Worry among Latinx College Students: Relations to Anxious Arousal, Social Anxiety, General Depression, and Insomnia

Abstract

Objective:

Latinx young adults in college (ages 18-25 years) are at a heightened risk for health disparities and there is a need to understand individual-based characteristics that are related to such health inequalities including anxiety, depression, and sleep disturbances. The cross-sectional current study investigated the role of worry in relation to anxious arousal, social anxiety, general depression, and insomnia among Latinx college students.

Participants/Method:

Participants included 401 (M_age = 21 years; SD =2.01; 83% female) Latinx students at a large, southwestern university.

Results:

Results indicated that greater levels of worry were related to increased levels of anxious arousal, social anxiety, general depression, and insomnia. These findings were evident above and beyond variance accounted for by age, sex, and subjective social status.

Conclusions:

The current investigation suggests that elevated levels of worry among Latinx young adults may be associated with greater levels of anxiety, depression, and insomnia.

In the United States (U.S.) a large proportion of college students have reported difficulties related to mental and behavioral health concerns including symptoms related to anxiety, depression, and sleep difficulties.¹ Moreover, the Latinx population represents a growing racial/ethnic minority group among college campuses with an estimated 12% of students who are Latinx.^1,2 Like other racial/ethnic minority groups within the U.S. and the world more generally, Latinx individuals are more apt to suffer from mental health disparities relative to the majority non-Latinx White population.^3,4 Specifically, research suggests that Latinx persons are more apt to experience certain mental health problems⁴ and may lack access to care or evidenced-based therapeutic approaches for such problems.⁵ Within the larger Latinx population, young adults in college within the age ranges of 18-25 years of age are at a heightened risk for mental and behavioral health problems,^6,7 presumably due to transitioning to greater independent roles and interacting in environments that may differ from home communities.^8,9 This is a developmental window for many emerging adults wherein psychological symptoms and problems emerge and thereafter are maintained over time.^7,10,11

Of psychological and behavioral health problems, anxiety and depression as well as insomnia are the most common among Latinx persons.^12-14 Indeed, extant work suggests that the Latinx population, including young adults, experiences greater levels of anxiety and depressive symptoms compared with non-Latinx Whites.⁴ These mental health concerns may be exacerbated by several issues among Latinx persons, including access to care in general, or evidenced-based care specifically.¹⁵ Such negative emotional symptoms also tend to co-occur with sleep dysfunction,¹⁶ especially insomnia. Research expressly focused on sleep disorders among Latinx persons, including youth, is less common than that focused on many other racial/ethnic minority groups. However, the available studies suggest that insomnia, the most common type of sleep problem, may be overrepresented among Latinx youth compared to non-Latinx Whites¹⁷ and be related to an increased risk of anxiety/depression.¹⁸ Overall, such work is broadly in line with evidence that there may be racial/ethnic differences in sleep-related breathing.¹⁹

To better understand the nature of emotional vulnerability and behavioral health problems, such as insomnia, among Latinx young adults, there is apt to be clinical utility in identifying individual-based characteristics that are related to increased risk of anxiety, depression, and sleep disturbances. The tendency to worry is one construct that has shown theoretical and empirical promise in terms of better understanding mental and other behavioral health problems (e.g., sleep dysfunction). Worry is defined as the apprehensive expectation about future negative events.²⁰ The tendency to worry is a dimensional construct that tends to be relative stable at the individual difference dimension.²¹ Worry also is distinguishable from other negative affect states and constructs, such as rumination.^22,23 Research largely focused on non-Latinx Whites has found that greater levels of worry are associated with a range of psychopathological processes, including avoidance, emotional dysregulation, and interpersonal deficits.^21,24 Worry is also the predominant characteristic --- at heightened levels --- of generalized anxiety disorder (GAD).

