with lower English proficiency scores and higher Mexican acculturation relative to participants in Cohort 2. These differences were expected because we recruited mothers of preschool children for Cohort 1, who were likely to be younger and more recent immigrants than those recruited for Cohort 2 who were mothers of elementary school age children. Participants had High Beginning English proficiency, which indicates they could understand, speak, read, and write common words, phrases, and sentences. Participants were high in psychological distress, scoring in the 90^th percentile of the normative adult female population on the BSI distress scale. They scored in the nonclinical range for trait anxiety and anxiety sensitivity. Mean ASI scores were similar to those reported in normative populations in Spain and Puerto Rico, ^{[32, 69]} as well as those observed among English speaking normal controls.^[62]

Missing Data

Data were incomplete for 24% of initially enrolled participants, equally distributed across study variables with the exception of English proficiency. We used regression estimates to impute missing English Proficiency scores (n =10). Sensitivity analyses were conducted to determine whether deletion of this variable would yield significant changes in regression estimates compared to inclusion as an imputed variable. The regression estimates did not change, so the imputed variable was used. There were no significant differences found between included (n = 82) and excluded participants (n = 17).

Bivariate Analyses

Pearson product-correlation analyses (Table 2) and bivariate logistic/linear regression analyses were conducted as an initial step. Odds for self-reported lifetime ATQ were higher among those with positive histories of PNRV (Odds Ratio [OR]=4.96; 95% Confidence Interval [CI]=(1.81, 13.59); p<.01). Odds for lifetime PNRV were higher among those with psychological distress (OR=6.14; 95% CI=2.07, 18.16; p<.001), acculturative stress (OR=1.23; 95% CI=1.00, 1.50; p<.05), and trait anxiety (OR=1.07; 95% CI=1.02, 1.12; p<.01). Current acculturative stress was related to lifetime ATQ (β=.21, p<.05), anxiety sensitivity (β=.45, p<.001), and trait anxiety (β=.48, p<.001). Current psychological distress was related to lifetime PNRV (β=.36, p<.001), current acculturative stress (β=.49, p<.001), anxiety sensitivity (β=.51, p<.001), and trait anxiety (β=.68, p<.001).

Table 2.

Intercorrelations between primary predictor, demographic, and dependent variables (N = 82)

	1	2	3	4	5	6	7	8	9	10
1. Age	1.00
2. History of ATQ	−0.14	1.00
3. History of PNVR	−0.13	.36**	1.00
4. Psychological Distress	−0.05	0.16	.39**	1.00
5. Acculturative Stress	0.13	0.14	.23*	.55**	1.00
6. US American Acculturation	0.12	0.04	0.05	−0.06	0.18	1.00
7. Mexican Acculturation	−0.06	0.16	0.05	−0.05	0.07	0.15	1.00
8. Anxiety Sensitivity	0.11	0.07	0.09	.54**	.50**	0.14	−0.08	1.00
9. Trait Anxiety	0.00	0.19	.30**	.67**	.45**	0.00	−0.16	.40**	1.00
10. English Proficiency	.23*	0.04	−0.03	−0.05	−0.03	0.09	−0.11	0.04	0.12	1.00

Note.

^*p < 0.05,

^**p < 0.01,

^***p < 0.001

Are ATQ/PNRV Associated with Acculturative Stress, Acculturation, and Anxiety?

To examine if acculturative stress, acculturation, and anxiety resulted in increased odds of lifetime ATQ/PNRV after controlling for age and English Proficiency, two sets of hierarchical logistic regressions were conducted with lifetime ATQ or PNRV as the outcomes. Predictors were entered in five blocks (Table 3). Based on “initial step” analyses, psychological distress was excluded due to high correlation with trait anxiety and evidence of multicollinearity.^[70]

Table 3.

Summary of Logistic Regression Analyses for Variables Predicting Cultural Syndromes

	Ataque de Nervios				Padecer de Nervios
	Model 4a		Model 5a		Model 4b		Model 5b
Step/Variable	Odds Ratio	95% CI	Odds Ratio	95% CI	Odds Ratio	95% CI	Odds Ratio	95% CI
Constant	0.00†		0.00†		0.03		0.07
1. Age	0.94	(.86, 1.03)	0.95	(.86, 1.04)	0.95	(.87, 1.04)	0.96	(.88, 1.06)
1. English Proficiency	1.02	(.98, 1.06)	1.02	(.98, 1.06)	1.00	(.96, 1.04)	0.99	(.95, 1.03)
2. Acculturative Stress	1.06	(.85, 1.33)	1.02	(.81, 1.29)	1.16	(.90, 1.50)	1.14	(.88, 1.47)
3. U.S. American Acculturation	1.13	(.41, 3.17)	1.07	(.36, 3.17)	1.21	(.41, 3.52)	1.34	(.41, 4.39)
3. Mexican American Acculturation	3.04†	(.81, 11.39)	2.7	(.73, 9.98)	1.40	(.42, 4.65)	1.00	(.28, 3.57)
4. Anxiety Sensitivity	1.00	(.95, 1.05)	1.00	(.95, 1.06)	0.98	(.93, 1.03)	0.98	(.93, 1.04)
4. Trait Anxiety	1.04	(.99, 1.10)	1.02	(.97, 1.09)	1.07*	(1.01, 1.13)	1.07†	(1.00, 1.13)
5. Positive History of PNRV			4.02*	(1.37, 11.81)			--	--
5. Negative History of PNRV			REF	REF			--	--
5. Positive History of ATQ							4.31**	(1.44, 12.90)
5. Negative History of ATQ							REF	REF
χ2 (df, df)	8.74 (1,7)		15.61 (1,8)*		11.96 (1, 7)		19.41 (1, 8)*
N	82		82		82		82
Nagerlkerke R2	0.14		0.23		0.18		0.28
−2 Log Likelihood	102.54		95.66**		99.32*		91.87**

Note.

^†p<.10,

^*p < 0.05,

^**p < 0.01,

^***p < 0.001

Hosmer and Lemeshow Goodness of Fit Test indicated good fit for all Models

The variables in Model 5a (Table 3)¹ explained 23% of the variance in lifetime ATQ. As in the unadjusted model, positive history of PNRV resulted in three-fold increase of lifetime ATQ relative to those with negative history of PNRV. Lifetime PNRV uniquely accounted for 9% of the variance. Similarly, the variables in Model 5b explained 28% of the variance in lifetime PNRV. As in the first set of analyses, lifetime ATQ was related to higher odds for lifetime PNRV relative to those with unremarkable lifetime ATQ; lifetime ATQ was responsible for the majority of the variance that explained lifetime PNRV (R² = 10%). Trait anxiety was significant in predicting lifetime PNRV in Model 4b and marginally significant in Model 5b when lifetime ATQ was included. Mexican acculturation and acculturative stress were not related to lifetime ATQ or PNRV.

Is Current Psychological Distress Predicted by ATQ/PNRV?

Hierarchical linear regression was used to examine the extent to which lifetime ATQ/PNRV were unique predictors of past-week psychological distress adjusting for age, English proficiency, and anxiety. Mexican and U.S. American acculturation subscales were excluded because they were not statistically related to psychological distress at the bivariate level. The predictors were entered in five blocks (Table 4). The variables in the fifth model explained 62% of the variance in psychological distress. Acculturative stress (β=.19, p<.05), anxiety sensitivity (β=.28, p<.01) and trait anxiety (β=.43, p<.001) all significantly predicted psychological distress when age and English proficiency were held constant. Positive history of PNRV (β=.20, p<.05) added a small but significant amount of additional predictive power, accounting for an additional 3% of the variance in psychological distress, beyond that explained by the other variables.

Table 4.