Surprisingly, worry has not been studied among Latinx persons to the same extent it has among other racial/ethnic groups. The extant scientific literature among Latinx in North America persons suggests that worry-related problems, such as GAD, may be overrepresented among this group compared to other racial/ethnic groups.^25,26 Other research has found that greater levels of worry may be more common among Latinx youth when compared to non-Latinx Whites.^27,28 Research that has included Latinx persons outside of the U.S. has similarly found elevated rates of worry among this group.^29,30 The available literature also suggests that the experience of worry among Latinx persons is evident across a wide array of life domains, including interpersonal, health, and other areas. Despite initial research showing the potential clinical relevance of worry among Latinx persons, to the best of our knowledge, the value of worry has not been tested among Latinx young adults in terms of some of the most common emotional (e.g., anxiety, depression) and behavioral health problems (e.g., insomnia) among this group.

Theoretically, Latinx young adults may experience elevated levels of worry^29,30 related to a number of different life areas.³¹ Accordingly, Latinx young adults be who are experiencing symptoms of worry may be at greater risk for mental and behavioral health concerns. Specifically, individuals who experience greater levels of worry may be more apt to maintain the capacity to acquire and maintain new feared associations, resulting in greater reported levels of anxiety and depressive symptoms.³² Moreover, individuals with greater levels of worry may also experience insomnia as a result of excessive nighttime worry (e.g., worry about the daytime consequences of not getting enough sleep). ^33,34 Yet, no work has tested this theory of the observed relations among a sample of Latinx young adults.

The purpose of the current cross-sectional study was therefore to empirically explore individual differences in the propensity to worry in relation to some of the most common emotional problems, as well as insomnia, among Latinx college students. It was hypothesized that greater levels of worry would be associated with anxious arousal, social anxiety, depression, and insomnia. This hypothesis is consistent with past research and theory among largely non-Latinx White samples that has found that the tendency to worry is related to more severe emotional and sleep problems.²¹ The expected worry effects were anticipated to be evident above and beyond the variance accounted for by variables related to mental and physical health among Latinx in past work, including age,³⁵ sex,³⁶ and subjective social status.³⁷

Method

Participants

The present sample is a subset of participants from a larger study of mental and physical health among students at a large, southwestern university. Inclusion criteria for the larger study included being between the ages of 18 and 64 and current enrollment at the university where the study took place. Latinx students who participated between September 2015 through May 2017 (N = 470; M_age = 22.3; SD = 4.4; age range: 18-53 years; 82.3% female) were selected for inclusion. Exclusion criteria for the larger study included being younger than age 18 and non-proficiency in English (to ensure comprehension of study questions). In addition, only participants who provided complete data were included. Specifically, in the current study we restricted the age range to the traditional college age for students given evidence that this period represents a significant environmental shift (corresponding to a host of new demands for autonomous social, behavioral, and cognitive functioning;) that is likely to fuel feelings of anxiety, depression, as well as sleep difficulties in this age group.¹¹ A final sample was comprised of 401 Latinx college students (M_age = 21 years; SD =2.01; Age range: 18-25 years; 83% female).

Measures

Demographic Questionnaire.

A demographic questionnaire was used to collect data including sex, race/ethnicity, and age for descriptive purposes. Age and sex were included as covariates in the current study.

Penn State Worry Questionnaire (PSWQ³⁸).

The PSWQ is a 16 item self- report measure used to assess worry. Items are rated on a 5-point Likert-type scale ranging from 1 (not at all typical of me) to 5 (very typical of me) and include questions such as “Many situations worry me.” Among college students, a cut off total score of 62 is suggested to be considered clinically significant.³⁹ This scale has demonstrated good psychometric properties, including internal consistency, test–retest reliability, and discriminate validity.^38-41 A total PSWQ score was used for the present study and yielded good internal consistency (α = .80).

Inventory of Depression and Anxiety Symptoms (IDAS⁴²).

The IDAS is a 64-item self-report instrument that assesses distinct affect symptom dimensions. The IDAS contains affective content, including 10 specific symptom subscales for suicidality, lassitude, ill temper, well-being, insomnia, appetite loss, appetite gain, panic, social anxiety, and traumatic intrusions, and two broad subscales of general depression and dysphoria. Respondents are asked to rate the degree to which they have experienced symptoms within the past two weeks, rated on a 5-point Likert-type scale from 1 (not at all) to 5 (extremely). The IDAS subscales show strong internal consistency, convergent and discriminant validity with psychiatric diagnoses and self-report measures; and short-term retest reliability (r = 0.79) with both community, and psychiatric patient samples.^42,43 The present study used the panic subscale (8 items; e.g. “My heart was racing or pounding”), the social anxiety subscale (5 items; e.g. “I found it difficult to talk with people I did not know well”), general depression subscale (20 items; e.g. “I felt depressed”), and the insomnia subscale (6 items; e.g., “I had trouble falling asleep”). The four subscales demonstrated good internal consistency among the present sample (panic α = 0.91; social anxiety α = 0.86; general depression α = 0.92; insomnia α = 0.83), consistent with past work among Latinx.⁴⁴

Subjective Social Status (SSS⁴⁵).