Summary of Linear Regression Analyses for Variables Predicting Psychological Distress (N = 82)

Step/Variable	Model 3		Model 4		Model 5
	B (SE B)	β	B (SE B)	β	B (SE B)	β
Constant	−.57 (.34)		−.58 (.34)		−.64 (.33)
1. Age	−.01 (.01)	−0.09	−.01 (.01)	−0.09	−.01 (.01)	−0.07
1. English Proficiency (BEST)	−.00 (.00)	−0.09	−.00 (.00)	−0.09	−.00 (.00)	−0.08
2. Acculturative Stress	.05 (.02)	0.21*	.05 (.02)	0.21*	.04 (.02)	0.19*
3. Anxiety Sensitivity	.01 (.01)	.26**	.01 (.01)	.26**	.01 (.00)	.28**
3. Trait Anxiety	.03 (.01)	.48***	.03 (.01)	.47***	.02 (.01)	.43***
4. Positive History of Ataques			.01 (.02)	0.02	−.01 (.02)	−0.04
5. Positive History of Nervios					.06 (.02)	.20*
F (df1, df2)	21.42 (5, 76)***		17.63 (6, 75)***		16.99 (7, 74)***
N	82		82		82
R2	0.59		0.59		0.62
ΔR2	.17***		0.00		.03*

Note.

^*p<0.05,

^**p < 0.01,

^***p < 0.001

Is Current Acculturative Stress Predicted by ATQ/PNRV?

Hierarchical linear regression was used to examine if lifetime ATQ/PNRV was predictive of three-month acculturative stress after controlling for age, English proficiency, and anxiety. The predictors were entered in four blocks (Table 5). Acculturation subscales were excluded because they were not statistically associated with acculturative stress at the bivariate level. Variables in the third model accounted for 34% of the variance. Anxiety sensitivity (β=.36, p<.001) and trait anxiety (β=.32, p<.01) were predictive of acculturative stress. Anxiety sensitivity (R²=24%) followed by trait anxiety (R²=9%) accounted for the majority of the variance in acculturative stress. Any effect of lifetime ATQ on acculturative stress disappeared when anxious predisposition was considered in the full model.

Table 5.

Summary of Linear Regression Analyses for Variables Predicting Acculturative Stress

	Model 1		Model 2		Model 3		Model 4
Step/Variable	B (SE B)	β	B (SE B)	β	B (SE B)	β	B (SE B)	β
Constant	10.39 (1.86)***		9.20 (1.64)***		6.48 (1.78)***		5.96 (1.82)
1. Age	.07 (.05)	0.15	.04 (.05)	0.10	.05 (.04)	0.12	.06 (.05)	0.14
1. English Proficiency (BEST)	−.01 (.02)	−0.07	−.02 (.02)	−0.08	−.02 (.02)	−0.11	−.02(.02)	−0.11
2. Anxiety Sensitivity			.11 (.02)	0.50***	.08 (.02)	.36***	.08 (.02)	.36***
3. Trait Anxiety					.08 (.03)	.32**	.07 (.03)	.28**
4. Positive History of Ataques							.06 (.13)	0.05
4. Positive History of Nervios							.14 (.13)	0.11
F (df1, df2)	.85 (2, 79)		9.10 (3, 78)***		10.11 (4, 77)***		7.02 (6, 75)***
N	82		82		82		82
R2	0.02		0.26		0.34		0.36
ΔR2			0.24***		.09**		0.02

Note.

^†p<.10,

^*p < 0.05,

^**p < 0.01,

^***p < 0.001

DISCUSSION

Herein we examined whether lifetime history of culturally sanctioned idioms of distress, ATQ and PNRV, provided more robust predictors of distress outcomes in a Mexican immigrant population than psychometrically-derived Western measures of anxious predisposition. Surprisingly, they did not – Western instruments measuring anxiety sensitivity (ASI) and trait anxiety (STAI-T) provided stronger predictors of psychological distress and acculturative stress than endorsed history of the culturally-bound “anxiety” syndromes. However, ATQ/PNRV did have differential associations to distress in this community, with PNRV providing some predictive power.

Across unadjusted and adjusted models, lifetime ATQ was largely associated with lifetime PNRV. Despite the interrelatedness, there were points of distinction. For example, lifetime PNRV emerged as a significant and consistent predictor of psychological distress, but lifetime ATQ did not. Lifetime PNRV was also associated with trait anxiety in unadjusted models and marginally so in adjusted models. Thus, it is possible that the relationship between ATQ and PNRV may be similar to the relationship between state level anxiety/negative affect and trait level neuroticism/negative emotionality. Neuroticism is a trait that is associated with general distress vulnerability and a tendency to experience negative affect.^[71] Endorsement of nervios in a Mexican immigrant population may reflect a related type of general distress vulnerability. Endorsing a history of ataques may reflect vulnerability to a more specific and temporary state of behavioral disruption that is less associated with general stress and distress. However, the data tentatively suggest that if seeking to detect a general predisposition to distress among Mexican mothers, whether acute or chronic, a measure of lifetime PNRV would be more useful than a measure of lifetime ATQ.

Disconfirming our hypotheses, measures of anxious predisposition were more consistent and robust predictors of distress outcomes in comparison to history of ATQ/PNRV. This result converges with previous research noting an association between anxious predisposition, psychopathology, and stress among Latina/os.^{[32, 65]} Anxiety sensitivity was a more robust predictor of acculturative stress than trait anxiety. Perhaps anxiety sensitivity and acculturative stress both tap the latent constructs of fear of uncertainty and lack of control typically discussed in models of normative and pathological worry, and intolerance of negative affect/emotion described in distress models. ^{[71, 72]} It is also possible that stressful contexts associated with the acculturation process may engender distress, which is then amplified by anxiety sensitivity, creating a positive feedback loop and escalating cycle. Anxiety sensitivity may be an important risk factor in the mental health of immigrants that warrants greater attention.

Our hypothesis that acculturation and English proficiency would be significant predictors of distress and lifetime ATQ/PNRV was also not supported. Our ability to detect statistically meaningful associations may have been undermined by low variability in acculturation status, use of a community rather than clinical sample,^[33] measurement concerns^[73], and emphasis on actual rather than perceived English proficiency.^[74]

This study had several limitations. Participants came from a small convenience sample of community respondents with restricted ranges in acculturation levels. Findings may have differed with a randomly selected sample. Results may also differ in a clinical sample, where rates of psychiatric disorders and distress are likely to be higher. Results have limited generalizability to other Latina/o ethnic groups with a wider range in acculturation status. The measure of ATQ and PNRV probed lifetime experience, not current or past-month; while other study variables involved current status. Results may have differed if recent experience of ATQ/PNRV (e.g., past-month, past six-months) was examined due to temporal ordering effects. Lifetime ATQ/PNRV was assessed with a dichotomous survey, which could be vulnerable to reporting biases or unreliability. Furthermore, our measure of ATQ/PNRV did not assess for lifetime or current severity, despite research evidence to support a relationship between ATQ severity and anxiety sensitivity.^[41] More extensive quantitative and qualitative data on the meaning and function of ATQ/PNRV in Mexican communities is needed to create greater confidence in the inferences made about the relationships between the culture-bound syndromes, acculturation, anxious predisposition, and psychopathology. Finally, we did not explore contextual factors that might influence mental health status, such as socioeconomic position, available social supports, perceived discrimination, and physical health.^{[75, 76]} Despite these limitations, this study tentatively suggests that the relative clinical utility and validity of assessing lifetime history of ATQ and PNRV may vary across diverse Latina/o ethnic groups.

CONCLUSION

Our results suggest that asking about lifetime history of nervios (not lifetime ATQ) might be a useful point of entry in talking to Mexican immigrant mothers about stress and distress, because it seems to be a meaningful “idiom of distress” – a culturally sanctioned way to discuss affliction that might reflect vulnerability to distress and emotional or behavioral dysfunction. However, anxious predisposition, as measured by standard and widely used psychometric instruments of anxiety sensitivity and trait anxiety, was the most consistent predictor of acculturative stress and psychological distress in this exploratory study. Thus, the constructs of anxiety sensitivity and trait anxiety may capture an important component of distress vulnerability that is applicable across cultures. Contrary to our expectations, these results suggest that Western approaches to treatment for anxiety sensitivity and high trait anxiety may be relevant to reducing distress vulnerability in this population. Further research is needed to determine the cross-cultural relevance of Western clinical approaches to anxious predisposition among Mexican immigrant clinical samples, and to develop empirically-informed diagnostic assessments that integrate culture-bound syndromes and Western measures of psychopathology.