Subjective social status was assessed with the community version of the MacArthur Scale.⁴⁵ Participants were presented with a picture of a 10-rung ladder. Respondents were asked to rate from 1 (worst) to 10 (best) where they stand in the community, relative to others, where higher rungs indicate higher status (i.e., more money, more education, and better jobs). This measure has demonstrated adequate reliability and validity in previous work^46-48 and has been utilized among racially diverse U.S. samples⁴⁹ and college students.⁵⁰ The SSS scale was utilized as a covariate in the current study.

Procedure

The present sample included university students at a large, southwestern university interested in receiving extra credit towards their psychology course as compensation. Participants were recruited via flyers and posting on the psychology subject pool for the department. All participants provided informed consent, which was completed over the internet before proceeding to the online self-report survey. Quality assurance questions were placed throughout the survey to protect against sporadic responding and to ensure the integrity of the data. Identifying information was not retained for each participant; there was no link between each participant’s identity and study responses. This study protocol was approved by the Institutional Review Board at the university where the study took place.

Analytic Strategy

Analyses were conducted using SPSS version 24. Data were first checked for quality assurance; no cases of sporadic responding were detected. Sample descriptive statistics and zero-order correlations among study variables were examined. To evaluate the incremental predictive power of worry, four separate two-step hierarchical regressions were conducted for each of the criterion variables; (1) anxious arousal, (2) social anxiety, (3) general depression, and (4) insomnia. For all analyses, step 1 covariates included age, sex (0 = male, 1 = female), and SSS. Step 2 included the PSWQ total score. Model fit for each of the steps was evaluated with the F statistic and an increase in variance accounted for as evidenced by a change in R². Squared semi-partial correlations (sr²) were used as measures of effect size.

Results

Bi-variate Relations

Zero-order correlations among all study variables are presented in Table 1. Worry was positively correlated with anxious arousal (r = .13, p = .009), social anxiety (r = .21, p < .001), general depression (r = .34, p < .001), and insomnia (r = .21, p < .001) In the current sample, 22.7% of Latinx students met for moderate levels of worry and the remaining 77.3% of Latinx students met for high levels of worry. Utilizing suggested clinical cut-offs among college students ³⁹, 41.4% of the sample met criteria for clinically significant worry.

Table 1.

Descriptive Statistics and Bivariate Correlations between Study Variables (N = 401)

Variable	1	2	3	4	5	6	7	8
1. Agea	-
2. Sexa	.10	-
3. SSSa	−.01	−.03	-
4. PSWQb	.07	.15**	−.08	-
5. Anxious Arousalc	.09	.04	−.13*	.13**	-
6. Social Anxietyc	−.07	−.02	−.14**	.21***	.65***	-
7. General Depressionc	.10*	.03	−.20***	.34***	.61***	.65***	-
8. Insomniac	0.09	.02	−.19***	.21***	.50***	.46***	.76***	-
Mean/n (SD/%)	21.01 (2.01)	333 (83%)	5.83 (1.79)	60.55 (3.14)	11.31 (5.25)	9.33 (4.63)	43.20 (13.84)	13.32 (5.29)

Note.

^***p < .001

^**p < .01

^*p < .05.

^aCovariate

^bPredictor

^cCriterion; Sex: % listed as females (Coded: 0 = male and 1 = female); SSS = Subjective Social Status-Community Subscale;⁴⁵ PSWQ = Penn State Worry Questionnaire;³⁸ Anxious Arousal = Inventory of Depression and Anxiety Symptoms-Panic subscale;⁴² Social Anxiety = Inventory of Depression and Anxiety Symptoms-Social Anxiety subscale;⁴² General Depression = Inventory of Depression and Anxiety Symptoms-General Depression subscale;⁴² Insomnia = Inventory of Depression and Anxiety Symptoms-Insomnia subscale.⁴²

Regression Analyses

Regression results are presented in Table 2. For anxious arousal, step one of the model with covariates was statistically significant (R² = .03, F(3, 397) = 3.35, p = .019); SSS was a significant predictor (b = −0.37, SE = 0.15, p = .011, sr² = .02). In step two, the model with PSWQ total score added was significant (R² = .04, F(4, 396) = 3.81, p = .005) and accounted for a significant increase in R² (ΔR² = .01, F(1, 396) = 5.09, p = .025). Specifically, the PSWQ total score was a significant predictor (b = 0.19, SE = 0.08, p = .025, sr² = .01).

Table 2.

Hierarchical Regression Results

Anxious Arousal
Model		R2	b	SE	β	t	p	sr2
1	Age	.03*	0.22	0.13	0.08	1.67	.096	.01
	Sex		0.41	0.70	0.03	0.59	.554	.00
	SSS		−0.37	0.15	−0.13	−2.55	.011	.02
2	Age	.04**	0.20	0.13	0.08	1.55	.121	.01
	Sex		0.19	0.70	0.01	0.27	.790	.00
	SSS		−0.34	0.15	−0.12	−2.37	.018	.01
	PSWQ		0.19	0.08	0.11	2.26	.025	.01
Social Anxiety
Model		R2	b	SE	β	t	p	sr2
1	Age	.02*	−0.15	0.11	−0.07	−1.32	.188	.00
	Sex		−0.17	0.61	−0.01	−0.28	.781	.00
	SSS		−0.36	0.13	−0.14	−2.82	.005	.02
2	Age	.07***	−0.18	0.11	−0.08	−1.57	.117	.01
	Sex		−0.53	0.61	−0.04	−0.88	.379	.00
	SSS		−0.32	0.13	−0.12	−2.53	.012	.02
	PSWQ		0.31	0.07	0.21	4.19	> .001	.04
General Depression
Model		R2	b	SE	β	t	p	sr2
1	Age	.05***	0.66	0.34	0.10	1.95	.052	.01
	Sex		0.64	1.81	0.02	0.35	.724	.00
	SSS		−1.53	0.38	−0.20	−4.05	.000	.04
2	Age	.15***	0.54	0.32	0.08	1.67	.095	.01
	Sex		−1.09	1.73	−0.03	−0.63	.531	.00
	SSS		−1.33	0.36	−0.17	−3.72	> .001	.03
	PSWQ		1.45	0.21	0.33	6.97	> .001	.10
Insomnia
Model		R2	b	SE	β	t	p	sr2
1	Age	.04**	0.23	0.13	0.09	1.77	.078	.01
	Sex		0.09	0.69	0.01	0.13	.894	.00
	SSS		−0.55	0.14	−0.19	−3.80	> .001	.03
2	Age	.08***	0.20	0.13	0.08	1.58	.115	.01
	Sex		−0.30	0.69	−0.02	−0.43	.664	.00
	SSS		−0.51	0.14	−0.17	−3.54	> .001	.03
	PSWQ		0.33	0.08	0.19	3.97	> .001	.04

Note. N for analyses is 401 cases.

^***p < .001

^**p < .01

^*p < .05.

SSS = Subjective Social Status-Community Subscale;⁴⁵ PSWQ = Penn State Worry Questionnaire;³⁸ Anxious Arousal = Inventory of Depression and Anxiety Symptoms-Panic subscale;⁴² Social Anxiety = Inventory of Depression and Anxiety Symptoms-Social Anxiety subscale;⁴² General Depression = Inventory of Depression and Anxiety Symptoms-General Depression subscale;⁴² Insomnia = Inventory of Depression and Anxiety Symptoms-Insomnia subscale.⁴²

In terms of social anxiety, step one with covariates was significant (R² = .02, F(3, 397) = 3.24, p = .022) with SSS significantly predicting social anxiety (b = −0.36, SE = 0.13, p = .005, sr² = .02). In step 2, when PSWQ total score was added, the model remained significant (R² = .07, F(4, 396) = 6.92, p < .001) and accounted for a significant increase in R²(ΔR² = .04, F(1, 396) = 17.56, p < .001); PSWQ total score was a significant predictor of social anxiety (b = 0.31, SE = 0.07, p < .001, sr² = .04).

In regard to general depression, the overall model with covariates was significant (R² = .05, F(3, 397) = 6.91, p < .001); age (b = 0.66, SE = 0.34, p = .052, sr² = .01) and SSS (b = −1.53, SE = 0.38, p < .001, sr² = .04) were significant predictors. PSWQ total score was added in step two, and the model remained significant (R² = .15, F(4, 396) = 17.95, p < .001) and accounted for a significant increase in R² (ΔR² = .15, F(1, 396) = 48.58, p < .001). Specifically, the PSWQ total score was a significant predictor of general depression (b = 1.45, SE = 0.21, p < .001, sr² = .10).

In predicting insomnia, the model with covariates was significant (R² = .04, F(3, 397) = 5.94, p = .001). Specifically, SSS (b = −0.55, SE = 0.14, p < .001, sr² = .03) significantly predicted insomnia. In step two, the model with PSWQ total score added remained significant (R² = .08, F(4, 396) = 8.56, p < .001) and accounted for a significant increase in R² (ΔR² = .04, F(1, 396) = 15.77, p < .001); PSQW total score was a significant predictor (b = 0.33, SE = 0.08, p < .001, sr² = .04).

Discussion

The purpose of the present investigation was to explore the relevance of worry in terms of some of the most common mental health symptoms and insomnia among Latinx college students. Although worry has a long history among non-Latinx Whites as a clinically important construct,²⁰ there is highly limited scientific work on this individual difference variable among Latinx. This limitation is unfortunate, as some work has suggested that worry may be an under-recognized emotional vulnerability factor for behavioral health problems among Latinx.²⁵

In line with expectations, the results indicated that greater levels of worry were significantly associated with each of the studied criterion variables. Specifically, worry was related to greater levels of anxious arousal, social anxiety, general depression, and insomnia. The worry effects were evident above and beyond the variance accounted for by age, sex, and subjective social status. The amount of the unique variance accounted for by worry ranged from small (1% of variance) for anxious arousal to medium (10%) for general depression. Such findings offer novel empirical insight into the relation between worry and a range of common mental health symptoms among Latinx college students. The present results build upon past work that has found worry-related problems, such as GAD, are overrepresented among Latinx²⁵ adults by being the first to link the severity of worry to three separate classes of mental health problems among Latinx young adults. Although the mechanism(s) explaining the observed worry severity-mental health association is not known, past work suggests that a range of factors could be involved (e.g., pathophysiological altercations, interpersonal stressors^24,51,52,53). Based upon the current data, it would be useful for future research to explore the severity of worry in terms of mental health and sleep problems and disorders among Latinx using prospective methodology, including time sampling approaches and longitudinal designs.

The current data suggest that worry may be an individual difference factor relevant to better understanding the high prevalence of anxiety, depression, and insomnia among Latinx young adults. Indeed, among the present sample, approximately 41.4% met criteria for clinically significant worry despite not being expressly recruited for a specific worry level. Thus, the prevalence of clinically severe worry appears to be high among Latinx young adults. Although past work has established that severe worry may exacerbate mental health problems among non-Latinx groups,²⁰ the clinical importance of worry has not been extensively explored among Latinx. Available theoretical models suggest that greater levels of the tendency to worry are associated with more severe anxiety and depressive symptoms and sleep disturbance.^53-55 Such models implicate problem solving deficits,⁵³ intolerance of uncertainty,⁵⁶ avoidance,⁵⁷ and emotion dysregulation as possible mechanisms involved in such relations. Although such models have not been culturally adapted for Latinx, they all predict that worry may impair the depth of emotional processing of salient life events. Such impaired emotional processing, in turn, theoretically exacerbates the severity and chronicity of emotional and other behavioral health problems such as sleep.

It also is important to note two other issues pertaining to worry in the present sample. First, the sample was not selected based on high levels of worry. Yet, the average worry score on the PSWQ was 60, which is above the typical clinical cut-off for GAD.^24,39,58 These prevalence data highlight the heightened level and commonality of worry among the young adult Latinx sample. Research is needed to further explore the prevalence of worry-related psychopathology using structured clinical interviews to cross-validate the findings from the current report. Second, worry was related to each of the dependent variables at the bi-variate level (see Table 1). However, despite correlating with each emotional variable and insomnia, worry maintained unique variance with these constructs. Indeed, the range of shared variance was only 1% to 4%. These data help make clear that worry is related to, but distinct from, other affect and health-related constructs. Future research could usefully be focused on exploring the relevance of worry in terms of other psychosocial problems common among Latinx, including discrimination,⁵⁹ acculturative stress,⁶⁰ as well as posttraumatic stress.⁶¹

The present findings have important clinical implications given mental health promotion-based interventions among Latinx individuals are highly limited. ⁶² If replicated and extended using longitudinal methodology, the findings suggest worry may represent a clinically important individual difference variable among Latinx young adults that should be assessed and perhaps a focus of prevention, health promotion, and treatment efforts among this group. For example, it may be beneficial to utilize worry as a potential screening mechanism within college health care settings for Latinx individuals who may be at risk for anxiety, depression, as well as insomnia. In doing so, these concerns may be identified and treated earlier through appropriate mental health referrals. Clinically, there are several evidenced-based intervention programs for GAD that target worry as a core psychopathological process.^63-65 There may be clinical utility in evaluating such intervention programs for Latinx young adults to offset vulnerability related to worry, as has been done for other groups.^66,67 It also may be useful to adapt such treatment programs to Latinx by integrating a culturally-oriented therapeutic element. For example, as in past work among Latinx, ^68,69 it may be beneficial to incorporate fotonovelas (i.e., small booklets that portray a story using photographs and captions) which may increase knowledge and awareness of worry and its treatment as well as reduce mental health care stigma. Moreover, in line with the American College Health Association’s Guidelines for Hiring Health Promotion Professional in Higher Education initiative,⁷⁰ these data may inform culture competency efforts within higher education settings. Specifically, the current study may provide insight into culturally competent evidence-based health promotion initiatives among health professionals within higher education including.

The current findings should be interpreted in the context of several limitations. First, the present data are cross-sectional. Consequently, there is a need to replicate and extend the findings over time to better characterize the nature of the observed relations. Second, the sample was disproportionately female. Although there were a meaningful number of males in the sample (n = 68), it would be ideal to have a more sex-balanced sample in future research. Third, we utilized a self-report assessment approach to measure the studied constructs. Although the measures employed are psychometrically sound in terms of validity and reliability, it would be ideal to have complimentary assessments using alternative modes of instrumentation. For example, there would be utility in using structured clinical interviews for the presence of psychopathology and assessment of sleep patterns using objective measures (e.g., actigraphy). Fourth, the current study represents an initial empirical evaluation worry among Latinx young adults, but there could be value in evaluating whether this construct maintains similar relations in younger and older populations of Latinx individuals. This type of work would facilitate understanding the relative generalizability of the findings reported to the larger Latinx population. Fifth, cultural factors beyond subjective social status were not modeled in the current investigation. It may be useful in future research to explore how worry and cultural factors of known importance among Latinx interrelate with one another and psychological and health problems among this group. For example, worry may moderate or mediate relations between such factors as acculturative stress and discrimination in terms of mental and physical health problems. Such research would facilitate understanding the interplay between individual differences in worry and sociocultural factors among the Latinx population. Finally, the current study did not evaluate the impact of generation of residency in the U.S. or primary language spoken at home. Future studies may benefit from examining these variables among the observed relations presented in the current study.

In summary, the current investigation found that high levels of worry were evident among an unselected sample of Latinx young adults and such worry was related to anxiety and depressive symptoms as well as insomnia. These data are consistent with the perspective that worry may be an important individual difference factor among Latinx young adults that could be involved in demonstrated behavioral health problems among this group. Drawing from this initial study, there is utility in further exploring the relative importance of worry in models of mental health and health behavior (e.g., sleep problems) among Latinx young adults.